? ? ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Homelessness: What?s in a Word? In: Hulchanski, J. David; Campsie, Philippa; Chau, Shirley; Hwang, Stephen; Paradis, Emily (eds.) Finding Home: Policy Options for Addressing Homelessness in Canada (e-book), Introduction. Toronto: Cities Centre, University of Toronto. www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 Introduction Homelessness: What?s in a Word? J. DAVID HULCHANSKI, PHILIPPA CAMPSIE, SHIRLEY B.Y. CHAU, STEPHEN W. HWANG, EMILY PARADIS As?we?write? this? introduction? in? 2009,? to? accompany? the? launch?of? an? electronic?book?that?brings?together?current?Canadian?research?on?home? lessness,?we?are?struck?by?the?way?in?which?the?term??homelessness??has? come?to?be?used???by?researchers,?by?the?media,?by?politicians,?by?service? providers.?Homelessness?has?been?called??an?odd?job?word,?pressed? into? service? to? impose?order?on?a?hodgepodge?of?social?dislocation,?extreme? poverty,? seasonal? or? itinerant?work,? and? unconventional?ways? of? life?? (Hopper?and?Baumohl,?1996,?p.?3). Why?do?we?have?such?a?term??Where? did? it?come? from??What?does? it?mean??What?does? it?conceal??These?are? all?essential?questions,?not?only?for?society?and?public?policy,?but?also?for? researchers.?What?are?we?researching?? The?invention?of?homelessness? A?search?of?the?New?York?Times?historical?database?covering?1851?to?2005? reveals?that?the?word?homelessness?was?used?in?4,755?articles,?but?87%?of? this?usage?(4,148?articles)?was?in?the?20?years?between?1985?and?2005.?Be? fore? the? 1980s,? it? is? rare? to? find? homelessness?used? to?designate? a? social? problem.?What?happened?in?that?decade?that?made?the?difference??? ? ? ? ? ? ? ? H O M E L E S S N E S S : ?W H A T ? S ? I N ? A ?W O R D ? / 2 ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Introduction: Homelessness: What?s in a Word? www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 In?1981,? the?United?Nations?announced? that?1987?would?be? the? In? ternational?Year? of? Shelter? for? the?Homeless? (IYSH).?What? the?United? Nations?intended?was?a?focus?on?the?fact?that?so?many?people?in?less?de? veloped? countries?were?unhoused.?There?was?no?mention? of?developed? countries?like?Canada?in?that?1981?UN?resolution.1?The?1981?UN?General? Assembly?resolution?also?did?not?use?the?word?homelessness.?The?term?as? the?name?of?a? social?problem?was?not? in?common?use?at? the? time.?The? 1981?UN?resolution?was?intended?to?draw?attention?to?the?fact?that?many? millions? of? households? in? developing? countries? had? no? housing.? They? were?unhoused,?homeless.?They?needed?adequate?housing.?? But?by?1987,? the? focus?of? the? International?Year?had? shifted? to? in? clude?homeless?people?in?the?developed?nations?of?the?world,?including? Canada.?In?that?year,?many?of?the?people?whose?work? is?represented? in? this? electronic? book? attended? conferences? on? homelessness? in? Canada? that?focused?on?the?growing?number?of?unhoused?people?in?Canada,?not? those?in?developing?countries.?? Before?the?1980s,?people?in?developed?countries?did?not?know?what? it?was? like?to?be?unhoused?or?homeless.?They?had?housing,?even? if?that? housing?was?in?poor?condition.?Some?transient?single?men?in?cities?were? referred?to?at?times?as??homeless.??But?the?term?had?a?different?meaning? then.?? In?1960,?for?example,?in?a?report?titled?Homeless?and?Transient?Men,?a? committee? of? the? Social? Planning?Council? of?Metro? Toronto? defined? a? ?homeless?man??as?one?with?few?or?no?ties?to?a?family?group,?who?was? thus?without? the? economic? or? social? support? a? family? home? normally? provides.? The? committee?made? a? clear? distinction? between? house? and? home.?The?men?were? homeless,? not? unhoused.?Home? refers? to? a? social,? psychological? space,? not? just? a? house? as? a? physical? structure.? These? homeless?men?had?housing,?albeit?poor?quality?housing???rooming?hous? 1??That?an?international?year?devoted?to?the?problems?of?homeless?people?in?ur? ban?and?rural?areas?of?the?developing?countries?...?to?focus?the?attention?of?the? international?community?on?those?problems,?Recognizing?the?grave?and?gener? ally?worsening?situation?of?the?homeless?in?the?developing?countries...??U.N.? General?Assembly,?Resolution?36/71.?International?Year?of?Shelter?for?the? Homeless,?4?December?1981.?? ? ? ? ? ? ? ? H O M E L E S S N E S S : ?W H A T ? S ? I N ? A ?W O R D ? / 3 ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Introduction: Homelessness: What?s in a Word? www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 es? or? accommodation? provided? by? charities.?Canada? at? that? time? thus? had?homeless? individuals,?but?no?problem?called??homelessness.??Most? of?the?homeless?individuals?at?that?time?were?housed,?though?their?hous? ing?was?of?poor?quality.? Similarly,?in?1977,?the?City?of?Toronto?Planning?Board?released?a?Re? port? on?Skid?Row.?This? report?never?uses? the?word??homelessness??and? uses?the?word??homeless??only?a?few?times.?These?men???and?they?were? mainly?men2? ??were?characterized?by? their??residence? in?a?deteriorated? mixed?commercial?residential?area? in?older?sections?of? the?city,??by? fre? quent? changes? in? residence,? and? by? the? low? rent? they? paid.?They? had? housing,?but?they?were?homeless.?? The? word? ?homelessness?? came? into? common? use? in? developed? countries?in?the?early?and?mid?1980s?to?refer?to?the?problem?of?dehousing? ??the?fact?that?an?increasing?number?of?people?who?were?once?housed?in? these?wealthy?countries?were?no? longer?housed.?Canada?had?started? to? experience?dehousing?processes.?? Until?the?1980s?Canadian?urban?planners,?public?health?officials,?so? cial?workers? and? related?professionals? had? been? focused? on? rehousing? people?into?better?housing?and?neighbourhoods.?This?was?because,?dur? ing?the?Depression?and?the?Second?World?War,?very? little?new?housing? was?built?and?many?people?were?living?in?poor?quality,?aging,?and?over? crowded?housing.?After?the?war,?Canadians?revived?the?housing?market,? created? a? functioning?mortgage? system?with?government?mortgage? in? surance,?built?social?housing,?and?subsidized?private?sector?rental?hous? ing.?About?20,000?social?housing?units?were?created?every?year?following? the?1973?amendments?to?the?National?Housing?Act.? In?addition,?starting? in?that?postwar?period,?people?who?needed?to? be?protected?during?difficult?economic?times?and?supported?in?ill?health? 2??It?was?only?a?few?years?later,?in?her?1982?book?The?Lost?and?the?Lonely,?that? McGill?sociology?professor?Aileen?Ross?examined??a?new?social?problem?that? of?homeless?women.??By?this?time,?destitute?women,?too,?were?finding?them? selves?in?Skid?Row?housing?and?even?on?the?street.?Ross?used?the?term??home? lessness??to?underscore?that?whether?housed?or?unhoused,?these?women?fell? outside?the?gendered?norms?associated?with?home:??Home?making?has?always? been?thought?of?as?a?much?more?important?part?of?a?woman?s?identity?than?a? man?s?Most?of?the?women?had?lost?this?part?of?their?identity.??? ? ? ? ? ? ? ? H O M E L E S S N E S S : ?W H A T ? S ? I N ? A ?W O R D ? / 4 ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Introduction: Homelessness: What?s in a Word? www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 and?old?age?received?the?assistance?they?needed.?Universal?health?insur? ance,?Unemployment?Insurance,?Old?Age?Pensions,?and?the?Canada?As? sistance?Plan?were? all? introduced? or? improved? as? national? cost?shared? programs?during?those?years.?? In? introducing? the?1973?housing? legislation,? the?Minister?of?Urban? Affairs???a? federal?ministry?we?no? longer?have? today?but?which?existed? during?most?of?the?1970s???clearly?asserted?that?our?society?has?an?obliga? tion?to?see?that?all?people?are?adequately?housed.?? When?we?talk???about?the?subject?of?housing,?we?are?talking?about?an? elemental?human?need???the?need?for?shelter,?for?physical?and?emotional? comfort?in?that?shelter.?When?we?talk?about?people?s?basic?needs???the?re? quirements?for?survival???society?and?the?government?obviously?have?an? obligation?to?assure?that?these?basic?needs?of?shelter?are?met.? I?have?already?acknowledged?this?obligation?in?stating?that?good?hous? ing?at?reasonable?cost?is?a?social?right?of?every?citizen?of?this?country.??? [This]?must?be?our?objective,?our?obligation,?and?our?goal.?The?legislation? which?I?am?proposing?to?the?House?today?is?an?expression?of?the?govern? ment?s?policy,?part?of?a?broad?plan,?to?try?to?make?this?right?and?this?ob? jective?a?reality?(Basford,?1973,?p.?2257).?? Undoubtedly?we?would?not?have? the? social?problem?of?homeless? ness?today?if?this?1970s?philosophy?had?continued?through?the?1980s?and? 1990s,?to?the?present?day.?? By?the?1980s,?however,?Canada?had?a?social?problem?that?was?and? has? ever? since? been? called? homelessness.? The? proceedings? of? Canada?s? 1987? national? IYSH? conference,? for? example,? included? a?document? en? dorsed? by? the? conference,? called? the? ?Canadian?Agenda? for?Action? on? Housing? and?Homelessness? through? the? Year? 2000.?? This? agenda? in? cluded?the?following?explicit?summary?of?the?federal?government?s?fail? ure?to?take?action?on?the?growing?national?affordable?housing?crisis.? A?significant?component?of?the?homelessness?problem?is?that?housing?has? not?been?a?high?priority?for?governments?at?any?level?.?[O]nly?a?small? proportion?of?government?resources?are?directed?to?improving?housing? conditions?.?In?all?regions?of?the?country,?the?demand?for?housing?that?is? adequate?and?affordable?to?low?income?persons?and?the?willingness?of?lo? cal?organizations?ready?to?build?greatly?exceed?the?availability?of?gov? ? ? ? ? ? ? ? H O M E L E S S N E S S : ?W H A T ? S ? I N ? A ?W O R D ? / 5 ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Introduction: Homelessness: What?s in a Word? www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 ernment?funds?to?carry?out?effective?social?housing?programs?(Canadian? Association?of?Housing?and?Renewal?Officials,?1988,?p.?122.).? The?cutbacks?in?social?housing?and?related?programs?began?in?1984.? The?government?ignored?the?1987?Agenda?for?Action.?In?1993?all?federal? spending?on?the?construction?of?new?social?housing?was?terminated?and? in?1996?the?federal?government?further?removed? itself?from? low?income? housing? supply?by? transferring? responsibility? for?most? existing? federal? social?housing?to?the?provinces.?Reliance?on?the?private?market?for?hous? ing?provision?puts?at?a?disadvantage?not?only? those?with? low? incomes,? but?also? those? facing?discrimination? in? the?housing?and? job?markets?on? the?basis?of? race,?gender,? family? status,?disability,? immigration,?age,?or? other?factors.?? Over?the?past?two?decades?we?relied?on?an?increasingly?deregulated? society?in?which?the??genius?of?market?forces??would?meet?our?needs,?in? which?the?tax?cuts,?made?possible?by?program?spending?cuts?that?usually? benefited? poor? and? average? income? people,?were? supposed? to? ?trickle? down??to?benefit?those? in?need.?The?competitive?economy?required,?we? were? told,?wage? suppression? and? part?time? jobs?with? no? benefits.?We? may?now?be?entering?a?new,?very?different?period?caused?by?the?global? financial?crisis???although?this?remains?to?be?seen.? The?dehousing?of?so?many?Canadians?starting?in?the?1980s?was?not? the?result?of?a?natural?disaster?(an?earthquake,?a?flood,?an?ice?storm).?Ca? nadians?are?quick?to?rehouse?people?whenever?a?natural?disaster?leaves? people?homeless.?But?over? the?past? two?decades,? instead?of? continuing? public? policies,? including? appropriate? regulation? of? the? private? sector? where?necessary?for?the?general?public?good,?we?did?the?opposite.?? By? the?early?1980s?countries? like?Canada?needed?a?new? term? for?a? widespread?mass? phenomenon,? a? new? social? problem? found? in?many? wealthy,?developed?nations.?The??odd?job?word,??homeless?ness,?filled?the? gap.?Adding? the? suffix???ness??turns? the? adjective? homeless? into? an? ab? stract?noun.?As?such,?it?allows?readers?and?listeners?to?imagine?whatever? they?want.?It?tosses?all?sorts?of?problems? into?one?handy?term.?We?thus? have? the?ongoing?problem?of?defining?what?homeless?ness? is?and? isn?t.? There?is?no?single?correct?definition,?given?the?different?mix?of?problems? that?goes?into?the?hodgepodge?of?issues,?and?depending?on?who?is?using? the?term.?? ? ? ? ? ? ? ? H O M E L E S S N E S S : ?W H A T ? S ? I N ? A ?W O R D ? / 6 ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Introduction: Homelessness: What?s in a Word? www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 In? short,?we?have?not?used? the?word?homelessness? for?very? long.? It? was?rarely?used?before?the?1980s.?It?is?a?catch?all?term?for?a?host?of?seri? ous?social?and?economic?policy?failures???more?serious?than?in?the?past.? Its?widespread?usage?reflects?what?has?happened?to?Canadian?society??? the?way?we?organize?who?gets?what,?and?our?failure?to?have?in?place?sys? tems?for?meeting?basic?human?needs?in?a?universal,?inclusive?fashion.?It? also?reflects? the? institutionalization?of?a?problem.?We?now?have?a?huge? social? service,? health,? mental? health,? and? research? sector? focused? on? homeless? or? dehoused? people.? This? requires? special? skills? and? knowl? edge.?? What?homelessness?means? We?need?to?be?careful?when?we?use?the?words?homeless?and?homelessness.? While?it?is?true?that?all?societies?through?history?tend?to?have?some?peo? ple?who?are?homeless???without?a?home???we?have?not?always?had?the?set? of?social?problems?we?associate?with?the?word?homelessness.?? Starting? in? the?1980s?homelessness?came? to?mean?a?poverty? that? in? cludes?being?unhoused.? It? is? a?poverty? so?deep? that? even?poor?quality? housing?is?not?affordable.?Canada?has?always?had?many?people?living?in? poverty.?But?it?was?only?in?the?1980s?that?more?and?more?people?found? themselves?not?only?poor,?but?unhoused.?? We?can?at?least?separate?out?the?one?common?feature?shared?by?all? homeless?people? from?all? the?other?complex?social?situations?associated? with?the?word?homelessness.?The?best?summary?of?the?core?of?the?problem? came?from? long?time?U.S.?housing?researcher?and?activist?Cushing?Dol? beare?about?10?years?ago.?It?is?a?statement?I?quote?often.?She?wrote:? The?one?thing?all?homeless?people?have?in?common?is?a?lack?of?housing.? Whatever?other?problems?they?face,?adequate,?stable,?affordable?housing? is?a?prerequisite?to?solving?them.?Homelessness?may?not?be?only?a?housing? problem,?but?it?is?always?a?housing?problem;?housing?is?necessary,?al? though?sometimes?not?sufficient,?to?solve?the?problem?of?homelessness? (Dolbeare,?1996,?p.?34).? Some? people? disagree,? saying? that? homelessness? is? an? individual? problem,? not? a? housing? problem.?Housing? is? an? expensive? problem? to? address.?It?is?simpler?and?cheaper?to?blame?people?for?their?personal?fail? ? ? ? ? ? ? ? H O M E L E S S N E S S : ?W H A T ? S ? I N ? A ?W O R D ? / 7 ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Introduction: Homelessness: What?s in a Word? www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 ures.?We?all?have?our?personal?failures.?But?only?for?some?does?it?mean? finding?themselves?and?their?families?unhoused.?? Homelessness?means?that?we?have?two?kinds?of?health?and?mental? health?care:?one?for?the?housed?population?and?another?for?the?unhoused? population.?? It?means? that? those?already? facing? systemic? inequities,?discrimina? tion,?and?violence?on? the?basis?of?gender,? race,?age,?poverty,?disability,? sexual?orientation,?immigration?or?Aboriginal?status,?now?face?the?possi? bility?of?becoming?dehoused?as?a?result.? It?means?that?we?work?to?create?more?and?better?emergency?shelters? rather?than?assisting?unhoused?people?to?settle?into?adequate,?stable?and? affordable?housing.? It?means? that?Canada?does?not?have?a? tenure?neutral?housing?sys? tem;?that?owners?and?renters?are?treated?very?differently?in?terms?of?sub? sidies?and?helpful?regulations.? This? huge? imbalance? in? the? allocation? of? resources? continues.?We? have? limited? resources? for? the? prevention? of? dehousing? and? for? quick? rehousing.?Most? resources? and? professional? attention? are? focused? on? supporting?people? in? their?homelessness.?This? is? the?situation? in?which? we?are?stuck?today.?We?have?all?the?evidence?we?need?about?the?health? impacts,?including?premature?death,?of?being?unhoused?for?any?extended? period?of? time.?Yet?we? still?give?priority? to? the?homeownership? sector? and?ignore?the?rental?and?social?housing?sectors.?? It?used?to?be?possible?to?say?that?no?one?in?Canada?was?born?home? less.?Unfortunately,?with?so?many?homeless?families? in? temporary?shel? ters,? children? are? today? being? born? into? unhoused? families? across? the? country.?Here?is?a?quote?from?an?experienced?Canadian?veteran?of?home? lessness:?? I?don?t?ever?want?to?go?back?to?being?homeless.?I?d?rather?try?to?do?some? thing?to?prevent?that?happening,?because?everybody?deserves?their?own? place?to?call?home.? This?Canadian?veteran?of?homelessness?is?a?12?year?old?Calgary?girl.?? As?we?write?in?2009,?postwar?progress?in?building?a?middle?income? inclusive? society? in?which? everyone? is? adequately? housed? has? halted.? ? ? ? ? ? ? ? H O M E L E S S N E S S : ?W H A T ? S ? I N ? A ?W O R D ? / 8 ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Introduction: Homelessness: What?s in a Word? www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 Instead? of? rehousing? processes? and?mechanisms,?we? have? had,? for? at? least?two?decades?now,?dehousing?processes?and?mechanisms.?? Hiding?behind?the?word?homelessness? Who?is?in?favour?of?homelessness??Who?lobbies?for?homelessness??Which? economists? tell?us?homelessness? is?good? for? the? economy?? If?no?one? is? doing?these?things,?why?does?homelessness?persist??? Homelessness?does?not?occur? in? a? social?or?political?vacuum.?The? events? that?make?people?homeless?are? initiated?and?controlled?by?other? people.?The?primary?purpose?of?these?activities?of?others?is?not?to?make? people?homeless?? but,?rather,?to?achieve?socially?condoned?aims?such?as?making?a?living,? becoming?rich,?obtaining?a?more?desirable?home,?increasing?the?efficiency? at?the?workplace,?promoting?the?growth?of?cultural?institutions,?giving?cit? ies?a?competitive?advantage,?or?helping?local?or?federal?governments?to? balance?their?budgets?or?limit?their?debts.?Homelessness?occurs?as?a?side? effect?(Jahiel,?1992,?p.?269).? Homelessness?is?the??natural??outcome?of?the?way?we?have?organ? ized?our?housing?system,?and? the?way?we?allocate?or? fail? to?allocate? in? come?and?support? services?when? they?are?desperately?needed.?Though? no?one?favours?homelessness,?many?contribute?to?it?by?doing?what?socie? tal?norms?and?government?laws?and?regulations?allow.?? For? a? long? time? sociologists? and? social?policy? experts?have? recog? nized? the?especially?difficult?nature?of?some?social?problems???which? is? why?some?persist.?Here?is?one?explanation:? a?social?problem?is?an?enterprise?in?finding?ways?of?getting?something? done?or?prevented,?while?not?interfering?with?the?rights,?interests,?and?ac? tivities?of?all?those?who?are?involved?in?the?failure?to?do,?or?the?persistence? in?doing,?what?is?the?subject?of?the?problem?(Frank,?1925).?? This?observation,? from?a?1925?article?on? the?nature?of? social?prob? lems,?refers?to?what?we?might?call?the?tyranny?of?the?status?quo.?A?sig? nificant?majority,?or?at? least?an? influential?minority,?are?doing? fine?and? have?so?far?benefitted?from?the?changes?that?were?made? in?the?1980s?to? the?present.?? ? ? ? ? ? ? ? H O M E L E S S N E S S : ?W H A T ? S ? I N ? A ?W O R D ? / 9 ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Introduction: Homelessness: What?s in a Word? www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 So?keeping?things?the?same?and?tinkering?at?the?edges,?acting?only?at? the? local?community? level?and? individual? level?of? the?problem,?without? addressing? the? larger?dynamics? that? are?producing? the?problem? in? the? first?place,?means,?obviously,?that?the?problem?will?persist.? By?hiding? a?broad? set?of? socially?undesirable?outcomes?under? the? rubric?of?homelessness,?society?can?recognize?and?condemn?the?undesir? able?social?outcome?we?call?homelessness.?No?one?I?know?of?is?in?favour? of?homelessness.?But?simply?condemning?the?problem?while?at?the?same? time?not?doing?anything?to?change?the?social?dynamics?that?produce?the? undesirable? outcomes,?means? that? things?will? stay? the? same? ?? or? get? worse.?In?addition,?the?social?dynamics?creating?the?problem?remain?un? named,?subsumed?under?the?rubric?of?the?abstract?term?homelessness.?The? homeless?makers? carry? on? their?work? and? the? homeless?making? proc? esses?continue.?? Responding?to?homelessness? If?we?are?to?ensure?that?things?do?not?simply?stay?the?same,?or?get?worse,? we?need?to?act?on?two?main?fronts.? First,?bearing?in?mind?Cushing?Dolbeare?s?insight?that?homelessness? is?always?a?problem?of?housing,?we?need?to?focus?on?rehousing?those?who? have?become?unhoused.?Unfortunately,? at?present?we?have? limited? re? sources? for? the?prevention?of?dehousing?and? for?quick?rehousing.?Most? resources?and?professional?attention?are?focused?on?supporting?people?in? their? homelessness.? It? is? urgent? that?we? refocus? our? efforts? on? getting? people?rehoused.? The?second?thing?we?need?to?do?is?to?recognize?that?homelessness?is? not?a?complex?problem.?Yes,?it?is?not?a?complex?problem.?? After?all?these?years?of?research?and?policy?analysis?and?document? ing?the?lived?experience?of?those?affected?and?those?who?provide?support? services,?we?know?what? the?causes?of? the?problem?are.?That?means?we? know?what?the?solutions?are.? When? individuals?or? families? run? into? serious?difficulty? in?one?or? more?of?the?three?key?areas?that?support?a?decent?standard?of?living,?they? may? find? themselves?unhoused?and?potentially?on?a?downward? spiral.? The? three?areas?are:?housing,? income,?and? support? services.?Groups?al? ready?facing?inequities,?discrimination,?and?violence?are?often?the?first?to? ? ? ? ? ? ? ? H O M E L E S S N E S S : ?W H A T ? S ? I N ? A ?W O R D ? / 1 0 ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Introduction: Homelessness: What?s in a Word? www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 face?difficulties?in?these?areas?when?the?economic?tide?changes.?Starting? in? the?1980s,?more?and?more? individuals?and? families?could?not?afford? housing,?or?could?not? find? jobs?or? income? support?at?a? living?wage,?or? could?not?obtain?appropriate?addiction?or?mental?health?support.?? An?adequate?standard?of?living?means?that?a?good?society?not?only? ensures? that?good?quality?health?care? is?available? to?everyone,?but?also? access?to?adequate?housing,?employment?at?a?living?wage,?and?essential? support?services?must?also?be?available?for?everyone,?not?just?those?who? can? afford? them? ?? and? that? systemic? inequities? are? addressed? in? social? policy.?? Homelessness?in?the?plural? We?have? tried? in? this? electronic?book? to? avoid? lumping? together?prob? lems?that?are?distinct,?and?lumping?together?people?who?may?have?little? in? common?beyond? the? fact? that? they?have? experienced? the?dehousing? processes?at?first?hand.?There?is?no?one?face?of??homelessness.??? We? have? organized? the? chapters? into? themes.?One? set? of? themes? represents? disadvantaged? groups? (women,? children? &? youth,? immi? grants,?Aboriginal?people),?another?represents?policy?areas?implicated?in? homelessness? (housing,?health,?or? the? justice?system).?We?also?have? in? cluded?a?section?on? research? issues.?The?contents?of?each?chapter?often? cross?various?themes.?As?this?electronic?book?evolves?over?time,?we?may? need?to?add?further?themes,?if?new?research?emerges?that?does?not?fit?the? existing?themes.? Why?women??Research?shows?that?women?s?reasons?for?homeless? ness?are?often?different?from?men?s,?and?abuse?in?the?home?is?a?primary? factor.?Women?who? are? homeless? are? often? accompanied? by? children,? and?the?housing?and?supports?they?require?differ?as?a?result.?Policy?and? service?responses?to?homelessness?must?take?gender?into?account.? Why?children?and?youth??The?experience?of?homelessness?varies?ac? cording? to?one?s? age?and?defies? easy? categorization.?The?growing?phe? nomenon? of? ?family? homelessness??means? that? increasing? numbers? of? children?experience?homelessness,?with?serious?and?lasting?consequences? for? their? well?being,? development,? and? education.? Youth,?meanwhile,? outnumber? any? other? homeless? group? in? Canada.?Alongside? housing,? income,?and?support?services,?the?child?welfare?system?is?also?implicated? ? ? ? ? ? ? ? H O M E L E S S N E S S : ?W H A T ? S ? I N ? A ?W O R D ? / 1 1 ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Introduction: Homelessness: What?s in a Word? www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 in?homelessness?among?children?and?youth???both?when?children?cannot? be? reunified?with? their? parents? due? to? inadequate? housing,? and?when? youth?become?too?old?for?the?services?of?child?welfare?agencies?but?still? have? insufficient?supports?for?obtaining?and?maintaining?housing.?How? service?providers? respond? to?homeless?people?depends?on?how?we?un? derstand?who?is?homeless?and?why,?and?how?we?recognize?family?rela? tionships?in?that?response.? Why?Aboriginal?people??People?of?Aboriginal?descent?are?overrep? resented?among?homeless?populations?across?Canada.?The?effects?of?eco? nomic?marginalization,?social?exclusion,?and?Aboriginal?policy? intersect? with?devastating?results?for?Aboriginal?individuals?and?families.?Deplor? able?housing?conditions?on?reserves,?high?rates?of?family?violence,?inade? quate?housing?and?supports? for?Aboriginal?people? living? in?urban?cen? tres,? and? historical? legacies? of? residential? schools? and? community? displacement?all?play?a?role?in?Aboriginal?homelessness.?? Why?immigrants?and?refugees??Homelessness?among?these?groups? is?increasing,?because?of?inequities?in?employment,?discrimination?in?the? housing?market,? lack?of? family?and?social?supports,?and?differential?ac? cess?to?services?based?on?immigration?status.?Homelessness?in?immigrant? and? refugee? communities?poses? a?particular? challenge? to?homelessness? services?in?Canada?s?largest?urban?centres,?and?regions?such?as?Southern? Ontario?where?the?majority?of?new?immigrants?settle.? Why?health??Research?clearly?shows?that?the?consequences?of?home? lessness?include?effects?on?health?and?mental?health???and?some?of?these? effects?persist?even?after?homeless?people?secure?housing.?The?response? to?homelessness?must? take? into?account? these?problems?and? their? long? term?repercussions.? Why? the? justice?system??Another? finding? from? the? research? is? that? homeless?people?may?be?criminalized?within?the?justice?system,?and?also? that? they? are? disproportionately? the? victims? of? crime.? Responses? to? homelessness?must?take? into?account?the?way? in?which?those?who?have? been? incarcerated?may? become?dehoused,? and?how? those?who? are?de? housed?may?become?involved?with?the?justice?system.? Why?include?research?issues??In?conducting?research,?it?is?important? not?to?make?quick?assumptions?about?the?lives?and?pathways?into?home? lessness? of? the? many? different? individuals? who? find? themselves? de? ? ? ? ? ? ? ? H O M E L E S S N E S S : ?W H A T ? S ? I N ? A ?W O R D ? / 1 2 ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Introduction: Homelessness: What?s in a Word? www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 housed.?This?section? is? intended? to?challenge?researchers? to?ensure? that? they?are?rigorously?testing?their?assumptions?and?keeping?an?open?mind? towards?new?research?and?new?evidence?in?this?field.?We?have?also?tried? to?include?information?from?across?the?country,?since?homelessness?takes? different?forms?in?different?places???big?cities,?smaller?centres,?suburban? and? rural? areas,? or? the?North.?We? hope? eventually? that? the? book?will? provide?good?coverage?of?the?whole?country.? Policy?options:?Housing,?income,?and?support?services?? Some?might?wonder?about?the?subtitle?of?this?book:?Policy?Options?for?ad? dressing? Homelessness? in? Canada.? The? individual? chapters? are? mainly? about?specific?issues?for?specific?population?groups.?Where?are?the?policy? options??? The?policy?options?emerge? from?a?better?understanding?of?specific? aspects?of?the?many?problems?tossed?into?the?word?homelessness.?People? become?homeless?because?of?serious?problems?arising?in?one?or?more?of? the?three?key?necessities?of?an?adequate?standard?of? living:?housing,? in? come,? and? support? services.?When? systemic? inequities,?misfortune,? ill? health,?or?abuse?interfere?with?people?s?ability?to?hold?on?to?or?obtain?one? or?more?of?these,?a?serious?personal?crisis?can?result,?especially?for?those? with? a? limited?knowledge?of? available?options?or? a?weak? support?net? work.? A?host?of?problems?became?lumped?together?under?the?word?home? lessness? starting? in? the? 1980s? because? the? public? sector?s? provision? of? housing,? income?and? support? services? to? those?most?disadvantaged?by? the?market?system,?which?were? far? from?adequate? to?began?with,?were? systematically?and?continually?cut?back?or?eliminated.?What?do?we?call?a? new?form?of?deeper?and?widespread?destitution?that?now?included?being? unhoused? for?periods??The?word?homelessness?provides? a?good? cover? for?the?impacts?and?outcomes?of?public?policies,?programs,?and?tax?cuts? that? benefited? mainly? higher?income? groups.? If? people? became? un? housed,?it?was?their?fault.?? Neo?conservative?policies?(known?in?the?research?literature?as??neo? liberal?)?were?first?implemented?in?Margaret?Thatcher?s?United?Kingdom? (elected?1979),?Ronald?Reagan?s?United?States? (elected?1980),?and?Brian? ? ? ? ? ? ? ? H O M E L E S S N E S S : ?W H A T ? S ? I N ? A ?W O R D ? / 1 3 ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Introduction: Homelessness: What?s in a Word? www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 Mulroney?s?Canada? (elected?1984).3?Deregulation,?public?spending?cuts,? and?tax?cuts?for?the?well?off?were?supposed?to??trickle?down??to?the?less? fortunate.?The?other?popular?clich??was? that??all?boats?would? rise?with? the?rising?tide??of?wealth.?In?2004,?TD?Senior?Economist?Don?Drummond? and?his?colleagues?at?TD?Financial?concluded:??In?sum,?the?evidence?that? a?rising?tide?lifts?all?boats?is?spotty?at?best???though,?certainly,?it?is?supe? rior? to? a? situation?where? all? the? boats? are? sinking?? (Drummond? et? al.,? 2004,?p.?25).?Five?years?later,?in?fall?2008,?we?learned?that?most?boats?are? indeed?sinking.?? The?huge?economic?surplus?generated?during?the?prosperity?of?the? past? two? decades,? we? now? know,? never? could? and? never? did? trickle? down.? The? global? financial? crisis? and? the? economic? depression? it? pro? duced? are? the? result? of? the? same? policies? that? stripped? lower?income? households? of? essential? housing,? income,? and? support? services.? Even? during?the?period?of?economic?growth,?the?bottom?10?to?20?percent?suf? fered?a?loss?in?their?real?(inflation?adjusted)?standard?of?living,?while?the? top?10?to?20?percent,?even?after?the?financial?collapse?of?late?2009,?reaped? most?of?the?benefits.?With?the?onset?of?the?financial?crisis?we?are?all?now? paying?for?the?implementation?of?an?ideology?that?benefited?those?seek? ing? elected? office? in? Western? liberal? democracies? and? those? able? to? change,?bend,?or?break?the?rules?in?the?financial?sector?in?their?favour.? We?know? the?policy?options.?We?need? social?protections? that?pre? vent?Canadians? from?becoming?unhoused.?We?need?programs? that?en? sure?that?all?Canadians?have?what?they?need?for?an?adequate?standard?of? living,?so?that?no?one?will?be?unhoused?for?more?than?a?very?brief?period? should?a?crisis?of?some?sort?arise.?We?need?policies? that?correct?historic? and? systemic? inequities,? and? that?provide? adequate,? affordable?and? se? cure?housing,?an?adequate?income?or?income?support?when?needed,?and? adequate? support? services? if? these? are? required? (for? addictions,?mental? health,?and?so?on).?Only?then?will?we?begin?to?solve?the?problem?we?now? call?homelessness.?? 3??For?a?history,?see?Harvey,?2000.?? ? ? ? ? ? ? ? H O M E L E S S N E S S : ?W H A T ? S ? I N ? A ?W O R D ? / 1 4 ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Introduction: Homelessness: What?s in a Word? www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 Health?impacts:?Homelessness?kills? If?this?book?were?about?homelessness?in?the?United?States,?lack?of?health? insurance?would? join? the? list?of?causes?and?solutions:?housing,? income,? support?services,?and?health?care.?Without?some?form?of?universal?access? to?adequate?health?care,?independent?from?one?s?income?or?wealth,?peo? ple?can?become?unhoused?and?homeless?after?trying?to?pay?for?the?health? care?services?required?by?a?loved?one.?? Canadians?are?fortunate?in?that?our?universal?health?care?insurance? system,? though? far? from?perfect,?eliminates?almost?all? financial?barriers? to?physician?and?hospital?services.?No?one?loses?his?or?her?housing?due?to? the? cost? of? health? care? ?? through?many? cannot? pay? for? the? prescribed? medications,? dental? care,? or? assistive? devices? they? require.? However,? people?often?are?dehoused?due?to?disability,?chronic?illness,?or?a?sudden? health?crisis,?because? income,?housing,?and?support?services?are? insuffi? cient?to?protect?people?in?these?situations.?Health?conditions?that?greatly? increase? the? risk? of? loss? of?housing? include? severe?mental? illness,? sub? stance?abuse,?and?medical?illness.?? As?well,?being?unhoused?for?any?period?of?time?can?cause?or?exacer? bate?health?problems.?Longer?periods? can? cause? long?term?and?perma? nent?harm?to?health.?People?who?have?been?unhoused?for?long?periods?of? time?die?younger? than? is? the?norm? for? the? rest?of? society.?The? research? literature?on?this?issue?is?unequivocal.?? Among?men?aged?25?to?44?years?who?use?emergency?shelters?in?To? ronto,?mortality?rates?are?four?times?higher?than?among?men?in?the?gen? eral?population?(Hwang?2000).?Lack?of?housing?takes?a?particularly?high? toll?on?women?under? the?age?of?45:? those?who?use?shelters?are? fully?10? times?more? likely? to?die? than? their?housed?counterparts? (Cheung?2004).? This? increased?mortality? risk? has? an? enormous? cumulative? effect? over? one?s?lifetime???in?Canada,?a?typical?25?year?old?man?has?a?64%?chance?of? surviving?to?age?75,?but?a?25?year?old?man?at?a?homeless?shelter?has?only? a?27%?chance?of?living?to?see?his?75th?birthday?(Hwang,? in?press).?These? extreme?health? inequities?can?only?be?addressed?through?a?combination? of?housing,?health?care,?and?social?interventions.? ? ? ? ? ? ? ? H O M E L E S S N E S S : ?W H A T ? S ? I N ? A ?W O R D ? / 1 5 ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Introduction: Homelessness: What?s in a Word? www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 Finding?Home???and?losing?homelessness?? In? this?electronic?book,?we?hope? that?by? taking?apart? the?word??home? lessness??and?revealing?the?many?social?issues?it?conceals,?we?can?begin? to?develop?appropriate?responses.?As? the?e?book?evolves,?we?hope? that? more?and?more?of? the?chapters?will?contain? information?about? the?pro? gress?we?are?making? in?eradicating?homelessness.?Most?of?all,?we?hope? that?in?another?20?years??time,?the?term?homelessness?will?be?obsolete.? This?e?book?project?is?funded?by?the?Social?Sciences?and?Humanities? Research?Council?of?Canada? (SSHRC).? It?represents?an?educational?and? dissemination?partnership?with?Canada?s?national?homelessness?charity,? Raising? the?Roof,?and?with? the?newly?created?SSHRC?funded?Canadian? Homelessness?Research?Network? (CHRN).? The? e?book?will? be? perma? nently? available?on? the?CHRN?s?website,?The?Homeless?Hub,?where? it? will?be?periodically?updated.?www.homelesshub.ca/FindingHome.? ? J.?David?Hulchanski?is?Associate?Director?of?Research?for?the?Cities?Centre?and? a? professor? in? the? Faculty? of? Social?Work,? University? of? Toronto.? Philippa? Campsie? is? an? adjunct? professor? in? the?Program? in?Planning,?Department? of? Geography,?at?the?University?of?Toronto.?Shirley?B.Y.?Chau?is?an?assistant?pro? fessor?in?the?School?of?Social?Work,?Faculty?of?Health?and?Social?Development,? University?of?British?Columbia?(Okanagan).?Stephen?W.?Hwang?is?an?associate? professor?in?the?Department?of?Medicine,?University?of?Toronto?and?a?research? scientist?at?the?Centre?for?Research?on?Inner?City?Health,?St.?Michael?s?Hospi? tal,?Toronto.?Emily?Paradis?is?Research?Associate?at?Cities?Centre,?University?of? Toronto.? References? Basford,?Ron?(1973).?Introduction?to?the?1973?NHA?amendments?by?the?Minister? of?State? for?Urban?Affairs,?Canada,?House?of?Commons,?Hansard,?March? 15,?1973,?p.?2257.? Canada,?Department?of?Finance?(2008).?Fiscal?Reference?Tables,?Table?8.? Canadian?Association?of?Housing?and?Renewal?Officials?(1988).?New?Partnerships? ??Building?for?the?Future:?Proceedings?of?the?Canadian?Conference?to?Observe?the? International?Year?of?Shelter? for?the?Homeless,?September?13?16,?Ottawa:?Cana? dian?Association?of?Housing?and?Renewal?Officials,?1988? ? ? ? ? ? ? ? H O M E L E S S N E S S : ?W H A T ? S ? I N ? A ?W O R D ? / 1 6 ? J. David Hulchanski, Philippa Campsie, Shirley B.Y. Chau, Stephen W. Hwang, Emily Paradis. Introduction: Homelessness: What?s in a Word? www.homelesshub.ca/FindingHome ? Cities Centre, University of Toronto, 2009 ISBN 978-0-7727-1475-6 Cheung,?Angela?&?Hwang? Stephen?W.? (2004).?Risk? of?death? among? homeless? women:?a?cohort?study?and?review?of?the?literature.?Canadian?Medical?Asso? ciation?Journal?170,?1243?1247.? City?of?Toronto?Planning?Board?(1977).?Report?on?Skid?Row.?Toronto.? Dolbeare,? Cushing?N.? (1996).? ?Housing? Policy:? A? General? Consideration.?? In? Baumohl,?J.?(ed.),?Homelessness?in?America.?Phoenix:?Oryx?Press.? Drummond,?Don,?Derek?Burleton?&?Gillian?Manning.?(2004).??Affordable?hous? ing? in?Canada:? In? search? of? a? new? paradigm.?? In:?Hulchanski,? J.D.,? and? Shapcott,?M.?(eds)?Finding?Room:?Policy?Options?for?a?Canadian?Rental?Housing? Strategy.?Toronto:?Centre?for?Urban?and?Community?Studies,?University?of? Toronto.? Frank,?L.K.?(1925).??Social?Problems,?The?American?Journal?of?Sociology,?30(4):468.? Hambrick,?Ralph? S.? Jr.? and?Gary?T.? Johnson? (1998).? ? ?The? future?of?homeless? ness,??Society,?Sept./Oct.? Harvey,?David?(2007).?A?Brief?History?of?Neoliberalism.?Oxford:?Oxford?University? Press.? Hopper,?Kim,?and?Jim?Baumohl?(1996).??Redefining?the?Cursed?Word.??In?Bau? mohl,?J.?(ed.),?Homelessness?in?America.?Phoenix:?Oryx?Press.? Hwang,? Stephen?W.? (2000).?Mortality? among?men? using? homeless? shelters? in? Toronto,?Ontario.?Journal?of?the?American?Medical?Association?283,?2152?57.? Hwang? Stephen?W.,?Wilkins,? R.,? Tjepkema?M.,? O?Campo? P.J.,? Dunn? J.R.? (in? press).?Mortality?among?residents?of?shelters,?rooming?houses,?and?hotels?in? Canada:?An?11?year?follow?up?study.?British?Medical?Journal.? Jahiel,?Rene? I.? (1992).??Homeless?Makers?and?Homeless?Making?Processes.?? In:? Jahiel,?R.?(ed.)?Homelessness:?A?Prevention?Oriented?Approach,?Baltimore:?The? Johns?Hopkins?University?Press.? Ross,?Aileen?D.?(1982).?The?Lost?and?the?Lonely:?Homeless?Women?in?Montreal.?Mont? real:?A.D.?Ross,?distributed?by?the?Canadian?Human?Rights?Foundation.? Social?Planning?Council?of?Metropolitan?Toronto? (1960).?Report?on?Homeless?and? Transient?Men,?Toronto.? ? ? Chapter 1.1 Transitional Housing Models in Canada: Options and Outcomes SYLVIA N OVAC , J OYCE BR OWN , AND CARMEN B OURB ONNAIS There is a growing recognition that some adults, youth, and families who have experienced homelessness need support as well as housing to stabi- lize their lives. Histories of abusive treatment, residential instability, ad- dictions, and mental health issues add to the trauma of homelessness itself. Transitional housing is intended to offer a supportive living envi- ronment, opportunities, and tools for skill development, and promote the development of community among residents. These can be critical in enabling people to participate in employment or training programs, en- rol in educational facilities, address addiction or mental health issues, and ultimately move to independent living in the larger community. Examination of the transitional housing model is timely. Since De- cember 1999, several federal programs ? Supporting Communities Partnership Initiative (SCPI), Shelter Enhancement Program (SEP), and Residential Rehabilitation Assistance Program (RRAP) ? have funded new transitional housing projects for people who are homeless or at risk of homelessness, including Aboriginal people, youth, women, men, fami- lies, and people with health problems or severe mental illness and addic- tions. These programs add to the unknown number of transitional hous- ing programs serving similar populations across the country. Although the transitional housing concept is increasingly being ap- plied to help people ?exit? homelessness, there is no single program model. Just as those affected by homelessness are a heterogeneous lot, transitional housing projects vary widely in the groups served, the goals adopted, the types and levels of services provided, and the outcomes expected. What is Transitional Housing? Transitional housing is an intermediate step between emergency crisis service and long-term permanent housing. It is more long-term, service- intensive, and private than emergency shelters, yet remains limited to stays of between three months and three years (Barrow & Zimmer, 1999). It is intended for people who need some degree of structure, support, supervision, and skill building to move from homelessness into stable, permanent housing. It provides an intermediate step for people who need a safe, supportive place where they can overcome trauma, begin to address the issues that cause homelessness or kept them homeless, and begin to rebuild their support network (Nesselbuch, 1998). Transitional housing programs are usually building-specific and of- fer residents less private space than permanent housing (Sprague, 1991b). Building form and living arrangements range from dormitories to shared rooms with common facilities, single-room-occupancy hotels, dedicated apartment buildings, and scattered-site apartments. The services, which are provided on-site or through community partners, typically include case management and range from alcohol and drug abuse treatment to financial counselling and employment services. Some provide specialized services for childcare, domestic violence coun- selling, and services for HIV/AIDS patients (Burt et al., 1999). As resi- dents become stabilized, the program is expected to help them find per- manent housing (Burt et al., 2002). Programs tend to cluster at the ends of a continuum, from service- intensive facilities with rigorous expectations of residents (i.e., high de- mand) to programs with flexible requirements and optional services (i.e., low demand). Low-demand transitional housing programs are designed for chronically homeless individuals and added to outreach or drop-in services; high-demand programs are designed for families and individu- als with multiple problems (Barrow & Zimmer, 1999). The distinction between emergency shelter services and transitional housing may become blurred when shelter stays lengthen. For example, there is currently no standard length of stay in Toronto shelters, and it is not uncommon for families to stay up to one year. Shelters are becoming ?more specialized and flexible to meet new needs within the homeless population? (City of Toronto, 2002, p. 4). Transitional housing resembles supportive housing. Novac and Quance (1998) distinguish transitional from supportive housing only in terms of length of residency ? supportive housing is permanent. Both models encompass a combination of housing and support service provi- sion that varies in terms of housing form, type and level of support ser- vices, target population, and relationship between the housing provider and the support service provider, if different. They differ in that transi- tional housing is a stage in a progression from which residents are ex- pected to ?graduate? to more independent or ?normal? housing (Barrow & Zimmer, 1999). There is also an assumption that some kind of personal change will occur. Another difference is that supportive housing resi- dents commonly have full tenure rights. Residents of transitional hous- ing are expected to vacate when they have completed the program and can be ?dis-enrolled? (evicted) at any point if they violate the program?s rules or do not fulfil its expectations. It is typical to require residents to agree to a contractual requirement to work towards particular goals dur- ing their stay (Sprague, 1991b). The distinction between transitional housing and residential treat- ment programs of recovery and rehabilitation is also murky, in part be- cause of the prevalence of severe mental illness and substance abuse among the visibly homeless (Barrow & Zimmer, 1999). Other terms used for transitional housing include second-stage or bridge housing and ser- vice-enriched housing. Sprague (1991a) uses the term lifeboats to describe transitional housing projects designed for lone-mother-led families, many of them homeless because of family violence. In Canada, the term ?second-stage housing? is applied to transitional housing for women who have come from family violence shelters. Although similar in many respects, this type of transitional housing is not considered here. Who Does Transitional Housing Serve? The need for transitional housing for people in certain circumstances is not new. Victims of crises or family violence, substance abusers, persons with chronic medical problems, immigrant populations, and deinstitu- tionalized persons of all ages have traditionally required transitional housing on the road to independent community living. Halfway houses, independent living programs, and homes for unwed mothers are all fa- miliar examples of transitional programs (Sprague, 1991b). People who benefit from the longer time frame and targeted ser- vices provided by transitional housing include those who: ? are recovering from traumas such as domestic violence or extended homelessness; ? have a background of multi-generational poverty and do not have a kinship network or role models to support their move to self- sufficiency; ? are emancipated youth or younger adults coming out of institutions or having little or no independent living experience; ? are in need of education and job skills in order to obtain an income level sufficient to afford housing; or ? have other on-going service needs such as mental health problems, drug or alcohol treatment, or HIV/AIDS (Nesselbuch, 1998, p. 2). Sprague (1991b) has identified additional groups who are assisted by transitional housing that provides peer support, life skills training, or extensive supervision: ? young mothers and pregnant teenagers; ? physically or mentally disabled persons; ? those leaving prison; ? immigrants. The first major survey of transitional housing programs in the United States showed most serve more than a single group. Of those that specialize, most serve people with mental health or addiction problems; the other major groups in descending order of frequency are abused women, families, youth, and people with HIV/AIDS (Burt et al., 1999). Transitional housing is considered more appropriate for some groups than others. People in recovery from substance abuse was the group most frequently named by service providers as needing the transi- tional environment, to keep them from returning to neighbourhoods and acquaintances where they would have trouble avoiding drugs and alco- hol. U.S. policy and funding programs have favoured the provision of transitional housing for homeless families, but families are increasingly being placed in permanent housing units coupled with supportive ser- vices until their crisis has passed (Burt et al., 2002). Burt et al. (2002, p. 41) characterize transitional housing programs as ?interim placement for persons who are not ready or do not have ac- cess to permanent housing.? Achieving ?housing readiness? implies in- dividual change in behaviour or circumstances; this is the essence of transitional housing. But to what extent do transitional housing pro- grams temporarily house people who simply lack access to permanent housing? This question reveals the core debate on the transitional hous- ing model. A Model under Debate Transitional housing has operated for more than two decades in the United States and continues to be developed for this purpose, but some communities are reconsidering the importance, role, and appropriate clients of transitional housing and prefer to offer permanent housing with transitional support services (Burt et al., 2002). Communities with very low vacancy rates and little affordable housing tend to place a higher priority on the need for transitional housing. In other words, in- creased reliance on transitional housing can be an outcome of insufficient affordable housing units (Nesselbuch, 1998). As a remedy for homelessness, transitional housing is controversial. While proponents consider it the best way to ensure that homeless fami- lies and individuals get services that enable them to attain and sustain self-sufficiency as well as permanent housing, critics view it as stigmatiz- ing and a drain on resources better used for permanent housing (Barrow & Zimmer, 1999). Placing the emphasis on transitional support services rather than temporary housing appears to resolve much of the criticism. Many concerns raised by critics are addressed in newer models of transitional housing that help people access permanent housing and provide support services to enhance stability and self-sufficiency. Based on experience thus far, these new models seem to provide an effective way to assist people in the transition from homelessness without putting them in an institutional living environment (Nesselbuch, 1998, p. 5). Reviewing strategies used in European countries, Harvey (1999) distinguished three models of homeless resettlement strategies: ? normalization, which moves people directly into normal housing; ? tiered, which provides one or more stages before moving to normal housing; and ? staircase of transition, a series of stages, with sanctions in progress toward normal housing. The normalization model downplays personal problems among homeless people and stands in opposition to the model of transitional housing. In Germany, most participants have adapted to their new envi- ronments with little or no difficulty; only a minority of residents re- quired occasional intensive crisis support. The tiered model assumes that transitional housing is necessary for some homeless people. Scattered-site supervised apartments are used for a few months up to two years before participants move to permanent housing. In Vienna, 84 percent of the participants achieved residential stability and 30 percent obtained employment (the local unemployment rate was low at the time). A typical staircase process includes an assessment stage in a shelter, followed by two stages of transitional housing (e.g., six months in a ?training? apartment, then one year in an ordinary apartment), and fi- nally, a move to a regular apartment with full tenancy rights. At each step, the level of support services decreases and the level of tenancy rights increases. Tenants who have difficulties or cause problems may be ?demoted.? Social workers may enter units for inspections (e.g., drug testing), and programs may include mandatory work plans. The out- comes for participants have been mixed. Many homeless people stay stuck at the bottom of the ladder. Others remain stuck near the top, still subject to contractual agreements with private-sector landlords who are reluctant to relinquish control by granting them full rights. Levels of homelessness were not reduced in the Swedish cities that adopted the staircase system. Harvey argues for the normalization model (which he believes may be the most effective in reducing institutionalization) and against the staircase model (which tackles the management issues of capacity for independent living, ?difficult? tenants, and anti-social behaviour, but is intrusive). He concedes that the tiered model is the most common and can be effective, especially when employment status can be improved. What Harvey calls the tiered model best approximates the North American model of transitional housing. Each of the re-settlement mod- els provides participants with permanent housing on program comple- tion. While all transitional housing programs in North America provide participants with assistance in locating and obtaining permanent hous- ing on program completion, they do not all provide affordable, perma- nent housing; this would appear to be a key distinction in success rates. Program Objectives The objective of transitional housing is to provide people with the struc- ture and support they need to address critical issues necessary to main- tain permanent housing and achieve self-sufficiency. At a minimum, ?graduates? are not expected to use a shelter or become homeless again. Transitional housing programs nonetheless vary considerably in their demands and expectations of participants, according to the sub- groups targeted for services, the way barriers to stable housing are con- ceived and approached, and the guiding philosophy about how to over- come those barriers (Barrow & Zimmer, 1999). Some programs are flexible about what participants should do or accomplish during their stay. Some low-demand programs designed to get chronically homeless people off the street initially focus on attracting participants and then only gradually encourage them to alter their be- haviour, such as improving hygiene and accepting health care services. For example, a major objective of one such program is to re-engage cli- ents with the mental health system (Blankertz et al., 1992). Others have a core of activities in which participation is mandatory. Caseworkers may also establish individualized or tailored goals for participants. In many programs, participants are required to: ? open a savings account and initiate a savings plan; ? request a copy of their credit report as soon as they enter the pro- gram; ? participate in education, job training, or employment services; ? for clients with mental health disabilities, receive mental health ser- vices as recommended by a mental health professional; ? for clients in recovery, participate in drug and alcohol programs (Nesselbuch, 1998). Programs for families usually try to promote better parenting. Some family programs even have objectives specifically for children. A transi- tional housing program for families in Calgary includes two objectives for children: to improve their school performance and diminish their in- volvement with the law (Datta & Cairns, 2002). Family reunification may be a program objective. An innovative transitional housing project in New York City is designed to reunite children with their mentally ill homeless parents after lengthy separa- tions (Emerson-Davis, 2000). Strengthening social networks and improving community connec- tion may be included in program objectives. A Canadian program for refugee families was designed to increase the size of families? commu- nity social networks and reduce their sense of isolation (Wiltshire, 1993). This range of objectives has implications for evaluation; to the ex- tent that objectives differ, programs cannot be compared with one other. Since all programs aim to improve housing status, that aspect is compa- rable, although it may be measured in different ways. Indicators of Success Not surprisingly, since the predominant or underlying goal of transi- tional housing is to increase economic self-sufficiency, the most com- monly applied indicators of participants? success are: ? stable residency, once permanent housing is provided; ? greater reliance on employment earnings, rather than income sup- port programs; ? increased income from employment or benefit programs. What constitutes stable residency or ?exit? from homelessness? Re- searchers have applied different definitions of ?housing success? to evaluate the outcomes of transitional housing programs. In many stud- ies, achieving stable residency simply means not using a shelter again. Frequently, this determination is made when residents leave a program. Few evaluations have attempted to determine former residents? housing situation beyond a follow-up period of 12 months, so long-term housing stability has rarely been defined or measured. Wearne and Johnson (2002) argue that ultimately the type of ac- commodation secured on leaving transitional housing is the best meas- ure of a program?s success, with long-term housing generally regarded as the best possible outcome. But what qualifies as ?long-term housing?? And what constitutes adequate housing? Griggs and Johnson (2002) cite an Australian study of transitional housing in which 10 percent of the residents moved to trailer parks or hotels and argue that this should not be considered an adequate housing outcome. Griggs and Johnson (2002) also question the validity of conventional exit data (i.e., no recurrent use of the homeless service system and the housing outcome immediately following service intervention) as ade- quate measures for evaluating transitional housing programs. They rec- ommend an objective hierarchy of housing outcomes; the measurement of non-housing related outcomes, such as improved health; and the use of longer-term outcome measures, especially as homelessness tends to reflect a state of long-term housing instability. Stern (1994) notes the lack of clear operational, and thereby measur- able, definitions of terms such as ?adequate housing? and comments that while some housing options are obviously desirable, such as a family renting or owning an apartment, other options are not as clearly desir- able. Dordick (2002) discounted the outcomes of one program because most of the participants ended up moving in with family or friends. To press the point, Stern (1994) asks: would moving into an overcrowded house with relatives, while potentially permanent, be acceptable? Fischer (2000) considers this an acceptable solution, at least for cer- tain groups, and provided the situation is not overcrowded. Since not everyone can establish an independent household, he argues that mov- ing in with family or friends was the best possible outcome for some of the young mothers in the transitional housing project he studied. He concluded that transitional housing served as a temporary, yet stable, environment from which the young mothers could mend or build rela- tionships that could sustain them in future. However, he overlooks the question of family violence within the context of outcomes. When a meta-evaluation of about 500 transitional housing programs reported that the number of former residents that left to live with friends or family almost doubled (from 12 to 21 percent), Matulef et al. (1995) admitted that this outcome could be interpreted either positively (reuni- fication of children and parents) or negatively (could involve overcrowd- ing, domestic violence, or indicate lack of economic independence). To measure improvements in financial independence, researchers have generally relied on indicators such as employment, job training, and upgrading education credentials. In most cases, these are presented as dichotomous (i.e., yes/no) variables. Depending on the client group and their personal situation, other indicators related to changes in behaviour or skills have been formulated (e.g., abstinence for the alcohol- and drug-dependent, learning English or French for refugees, leaving prostitution for young sex-trade workers). In one study of transitional housing for homeless veterans with psychiatric disabilities, the indicators of success were defined as maintaining sobri- ety or stability and continuing to work without rehospitalization for the duration of the study (Huffman, 1993). For a transitional housing pro- gram for families, the measures included performance of various tasks: cooking regular meals, sending kids to school, washing clothes regularly, keeping house clean, paying bills, keeping appointments with others, having more stable relationships, and having feelings of greater control in their lives (Rice, 1987). Datta and Cairns (2002) used indicators of psy- chological well-being (self-confidence and self-respect), social skills (healthy relationships), and household management (budgeting skills). Other indicators used in evaluation of supportive housing include re- duced admissions to hospital and crisis centres, and reduced number of days of impatient care (e.g., Hawthorne, 1994). Many characteristics that may be valuable in avoiding homelessness are not easily quantified, such as self-esteem, job skills, access to re- sources, community involvement, increased physical well-being, and happiness (Stern, 1994). In some programs, individualized goals are ne- gotiated between worker and participant; these may be highly specific, such as learning particular parenting skills. Goal Attainment Scales using mutually determined indicators are sometimes used to track change. In some cases, the path to success is paved with many small steps. One low-demand respite residence in Toronto serves chronically home- less women who are considered non-compliant and treatment-resistant. Several ?soft? indicators of progress were derived from data collected during the program?s first two years of operation: ? the first cohort of residents gradually reversed their pattern of sleep- ing during the day (an adaptation to avoid attacks at night when sleeping rough) to sleeping at night; ? residents? relationships with each other and with staff improved; ? residents awareness of behavioural and spatial boundaries in- creased; ? residents? involvement in the development of rules increased. Slight improvements in the residents? behaviour, appearance, and physical health were recorded. After two years of operation, two out of fifteen residents had established households in self-contained apart- ments, and two returned to living on the street. Some of the other resi- dents made modest gains toward independence (Novac, Brown, & Gal- lant, 1999). Transitional housing programs have been developed on the as- sumption that the services provided during the transition period will equip homeless individuals and families to maintain residential stability after they move on. Only long-term outcome research can test the vari- ous assumptions, for instance, that clinical and life skills services will enable individuals and families to weather the kinds of events and crises that previously resulted in homelessness and thus will contribute to residential stability (Barrow & Zimmer, 1999). Program Outcomes: Canadian Research Program evaluation of homeless services is not a high research priority in Canada, despite its apparent usefulness for effective program design and implementation, user satisfaction, and responsiveness to clients? needs. A review of 70 homelessness studies conducted within or about the Greater Vancouver Regional District categorized only eight as evaluative; the majority consisted of environmental scans and needs as- sessments (Quantz & Frankish, 2002). Barrow and Zimmer?s (1999) synthesis of the U.S. literature on tran- sitional housing points to a lack of research on program outcomes and effectiveness, especially compared with the extensive documentation of service providers? experience and knowledge. Even the latter type of documentation is sparse for transitional housing programs in Canada. Studies of transitional housing projects in Canada are rare. Only two evaluative studies have been conducted on projects for families. Rice (1987) evaluated a two-year transitional housing program for multi-problem, poor families who lacked the skills and knowledge to cope with the demands of daily living. This included families with a his- tory of bad debts; an inability to pay rent on time; a record of abusive behaviour towards neighbours, property, and family members; and those considered ?poor risks? by landlords. The researchers followed the progress of 25 families who entered the program. Staff expected the families to stay for two years, but only one family did so. At first, weekly meetings were mandatory and fo- cused on life skills (i.e., child rearing, money management, nutrition, maintenance, and communication); these evolved into discussions of common issues and were eventually replaced by individual meetings with staff. Families resisted periodic evaluation and feared eviction for violation of expectations of unit maintenance and childcare. A more tra- ditional form of casework intervention evolved as clients withdrew and reacted with resistance to the structured programming. On average, the families that stayed longer demonstrated more im- provement in their skills. Of the 25 families, eight improved their level of functioning, ten stabilized their ability to function, and seven were worse off. Those with the least severe problems benefited most. Although the families were promised priority for permanent subsidized housing after completion of the program, this did not occur. Only one family moved into subsidized housing after staying in the program for 16 months. Rice concluded that participation in programs should not be mandatory, and families should be provided with permanent housing and transitional support services that are withdrawn over time. Wiltshire (1993) conducted a qualitative evaluation of a short-lived, innovative transitional housing project for government-sponsored refu- gees or refugee claimants identified as needing extra settlement support. Eleven households were placed either in townhouses within a multicul- tural housing co-op or in apartments in a residential area, all managed by the same organization, for up to one year. Family group meetings were initially offered every two weeks and attendance was voluntary. Earlier support sessions focused on discussing common problems and sharing information were better received than later workshop sessions on permanent housing and employment. Based on interviews with 18 individuals (program participants, staff, and board members) and a group interview with six volunteers, Wiltshire determined that the families appreciated the quality of housing provided and the support they received, especially practical assistance, such as opportunities to practice speaking English, and a lessened sense of isolation and alienation. The program succeeded in integrating the families in their neighbourhood community, especially those in the housing co-operative. In fact, the families resisted leaving their homes and the social networks they had developed in the co-operative or neighbourhood and the schools their children attended. The families housed in the co-operative were eventually allowed to become permanent co-operative members, in the process removing the housing stock from the program?s resources. Wiltshire suggested that displacement after one year did not meet the needs of the newcomer families and that a more suitable model would be a brief program of several weeks or permanent housing placement with support services that wane as program participants are integrated into Canadian society. She concluded that the transitional housing concept may have contravened the goal of settlement because refugees benefit from establishing a permanent household and informal support system as soon as possible. In fact, this is true of all families. Based on a review of studies, Bar- row and Zimmer (1999) found that scattered-site models of transitional housing that ?convert? to subsidized permanent housing are a cost effec- tive approach to helping families exit homelessness without the disrup- tion of support networks that facility-based approaches may entail. U.S. Research It is U.S. government policy to provide funds for supportive and transi- tional housing with the goal of reducing homelessness. A survey of pro- gram directors of 360 transitional housing projects funded under the Transitional Housing Program found that 40 percent of clients overall obtained housing and a source of income when they left the program. Families and couples without mental health or addiction problems were most likely to succeed (United States General Accounting Office, 1991). A national evaluation of about 500 transitional housing programs was conducted in the mid-1990s and provided more detail on the clients and program outcomes (Matulef et al., 1995). As the funding program targeted families and persons with disabilities, this influenced the char- acteristics of the groups served. Forty-three percent of the participant households were families with dependent children. More than one- quarter of the projects were intended to assist the severely mentally ill or substance abusers. Ten percent primarily assisted battered women. The proportion of projects assisting other target groups was small, but in- cluded runaway or abandoned youth, veterans, pregnant women, dually diagnosed, developmentally disabled, elderly, and ex-offenders. Virtually all of the projects offered case management, which in- cluded needs assessment upon entry, periodic reassessment and pro- gress monitoring, group meetings, and resident enrolment in commu- nity-based service programs. Most also provided housing location services, training in household management, prevocational training, and vocational counselling. Fewer than half offered prenatal care, medication monitoring, detoxification, English as a second language, physical ther- apy, sheltered workshops, or Parents Anonymous. Matulef et al. (1995) concluded that the Transitional Housing Pro- gram had achieved its goal of helping residents achieve self-sufficiency and find independent living situations. Overall, 57 percent of partici- pants who entered a program completed it. Of those who completed programs, 70 percent moved on to stable housing, some with rent subsi- dies, and most without services. This outcome varied by sub-group, ranging from 90 percent for families to 41 percent for abused women. Of those who withdrew from the program early or were dismissed, less than one-third entered stable housing. This difference in outcomes sug- gests that participation in transitional housing programs increased resi- dents? odds of obtaining stable housing; however, the reliance on data from project sponsors and service providers (some of whom did not have detailed records), lack of data on long-term outcomes, and the lack of a control group comparison limits the conclusions that can be drawn. Twice as many of the participants were employed part- or full-time by the end of the program (38 percent) or engaged in education and training (14 percent) than when they began. A small percentage (11 per- cent) had increased their monthly income and reduced their reliance on income support programs. This was not the case in projects serving abused women, among whom employment status remained unchanged. Barrow and Soto (1996) conducted one of the very few studies that have incorporated a comparison group in the research design. They evaluated six transitional housing programs serving distinctive but over- lapping segments of the street homeless population. Outcomes for a sample of 113 individuals were compared to those for a matched control group who received similar non-residential services (i.e., money man- agement, entitlements, physical and mental health care, substance abuse, legal, and family), but not transitional housing. At program exit, 62 per- cent of the residents went on to some form of longer-term housing (usu- ally to an apartment or room of their own; in some cases to live with family or friends) and remained housed at the three-month follow-up point. This outcome was significantly better than that of the control group in shelters ? 35 percent of them were housed after a period of receiving similar non-residential services. Transitional programs for homeless individuals with severe mental illness frequently emphasize clinical outcomes and include post-program moves to supportive housing and specialized residential care. For exam- ple, Blankertz et al. (1992) reported that more than three-quarters of the residents took their medication regularly; virtually all were receiving income assistance and other help; and two-thirds had no psychiatric cri- ses while in residence. Almost one-third moved to board and care sites; one-quarter attained independent living; and about one-tenth went to specialized care centres, back to family, or to other mental health facili- ties, respectively. Interpretation of results across programs is difficult, given high rates of attrition. For example, Murray et al. (1997) reported that 92 per- cent of residents who completed a transitional residential program main- tained their housing one year after discharge. However, more than half of the sample of 228 individuals failed to complete the program. Assessment of Resident Characteristics and Outcomes No single characteristic of residents assessed so far has distinguished individuals? odds of success. Barrow and Soto (1996, 2000) found no rela- tionship between housing outcomes and characteristics such as gender, age, psychiatric disability or addiction, ethnicity, length of time home- less, main means of support, sleeping place, and pre-baseline services. However, a particular constellation of characteristics was associated with negative outcomes. Those who left or were discharged without place- ment tended to be women, were in their forties, had the most severe psy- chiatric diagnoses, and were actively abusing substances when admitted to the program. Hawthorne et al. (1994) also determined that various socio-demographic and clinical factors, including diagnosis, age, gender, number of previous hospital or crisis centre admissions, employment and living situation, and length of stay, were not related to successful treatment outcomes. Low-Demand vs. High-Demand Housing Barrow and Zimmer (1999) found that adding low-demand transitional housing programs to outreach or drop-in services for homeless indi- viduals improved their likelihood of obtaining permanent housing. High-demand or highly structured facilities which double as treatment programs for people with severe mental illness and/or addictions appear to improve housing and clinical outcomes for participants who complete the programs. Such programs, however, have extremely high attrition rates and are not an effective route out of homelessness for most people. How Appropriate is Transitional Housing for Families? There is considerable disagreement on the appropriateness of transi- tional housing for families. Based on the results of a survey of 40 women living in transitional housing projects (mostly second-stage housing) in Canada, Wekerle (1988) concluded that while the primary goal of offer- ing residents a respite and services to assist them in becoming independ- ent was met, the risk of housing insecurity and homelessness remained. She argued that the transitional housing model was a stop-gap measure that delays rather than resolves the long-term housing problems of these hard-to-house women. Twiss (1993) argued that transitional housing is more appropriate for the deinstitutionalized, the mentally ill, and those with substance abuse problems than for families, especially if the housing form is group home arrangements. An early study by Phillips et al. (1988) reported that within a few months, families had lost the gains they had made during residency in a transitional housing program. Most (71 percent) of the parents who completed a three-month program for homeless families improved their parenting skills, but on follow-up three and six months later, the pro- gress families had made was lost, and their housing facilities had dete- riorated (e.g., there was no furniture). Yet certain families have been more successful in becoming re- housed than other groups in transitional housing programs. An evalua- tion of U.S. transitional housing programs by Matulef et al. (1995) showed that, of those who completed their programs, families were more successful in securing permanent housing than those with severe mental illness (74 percent), addictions (67 percent), or abused women (61 percent). Since these categories are not mutually exclusive, this result can be interpreted to mean that families without problems of severe mental illness, addictions, or recent family violence are more likely to be suc- cessful than families or individuals with these problems. An essential element in stabilizing families is the provision of hous- ing subsidies. Shlay (1993) followed two cohorts of families for more than a year after they completed a two-year transitional housing pro- gram. The families were selected for likelihood of success. They had been screened for chemical dependency, perceived motivation to achieve eco- nomic independence, and potential for becoming trained for the labour market or employed. The program graduates maintained their residen- tial stability after receiving housing subsidy vouchers, and both adults and children exhibited positive changes in their lives. The families, how- ever, did not become economically self-sufficient as indicated by com- plete independence from income maintenance programs (Shlay, 1994). The largest evaluation of transitional housing for homeless families, conducted by Rog et al. (1995), showed a similar result. Data on some 1,670 homeless families in nine cities found considerable housing stabil- ity over time among families who received housing subsidy vouchers, with 91 percent using them after 12 months and 75 percent after 30 months, but little difference in families? increased self-sufficiency. Even homeless families with very complex problems have become residentially stable with the provision of permanent subsidized housing and short-term support services. In a large study of services-enriched housing programs for chronically homeless families in nine U.S. cities, a high proportion of the 781 mothers experienced childhood risk factors, were poorly educated, had health problems, had experienced domestic abuse, and were alcohol and drug dependent (Rog et al., 1995). Despite these problems, 88 percent remained housed 18 months after they had been given housing subsidies and received at least four months of sup- port services. Families have achieved housing stability, especially when provided with affordable housing, but not the other main outcome frequently ex- pected of transitional housing ? financial independence. Gerstel et al. (1996) argue that transitional housing programs fail to help families be- come financially self-sufficient because support services, although help- ful to some residents, is not effective in re-housing participants unless the fundamental shortfall between income and housing costs is ad- dressed. Moreover, the social and physical isolation caused by transi- tional housing programs separates individuals from their support net- works and thereby undermines useful contacts and collaborative strategies of mutual assistance, especially those related to employment and informal housing resources. Fogel (1997) has challenged the premises of high-demand programs, asking how they can promote self-sufficiency when they require resi- dents to adhere to rules on parenting chores, living mates, eating times, entertainment, sleeping and waking times, smoking locations, visitors, mail, medication, money use, overnights, and limitations on bedroom space. Gerstel et al. (1996) also criticized the constraints on residents? daily activities, calling them a form of incarceration for families. They noted prohibitions against in-room visits by outsiders, curfews for adults as well as children, and limitations on the amount of time that residents could spend away from the housing, and found that some programs of- fered residents no opportunity for collective or collaborative decision making. A small-scale study by Dunlap and Fogel (1998) underscored the difficulties families face. A year after completing a transitional housing program, some families were on the verge of homelessness again (e.g., living in a motel, moving from place to place). Most parents were inse- curely employed in low-wage jobs with minimal benefits, and all re- quired public assistance to meet their basic needs. Even two years later, the families were only beginning to attain economic self-sufficiency. Given the challenges of raising children while living in poverty or on low incomes, it is unreasonable to expect all families to become finan- cially independent, but the evidence suggests that they can maintain permanent housing if it is affordable, and that permanent housing with transitional support services is more effective than transitional housing. Whether this is also the case for individuals cannot be answered with the limited research conducted to date. Research Gaps Major gaps limit our ability to assess the effectiveness of transitional housing as a means of addressing homelessness in Canada. The lack of rigorous research on outcomes makes it difficult to evaluate effectiveness The knowledge base for transitional housing practice and research is still too limited to ascertain which practices and program models are most effective in helping formerly homeless people to stay adequately housed. Published studies frequently lack control or comparison groups. ?To as- sess the effectiveness of transitional housing requires research designs that control for other factors that may influence outcomes while compar- ing transitional housing programs to policy-relevant alternatives? (Bar- row & Zimmer, 1999, p. 4). Case management is a common program component, but its connection to outcomes is not known Case management is the factor most often cited by program directors as contributing to client success (Datta & Cairns, 2002; Matulef et al., 1995). However, how it does so is unclear. We lack studies that would clarify the effects of various styles of case management and to determine which aspects of case management or its elements may be fundamental re- quirements for resident success. The long-term effects of transitional housing are unknown We lack sufficient data on whether people maintain their housing over the long term. The challenge is to devise valid indicators and outcome measures of the long-term success or failure of housing assistance pro- grams and of specific service practices and designs (Griggs & Johnson, 2002). Conclusions Transitional housing is an intermediate step between emergency crisis service and long-term permanent housing, the objective of which is to establish residency stability. It combines short-term housing and support services, which vary in type and degree of flexibility, for people who are not ?ready? for permanent housing; or, to its critics, for people who sim- ply lack access to housing. Transitional housing programs are more effective than services alone Short-term provision of housing is more effective in ending homeless- ness than services alone, although the evidence is limited. A comparison study of participants in transitional programs for the street homeless in New York City found that close to two-thirds of the experimental group members, who were provided with temporary housing as well as access to support services, were living in permanent housing three months after leaving the program, compared to only one-third of the comparison group members who had received the same level of services but were not provided with temporary housing (Barrow & Soto, 1996). There is evidence of short-term success in improving housing status Virtually all evaluative studies of transitional housing have demon- strated some degree of post-program improvement in housing status and a significant reduction in the number of residents who return to a state of homelessness on exiting the program. Overall, about half of par- ticipants go on to permanent housing; a much higher proportion obtain housing among those who complete their programs (Barrow & Zimmer, 1999). Some transitional housing projects have provided subsidized housing or housing subsidies for their graduates; not surprisingly, these projects have higher rates of success in achieving permanent housing. All programs offer assistance in locating and obtaining housing, but not necessarily housing that is affordable or desirable to participants. Some programs that encourage chronically homeless people with severe mental illness to accept moves to supportive housing have met resistance from residents who would prefer conventional private-sector rentals, even though such accommodation is generally unaffordable to them (Barrow & Soto, 2000). It is unknown whether improvements in housing status are maintained over the long term, but the small number of stud- ies that have followed former residents, usually for up to twelve months, have shown only a small degree of drop-off in housing status during that relatively brief time. Only modest improvements in financial independence are achieved Improvements in financial and employment status have been modest, especially among families. A variety of other changes in behaviour, ac- quisition of skills, or health status have been reported. Whether transi- tional housing is the best means of promoting such change is unknown. Canadian experience and research is limited and calls into question the appropriateness of the model for families Transitional housing is a relatively new model of service provision in Canada. Consequently, documentation of existing projects is scarce, and evaluative studies even more so. In part, this is because service providers lack the funding and other resources to conduct program evaluation. Both Rice (1987) and Wiltshire (1993) concluded their respective studies of a transitional housing project by questioning the appropriateness of the transitional housing concept for families and suggested that perma- nent (subsidized) housing with transitional support services best pro- motes stable social connections and neighbourhood supports. Permanent housing and community services are critical to the success of transitional housing There is a broad consensus that transitional housing can be an effective component of the range of resources required to prevent homelessness only if adequate permanent housing and supportive community-based services are also available (Barrow & Zimmer, 1999; Nesselbuch, 1998). There are important Canadian?U.S. differences in transitional housing It appears that the Canadian experience of transitional housing projects differs in some respects from that in the United States. There are fewer projects for families versus individuals in Canada, likely due to the higher costs of housing and support service provision for families, and, until recently, the lack of government funding to develop transitional housing or to target programs for homeless families. There may be more projects, proportionately, for single youth ? if so, it is unclear why. There may be a higher proportion of flexible programs that focus on access to services rather than individual change in behaviour. Key indicators of this distinction are eligibility criteria, the extent and rigidity of rules and restrictions, and the basis for involuntary pro- gram discharges. Programs that focus on behavioural change or treat- ment usually require applicants to demonstrate motivation and mandate participation in daily program activities. Programs that focus on access to services are more flexible about program compliance, more forgiving, and less structured. Some conduct outreach to entice those estranged from the service system to enter a program and only gradually encour- age any change in individual behaviour. Sylvia Novac is a research associate at the Centre for Urban and Community Studies, University of Toronto and has written extensively on homelessness. Joyce Brown is Executive Director of Ontario Council of Alternative Busi- nesses, an organization that provides employment for psychiatric con- sumer/survivors. References Barrow, S., & Soto, G. (1996). Closer to home: An evaluation of interim housing for homeless adults. New York: Corporation for Supportive Housing. Barrow, S., & Soto Rodr?guez, G. (2000). Closer to home: Interim housing for long- term shelter residents: A study of the Kelly Hotel. New York: Corporation for Supportive Housing. Barrow, S., & Zimmer, R. (1999). Transitional housing and services: A synthesis. In L. Fosburg & D. Dennis (Eds.) Practical lessons: The 1998 National Sympo- sium on Homelessness Research Washington: U.S. Department of Housing and Urban Development, U.S. Department of Health and Human Services. Blankertz, L., Cnaan, R., & Saunders, M. (1992). Assessing the impact of serving the long-term mentally disabled homeless. Journal of Sociology and Social Welfare, 19(4), 199-220. Burt, M., Pollock, D., Sosland, A., Mikelson, K., Drapa, E., Greenwait, K., & Sharkey, P. (2002). Evaluation of continuums of care for homeless people. Wash- ington, D.C.: U.S. Department of Housing and Urban Development, Office of Policy Development and Research. Burt, M., Aron, L., Douglas, T., Valente, J., Lee, E., & Ewen, B. (1999). Homeless- ness: Programs and the people they serve. Findings of the National Survey of Homeless Assistance Providers and Clients. Washington, D.C.: Interagency Council on the Homeless. City of Toronto (2002). Toronto shelter standards. Toronto: Toronto Community and Neighbourhood Services, Shelter, Housing and Support Department. City of Toronto (2003). The Toronto report card on housing and homelessness (execu- tive summary). Toronto: Author. Community Based Research Network of Ottawa (2002, April 18). Research and evaluation in a learning community. Symposium report, Ottawa. Datta, N., & Cairns, K. (2002). Families in transition: Case studies report. Unpub- lished manuscript, University of Calgary. Dordick, G. (2002). Recovering from homelessness: Determining the ?quality of sobriety? in a transitional housing program. Qualitative Sociology, 25(1), 7-32. Dunlap, K., & Fogel, S. (1998). A preliminary analysis of research on recovery from homelessness. Journal of Social Distress and the Homeless, 7(3), 175-188. Emerson-Davis Family Development Center (2000). Supportive residential ser- vices to reunite homeless mentally ill single parents with their children. Psychiatric Services, 51(11), 1433-1435. Fischer, R. (2000). Toward self-sufficiency: Evaluating a transitional housing pro- gram for homeless families. Policy Studies Journal, 28(2), 402-420. Fogel, S. (1997). Moving along: An exploratory study of homeless women and children using a transitional housing program. Journal of Sociology and Social Welfare, 24(3), 113-133. Gerstel, N., Bogard, C., McConnell, J. J., & Schwartz, M. (1996). The therapeutic incarceration of homeless families. Social Service Review, 70(4), 543-572. Griggs, S. & Johnson, G. (2002). Re-thinking exit points: A longitudinal research project designed to assess the durability of exits from transitional and crisis accommodation. Parity, March. Retrieved May 5, 2003, from http://www.infoxchange.net.au. Harvey, B. (1999). Models of resettlement for the homeless in the European Un- ion. In P. Kennett & A. March (Eds.) Homelessness: Exploring new terrain. Bristol: The Policy Press. Hawthorne, W., Fals-Stewart, W. & Lohr, J. (1994). A treatment outcome study of community-based residential care. Hospital and Community Psychiatry, 45(2), 152-155. Huffman, R. (1993). Transitional housing for veterans with a psychiatric diagno- sis. Journal of Psychosocial Nursing, 31(1), 11-14. Matulef, M., Crosse, S., & Dietz, S. (1995). National evaluation of the Supportive Housing Demonstration Program. Washington, D.C.: Department of Housing and Urban Development. Murray, R., Baier, M., North, C., Lato, M., & Eskew, C. (1997). One-year status of homeless mentally ill clients who completed a transitional residential pro- gram. Community Mental Health Journal, 33(1), 43-50. Nesselbuch, L. (1998). Transitional housing: A bridge to stability and self-sufficiency. Best practices in program design and delivery. San Francisco: HomeBase: The Center for Common Concerns. Novac, S., Brown, J., & Gallant, G. (1999). Women on the rough edge: A decade of change for long-term homeless women. Ottawa: Canada Mortgage and Hous- ing Corporation. Novac, S. & Quance, M. A. (1998). Back to community: An assessment of supportive housing in Toronto. Toronto: Mayor?s Homelessness Action Task Force. Phillips, M., DeChillo, N., Kronenfeld, D., & Middleton-Jeter, V. (1988). Homeless families: Services make a difference. Social Casework: The Journal of Contem- porary Social Work, 69(1), 48-53. Quantz, D., & Frankish, J. (2002). Homelessness research in the Greater Vancouver Regional District: An agency for moving forward. Vancouver: Institute of Health Promotion Research, University of British Columbia. Retrieved May 5, 2003, from http://www.ihpr.ubc.ca. Rice, J. (1987). Barton Street Project: Measuring the impact of a housing/life skills pro- ject on the functioning of multi-problem families. Ottawa: Canada Mortgage and Housing Corporation. Rog, D., Holupla, C. S., & McCombs-Thornton, K. (1995). Implementation of the Homeless Families Program: 1. Service models and preliminary outcomes. American Journal of Orthopsychiatry, 65(4), 502-513. Shlay, A. (1994). Running on empty: Monitoring the lives and circumstances of formerly homeless families with children. Journal of Social Distress and the Homeless, 3(2), 135-162. Shlay, A. (1993). Family self-sufficiency and housing. Housing Policy Debate, 4(3), 496-495. SPR Associates (1997). A place to go: An evaluation of the Next Step Program for sec- ond-stage housing in Canada. Ottawa: Canada Mortgage and Housing Corporation. Sprague, J. F. (1991a). More than housing: Lifeboats for women and children. Boston: Butterworth Architecture. Sprague, J. F. (1991b). More than shelter: A manual on transitional housing. Boston: The Women?s Institute for Housing and Economic Development. Stern, P. (1994). The Minnesota Transitional Housing Program: Structuring a study of its effectiveness. Minnesota: Minnesota Coalition for the Homeless. Twiss, P. (1993). Creating supported housing for homeless families with children: Im- plementing the Transitional Housing Program of the McKinney Act. Doctoral dissertation, University of Pittsburgh. United States General Accounting Office (1991). Homelessness: Transitional housing shows initial success, but long-term effects unknown (report to the Chairman). Washington, D. C.: Committee on Governmental Affairs, U.S. Senate. Washington, T. A. (2002). The homeless need more than just a pillow, they need a pillar: An evaluation of a transitional housing program. Families in Society: The Journal of Contemporary Human Services, 83(2), 183-188. Wekerle, G. (1988). Women?s housing projects in eight Canadian cities. Ottawa: Can- ada Mortgage and Housing Corporation. Wearne, H. & Johnson, G. (2002). Inner-city transitional accommodation. Retrieved April 15, 2003, from http://www.ashs.org.au/Main/info4.asp. Wiltshire, C. (1993). In search of a place to call home: A qualitative evaluation of a tran- sitional housing program for refugee families. Unpublished master?s thesis, Wilfrid Laurier University. Chapter 1.2 Shelters for the Homeless : Learning from Research ROCH HURTUBISE, PIERRE - O L IVIER BABIN, AND CAR OLYNE GRIMARD In spring 2006, in a regional round table discussion on homelessness in Quebec, we noticed that many practitioners were worried about the role of shelters and of their importance in addressing the problem. What is the mission of these shelters? Are they the solution or do they merely reproduce conditions one would find in an asylum? Can they offer a way out from the street? The questions are numerous. What can we learn from the research in social and human sciences? What is a shelter? Definitions of homelessness are numerous and subject to different interpretations (Gaetz, 2004; Roy & Hurtubise, 2007, 2004). The definition of a shelter is no less problematic. In its initial sense, a shelter is a place where one goes to avoid danger or a place where people who have no place else to go can gather. A brief survey of the terms in use sheds light on the diversity: shelter, hostel, emergency shel- ter. In French: refuge, maison d?h?bergement , auberge, h?bergement d?urgence. The shelter can be defined by the number of beds (from a few to several hundred) or by the nature of the services offered. In most cases, the services offered by the organisms are not limited to temporary hous- ing and food. Moreover, some organizations offer emergency housing services while refusing to be associated with shelters. For example, in many cities, one may find shelters that serve both abused women and homeless families, while in Quebec the network of shelters for women who are victims of domestic violence are largely independent of those for homeless people. It would probably be more appropriate to use the term emergency housing measures (Hopper, 2004). However, the larger shelters remain the best-known representatives of the services available to homeless people, as they are frequently mentioned in the media, par- ticularly in crisis situations, when they are often overcrowded. Who uses shelters? There seems to be a consensus on the necessity of distinguishing the population of these shelters and the population of people without a home. Not all homeless people are shelter users, and reducing the former to the latter often renders part of the homeless pop- ulation invisible. A history of shelters This review is in four sections: (1) a history of shelters; (2) a portrait of shelter users; (3) intervention practices associated with the primary mission of shelters; (4) criticisms of shelters. Traditionally associated with resources offered to beggars by religious communities, shelters have evolved over the years. The first shelters were created in cities where an influx of individuals seeking work in- creased the number of people without housing. This temporary housing developed in parallel to other solutions like shantytowns or camps. In- itially established as temporary services for the homeless population, they eventually became permanent (Dordick, 1996). The development of shelters at the end of the 19th century is related to developments in the economy (industrialization and urbanization) and the ethic of work as a way to distinguish the honest working man from the idler. Two waves of modernization of shelters occurred in the first half of the 20th century. The first improved the hygienic and sanita- ry conditions of the area by equipping the facilities with basic sanitary equipment; the second redefined the mission of the shelters by services to help shelter users reintegrate into society (Aranguiz, 2005; Aranguiz & Fecteau, 2000). In the postwar period, shelters were reaffirmed in their more traditional role of emergency housing, with the responsibilities of re-adaptation relegated to the realm of public services. Deinstitutionalization policies designed to maintain people with physical or mental health problems in the community changed the popu- lation using shelters and community services. For example, from 1984 to 1988, the number of people using shelters in New York increased from 5,000 to 8,000. Most suffered from addiction or mental health problems. Visions of the role of shelters differ: is it emergency housing in- tended to be used only in the short term, or support and protection of- fered in the long term (Gounis & Susser, 1990)? Towards the end of the 1980s, the creation by the city of Montreal of a referral centre for home- less people caused many problems: first, the challenge of setting up the centre; second, ensuring the safety of the users; third, opposition from the residents of the neighbourhood (Charest & Lamarre, 2000). Today, when new shelters are built or when old ones relocate, they often face opposition from local residents, merchants, property owners, and NIM- BYists (?not in my backyard?). In these debates, the people who have the strongest voice, who are able to block these projects, are generally own- ers of large private properties (Ranasinghe & Valverde, 2006). A Portrait of Shelter Users In the 1990s, critics of shelters became more harsh. Shelters were perceived as part of a system that tries to hide the homeless population. Because the presence of homeless people in public areas is seen as an annoyance and a menace, two strategies for fixing this problem emerged: designing these spaces so that they are less attractive to homeless people (architecture, streetscape) and controlling the behaviour of homeless people through litigation. This effort to rid cities of people deemed as ?undesirable? encouraged the development of shelters as a way of shiel- ding the population from homeless people (Johnsen et al., 2005). The Numbers It is important to distinguish the homeless population from the people using shelters. Too often, the number of nights in shelters is used as an indicator of the homeless population. Thus in certain cities, the absence of shelters would lead to an underestimation of the homeless population. Following a first generation of studies based on the opinions of ex- perts and witnesses, a second based on interviews with shelter users has emerged. A more complex array of investigative procedures has yielded more precise approximations (Firdion & Marpsat, 1998). Attempts are under way to standardize procedures and facilitate comparisons among cities ? for example, the creation of an information system on people and families without a home (HIFIS). In 2001, Statistics Canada estimated the number of people in shelters at 14,150 on census day, but this number must be interpreted with caution. Although the numbers vary, statistical studies help identify the converging trends of the users? characteristics. Three trends are clear: the increase in the homeless population, the diverse characteristics of home- less people, and the aggravation of problems linked to the situation. If certain censuses show stability in the number of users from 1990 to 2000 (United States Census Bureau, 2001), others show a considerable increase (Goldberg, 2005). The homeless population which uses shelters does not constitute a homogenous group (Hecht & Coyle, 2001; Novac et al., 2002). Generally, youths are less inclined to use public services and shel- ters for homeless people, and prefer life on the streets (Brooks et al., 2004; De Rosa et al., 1999). Use of shelters varies among different groups. In Canada, immi- grants and Aboriginal people are under-represented in shelters (Distasio et al., 2005; Fiedler et al., 2006), whereas in the United States, Blacks and Hispanics are over-represented (Gondolf et al., 1988). The number of men using shelters who are involved in the judicial system is four times greater than in the general population (Tolomiczenko & Goering, 2001). People in shelters may express more satisfaction about their envi- ronment than those on the street and do not associate the shelters with a loss of freedom (La Gory et al., 1990). However, youths may see shelters as too restrictive, and often distrust the staff and associated social work- ers. The community network takes over for other services (Levac & La- belle, 2007; Poirier & Chanteau, 2007). Profiles of users enable us to determine different types of homeless- ness: chronic, cyclical, or temporary (Acorn, 1993). Certain groups (the elderly, or those suffering from mental health problems, addictions, or physical problems) stay for prolonged and repetitive periods. A few sit- uations seem to be particularly problematic: users who have exhausted their personal and family resources and who are also rejected by the public system often express aggressive behaviour towards aid workers and other users. Racial origin seems to be strongly associated with the length of the stay in the shelters; Caucasian people stay less than half as long as Black people (Culhane & Kuhn, 1998). However, in winter, stays are usually longer because of harsher weather. Simard (2005) estimated the average stay in a large shelter to be 355 days. Most beds (60 percent) are used by individuals staying more than three months, and 30 percent are used by those who stay more than a year. Users of shelters are not necessarily unemployed; some may have precarious or low-wage jobs. Research has established profiles of usage according to people?s needs: transition towards stable housing, rest, emergency, usage in addition to day centre use (Grella, 1994). Epidemiological Profile In 2006, the ?tent crisis? in Paris raised the question of homeless people refusing to use shelters. The initiative of a humanitarian group consisting in distributing tents during the winter season to improve the living conditions of homeless people provoked a social crisis. Homeless people spoke out publicly about life in shelters and explained that the life inside the tent presented a better alternative (de Fleurieu & Cam- baud, 2006). Elsewhere, homeless people also refuse to use the housing resources available because they are deemed constraining and threaten- ing (Hopper, 2003). The health status of people using shelters presents a serious challenge (Carri?re et al., 2003, Hurtubise et al, 2008). The use of shelters can even cause health problems through sleep deprivation, personal hygiene diffi- culties, or limited space for storing personal goods (Power et al., 1999). Users are often hesitant to use normal health services, treatment and prevention practices, and suffer health problems as a result (Frankish et al., 2005; Harris, 1994). The mortality rate varies from two times to eight times higher than that of the general population (Barrow et al., 1999; Hwang, 2000). Despite a decrease in schizophrenia cases (Geddes et al., 1994), some studies suggest that between 40 and 60 percent of the homeless population suffer some form of mental health problem, such as anxiety, depression, or suicidal tendencies (Fournier & Mercier, 1996; Poirier et al., 2000). Drug and alcohol problems are common. Shelters may contri- bute to the spread of infectious diseases such as tuberculosis or parasites such as lice (Marks et al., 2000). The question of health calls for a better understanding of the strategies used by people who are homeless (Wadd et al., 2006). Users and Appropriations Homeless people have difficulty accessing resources to take care of themselves (Boydell et al., 2000; Laberge, 2000). Often, they also end up adding to their health problem by waiting too long before seeking help (Desai & Rosenheck, 2005). They often use the emergency services of hospitals (Kushel et al., 2001; Marks et al., 2000; Stein et al., 2000; Thi- baudeau, 2000). Despite their obvious needs, homeless people are poorly served when it comes to health services, either prevention or interven- tion (Roy et al., 2006; Webb, 1998). Firdion and Marpsat (1998) point out that the differences between short- and long-term shelter users are not clearly defined. A more dynamic ap- proach that focuses on users? characteristics is helpful. Four groups can be distinguished: (1) those who make maximum use of resources during medium- and long-term re-integration into society; (2) those who find their own solutions to problems, without the use of resources for the homeless; (3) those who make ad hoc use of emergency shelter resources; (4) those in precarious housing situations (cars, trailers, squatting). A person who has used up all his or her personal, family, and community resources, may turn to a shelter as a last resort (Poole & Zugazaga, 2003). A focus on understanding the different solutions used by the homeless to compensate for a lack of housing allows us to better study the survival methods of those involved (Elias & Inui, 1993). According to Hopper (2003), understanding the history of the people in homeless situations enables us to propose more complete intervention models. Life in shelters is far from ideal, and living conditions are often de- scribed as similar to those in traditional asylums (Simard, 2000). The atti- tudes of workers and the organizational structure of shelters may create a context favourable to violent behaviour among users (Liebow, 1993). Beyond Emergency Sheltering, Inte rvention Practices Dordick (1996) proposes a description of the ?social world? in shel- ters. For example, sexual practices, seldom mentioned in scientific litera- ture, are an important preoccupation, even in areas that offer little or no privacy. Couples may form in shelters. Rituals of engagement have been observed among these couples, which imply support and comfort in shelters as well as outside them. Functions and Approaches Research on programs for homeless people that involve shelters can be divided up into four categories: (1) functions and approaches; (2) shelter- ing and housing as a stepping-stone to social re-integration; (3) shelters as a place for intervention; and (4) program evaluations. Studies of how shelters are organized use two perspectives: (1) the de- sired approach of professional workers and volunteers, and (2) the rules and guidelines that regulate life in shelters. Shelter staff may develop an understanding of the life conditions of the homeless and of the state of mind and characteristics displayed by shelter users. Flexibility, understanding, the ability to listen and to adapt to a person?s needs are all qualities that are valued in practitioners who have to constantly adapt to very diversified needs. Most shelters set rules and regulations that outline acceptable and unacceptable behaviour for both shelter users and practitioners (Neale et al., 1997; Roy et al., 2000). For example, permission to enter the shelter may depend on the person?s mental state (intoxication, aggressiveness, under the influence of drugs), personal characteristics (gender, age, cul- tural background), or history with the shelter (limited number of visits). Once inside the shelter, there are rules governing personal hygiene (showering, changing clothes), curfew and wake-up times, respecting others (noise, aggressive behaviour, violence), and participation in group chores (food preparation, dishwashing, chores). Repeated failure to re- spect the rules will result in a penalty such as temporary or permanent exclusion, extra chores, or reduced access to services. Sheltering and Housing as a Stepping Stone for Social Insertion Studies of intervention practices sometimes take the form of typolo- gies of shelter operations (Mosher-Ashley & Henrikson, 1997; Pel?ge, 2004). Shelters gather information from users and can target problems to refer residents to appropriate resources or services. Some shelters pro- mote job readiness through in-house training centres, social enterprises, or by employment groups. Others focus on health needs and orient users towards services that correspond best to their needs. The challenge of accessibility is a central point; there are numerous examples of cases where needs were clearly defined, but accessibility was limited by cul- tural, organizational, or administrative barriers (Roy et al., 2006). Servic- es are not available in all shelters and the complex problems of the homeless are not always taken into consideration in those offered (Berg & Hopwood, 1991). For example, many of the homeless suffering from mental health problems use shelters as a substitute to permanent and more appropriate housing (Hopper et al., 1997). Housing is a right, a social norm, a behaviour stabilizer, and a status symbol (Dorvil & Morin, 2001; Fuller-Thomson et al., 2000; Laberge & Roy, 2001). A home is a social anchor point for individuals. This means distinguishing between shelter and housing: the first implies a temporary way of life that offers help that may include some form of rehabilitation, the second is a stable way of life that in no way implies any social or the- rapeutic needs (Dorvil et al., 2002). Numerous projects in shelters have tried to facilitate housing for the homeless. A stay in a shelter constitutes an ideal occasion to work on a per- son?s ability to manage their own home. Shelters allow users a tempo- rary experience in a stable and safe environment (Peled et al., 2005). From this point of view, the role of shelters is to favour the transition towards stable housing, a move that implies not only finding a place to live but also building a solid foundation and a social network in the community (Friedman, 1994). Follow-up after leaving the shelter is an essential condition to the success of reintegration, and is a lengthy process. By all accounts, residential stability is very fragile during the first year and only in the second year do most people settle down (Dun- lap & Fogel, 1998). During the 1990s, advocates for the homeless focused on the right to housing as an alternative to solutions that relied essentially on a quick response to a crisis and poverty situation. The right to housing took pre- cedence over the simple right to shelter in the platforms of many human rights groups (Bresson, 1997; Hopper, 1998). Subsequently, the Council of Europe in its final declaration to the Congress of Local and Regional Authorities (1994, p. 183) noted: ?The right of all human beings to de- cent, affordable housing of a certain standing, adapted to essential needs is a fundamental right recognized by, among others, the Universal Dec- laration of Human Rights and where implementation is an obligation for all of society without exception or discrimination? (translated from the French). Shelters as Places for Intervention Research on housing rights includes comparative analyses of the costs of the services used by the homeless (shelters, public services) and the costs associated with long-term stability in a dwelling. For example, investments in subsidized housing for the homeless would decrease the costs of other services for the homeless. The savings generated would largely cover the financing of subsidized housing. Moreover, improve- ments in the quality of life of homeless people suffering from mental health problems can translate into a reduction in shelter use, hospital visits, and the number of people incarcerated (Culhane et al., 2002). The impacts of this type of initiative are numerous: better quality of life, in- creased self-esteem, development of self-affirmation skills and rights advocacy, developing a network, rights of citizens and social participa- tion (Metraux et al., 2003; Novac & Brown, 2004; Roy et al., 2003). Even if it is difficult to determine just how efficient they are, it is obvious that the interventions taking place in shelters often succeed in reaching out to a population considered marginal and fearful of public services (Levinson, 2004). Research tends to focus on the resources and the inter- vention models that target specific sub-groups: women, youths, the el- derly, and individuals with mental health problems. There is less re- search focused on the interventions with adult males. Mental health tops the list of problems. Shelters offer basic support, but it is difficult to do so for those suffering from mental health prob- lems. Grella (1994) suggests that shelters should offer options related to helping the homeless population suffering from mental health problems. A follow-up after the initial intervention (Hall, 1991) and long-term ser- vices are useful when dealing with homeless people suffering from men- tal health problems. Applebaum (1992), Dattalo (1991), and Hall (1991) suggest removing barriers to services, coordinating services, emphasiz- ing patient participation, modifying rules on the protection of informa- tion, lobbying for social and psychiatric services, raising shelter work- ers?s awareness of mental issues, and improving training. More mental health services are offered inside shelters than physi- cal health services (Mosher-Ashley & Henrikson, 1997). This fact raises questions about the responsibilities of community organizations relative to public services. Some experts fear the development of a parallel health system for homeless people. The intervention practices developed in shelters must be analysed within the context of the transformation of health and social services (Racine, 1993). What are the best places and the most strategic moments to inter- vene and avoid a relapse? The post-shelter period is considered particu- larly crucial and follow-up to ensure the continuity of the process of emerging from homelessness is essential. Interventions through a net- work of community services can prevent the reoccurrence of homeless- ness (Susser et al., 1997). Numerous programs have focused on reducing homelessness through a more intensive approach. Min, Wong, and Rothbard (2004) looked at the Access to Community Care and Effective Services and Support (ACCESS) program in the United States from 1993 to 1998. The program adopted a treatment model in the community combined with individual management of each person?s case. The objective of the pro- gram was to help homeless people suffering from mental health prob- lems emerge from poverty. The results suggest that managing each per- son?s case individually can reduce the risk of chronic homelessness in people suffering from mental health problems. Health practices usually consist of guiding the person towards available resources. Some nurses have developed a practice that involves regular visits to shelters and follow-up with shelter users. Strategies fo- cus on the resolution of problems, empowerment, work with network personnel, and sharing resources (Denoncourt & Bouchard, 2006; Di Marco, 2000; Thibeaudeau, 2000). Some studies have evaluated the effec- tiveness of health services for homeless people. For example, a shelter- based convalescence facility can help workers supply health service needs better adapted to individual conditions, ensure a more complete treatment of medical and mental health problems, favour continuity of treatment, reduce drug dependency, and help individuals with social reintegration (Podymow et al., 2006). Many interventions target sub-populations, particularly women, youth, and the elderly. Work with homeless women may call for a new approach inspired by the feminist movement that focuses on offering safe living conditions, valuing the autonomy of women, and establishing a trusting relationship (Goldberg, 1999; Gondolf, 1998; S?vigny & Racine, 2002). Most youth crisis centres follow similar goals: respond to basic needs (food, clothing, showers, a place to sleep, entertainment) and work to end marginality by helping youths develop everyday skills, find a place to live, manage a budget, use available resources, find employ- ment, and, in certain cases, reconcile with their families. Approaches that combine education and behaviour change through a coping and stress management strategy facilitate the resolution of the crisis (Dalton & Pa- kenham, 2002; Teare & Peterson, 1994). A few studies about services for elderly homeless people indicate a significant increase in this population. Physical health problems are sig- nificant and the barriers to services are numerous (Abdul-Hamid, 1997). In these cases, homelessness is often associated with a loss of autonomy and a decrease in support network; turning to a shelter may increase the effect of these losses in elderly people whose cognitive abilities are de- clining (Elias & Inui, 1993). Researchers have also documented original initiatives ? the addition of judicial services in shelters (Binder, 2001), the introduction of occupa- tional therapy programs (Herzberg & Finlayson, 2001) or the use of eth- nographic approaches in clinical work (Grisgby, 1992). These studies tend to be descriptive and do not identify the most effective practices. Program Evaluations The literature includes evaluations of the impact and the efficiency of the services as well as the role of shelters in the fight against homelessness. Numerous studies examine the contribution of a stay in a shelter to allowing individuals to escape homelessness and find permanent hous- ing. There are many contradictory views. Short-term improvements may be followed by deterioration (the change is often temporary); in other situations, the transformations seem more permanent, especially when there is effective follow-up (Glisson et al., 2001; Peled et al., 2005; Pollio et al., 2006). In certain situations, specific services in shelters are evaluated, for example, a decrease in behavioural problems in children of women par- ticipating in a conflict management program in centres for abused wom- en (McDonald et al., 2006). The role of shelters in ending homelessness can be looked at in two ways: (1) shelters as partners in intersectorial alliances networks in the fight against homelessness; (2) shelters as part of the continuum of care. Shelter administrators use diverse strategies to maintain services, such as tightening accessibility rules, or bridging with other resources (Goodfel- low, 1999). With such a diverse clientele expressing complex problems, collaborations with external resources and the diversification of practices becomes a necessity. Developing partnerships involves relationship building, clarifying expectations, identifying needs, sharing expertise, and evaluating the collaboration (Snyder & Weyer, 2002). Shelters can be the first step into a system of services, a place from which it is possible to evaluate a per- son?s needs and begin implementing interventions. Coordination by a case manager can ensure follow-up and the continuation of the interven- tions (Feins & Fosburg, 1999). Effectiveness depends largely on the inte- gration of many resources around the needs of the individual: preven- tion, outreach, emergency shelter, transitional housing, supportive housing, and affordable housing (Burt, 2004; Carter, 2005). The continui- ty of services seems promising in homelessness, but this approach presents ethical challenges that should be scrutinized in future research. Critical Analysis of Shelters A Total Institution? A body of research questions the role and the place of shelters as solu- tions to homelessness. These studies look at the homeless problem from a different angle and reveal some less than positive aspects. Some authors favour Goffman?s approach for analyzing homelessness (Pichon, 2002). From this point of view, shelters are viewed as total insti- tutions, consuming all the time of their users and depriving them of freedom. The rules established to control the physical and social envi- ronment of shelters shape the users and reinforce their marginal identity. The culture of total institutions tends to alienate and depersonalize users, whose lives are defined by their belonging to the shelter. For users, this translates into a loss of autonomy and the feeling of domination and en- closure. This perspective allows us to understand conflicting roles and allegiances that are often viewed as irreconcilable (Stark, 1994). The rules of some shelters show how encompassing shelter life can become. Underground practices may add a ?black market economy? of sorts, such as food re-selling networks, protective services, control of privileges, and odd jobs. Three other factors affect life in shelters: (1) a majority of time is spent on organizing ?living? in shelters, which leaves little time for other things; (2) personal networks and friendships may be fragile and short-term; (3) obligations towards other people must be res- pected, and leaving the shelter may be seen as abandoning these obliga- tions (Dordick, 1996). For Marcus (2003), this analysis neglects the role of collectivity in the lives of homeless people. The idea that shelters isolate users obscures the fact that for homeless people, shelters are one resource among many, and their strategy for survival and escape from homelessness draws on public, community, family, and personal resources. Shelter users are not completely defined by a sub-culture; they share values, beliefs, and norms with the general population. Shelterization: Confinement in a Marginal Area Marginal affinity, the proximity between shelter users and professionals, denotes sharing of a common surrounding and the development of a sense of belonging to a marginal environment. This proximity is appar- ent in the participation of homeless people in daily chores, the fluidity of the roles of interveners, and the absence of standards. In fact, services intended to aid in recovering from homelessness actually favour the re- production and maintaining of shelter life. Personal failures encountered by users during their efforts to find stable housing sometimes reinforce their sense of belonging to shelters, the place that accepts them for who they are and doesn?t judge them (Gounis & Susser, 1990). The idea of shelterization has been discussed by Novac, Brown, and Bourbonnais (1996) and Kozol (1988). A certain social pathology engulfs people in lethargic situations, so that they become incapable of taking responsibility for their lives, neglect personal hygiene, and lose interest in escaping their situation. For users, this situation is defined by a loss of autonomy, a lack of self-respect, and a loss of responsibilities. Confining rules, the difficulty of being able to care for oneself, and personal prob- lems can create a larger dependency on the services and an enclosure in homelessness (Elias & Inui, 1993). Social Regulations and the Role of Policies Shelterization also emphasizes the social processes of enclosure in homelessness, similar to the concentration of poor populations in ghet- tos. The abuse of shelters is not just a personal problem. Users become psychologically and economically tied to social assistance programs, and adapt by developing survival mechanisms that keep them homeless. Shelterization creates a sub-culture based on a common language and the assimilation of shared ideas and values. Furthermore, tolerating de- linquent behaviours may lead to a redefinition of what is normal beha- viour. Regardless of the dangers and the depersonalization, users are reluctant to leave the shelters (Grunberg & Eagle, 1990). Shelters are not neutral sites, they represent the borders of marginality where street rules apply (Zeneidi-Henry, 2002). Many authors question the role of the state in perpetuating homelessness. The reduction of ser- vices has resulted in a housing crisis that forces certain people to use emergency shelters (Layton, 2000). At the same time, the medicalization of problems or assertions that certain situations represent individual failures mask the real causes of people?s difficulties (Damon, 2002; Mar- cus, 2003). Studies focus on two areas: housing policies and urban planning and security policies. Some authors believe that policies focusing on access to housing should be reinforced and could help solve homeless- ness (Roman & Berg, 2006). Here we see the debate between targeting the clientele as a necessary condition for the implementation of efficient solutions, and the adaptation of existing general services by favouring accessibility and support for people (Dattalo, 1991; Fontaine, 2000). The shelter plays an intermediary role between homeless people and the community; it becomes, for some, a type of affordable housing. The shelter system can be seen as an official willingness to neutralize a problem. In fact, the location of a shelter, its structure and operational modalities influence the type of reintegration that homeless people can expect in a community. Offering many services within a shelter contri- butes to people?s isolation, because there is no incentive for them to use outside services or to familiarize themselves with the location of re- sources and services (Hartnett & Harding, 2005). There are two types of shelter. Some offer little comfort and few fi- nancial resources, and refer users to other services. Others offer more comfort and better resources, providing a personal approach to people who have the potential to make the transition to permanent housing. The emergency shelter network is therefore hierarchal. The sheltered popula- tion is not an arbitrarily formed group; it is the result of a selection process. Homeless people who can convince officials that they have the potential to benefit from services often gain access to better quality cen- tres (Soulie, 1997). The hierarchy represents a social control process that, through prioritization and targeting certain clientele, allows service pro- viders to distinguish ?good? homeless people from ?bad? homeless people; the ?good? group may qualify for intensive interventions, be- cause their problems are often less intractable (Hurtubise, 2000). Research Challenges Shelters vary widely. Their history and their development reveal the var- ious ways that individuals have protested to fight against homelessness. For human and social science research, shelters are important partners. Shelter workers are key sources of information for the evolution of the face of homelessness and help researchers reflect on and analyse their approach. Some research suggests that shelters are the best way to handle the homeless problem. However, most studies indicate that any solution to homelessness must include many participants and involve numerous sectors: community organizations, city governments, health and social services institutions, law enforcement agencies, and private and com- munity practitioners. Research should continue to describe the various experiences of shelters and document their transformation. Studies are also needed to analyse programs and practices and identify the most ef- fective interventions. Furthermore, critical analysis must continue to question ideas that are taken for granted. Since the homeless problem is so complex, solutions must be adapted to the diversity of the contexts to which they are applied. Roch Hurtubise and Pierre - Olivier Babin References are with the D?partement de se r- vice social at the Universit? de Sherbrooke. Carolyne Grimard is at the D?pa r- tement de sociologie at the Universit? de Qu?bec ? Montr?al (UQ?M). They are associated with CRI ? Collectif de recherche sur l?itin?rance, la pauvret? et l?exclusion sociale (www.cri.uqam) . Abdul-Hamid, W. (1997). The elderly homeless men in Bloomsbury hostels: Their needs for services. International Journal of Geriatric Psychiatry, 12(7), 724- 727. Acorn, S. (1993). Emergency shelters in Vancouver, Canada. Journal of Community Health, 18(5), 283-291. Applebaum, P. S. (1992). Legal aspects of clinical care for severely ill, homeless persons. Bulletin of the American Academy of Psychiatry and the Law, 20(4), 455-473. Aranguiz, M. (2005). Les pouvoirs urbains et la prise en charge des vagabonds ? Montr?al: Le cas du refuge municipal Meurling (1914 -1929). D?viance et S o- ci?t?, 29(2), 181-199. Aranguiz, M., & Fecteau, J.-M. (2000). L??cole de la pr?carit?: V agabonds et er- rants ? Montr?al eu tourna nt du si?cle. In D. Laberge (Ed.) L?errance urbaine (pp. 11-27). Sainte-Foy, QC: ?ditions Mu ltiMondes. Barrow, S. M., Herman, D. B., Cordova, P., & Struening, E. L. (1999). Mortality among homeless shelter residents in New York City. American Journal of Public Health, 89(4), 529-534. Berg, I. K., & Hopwood, L. (1991). Doing with very little: Treatment of homeless substance abusers. Journal of Independent Social Work, 5(3-4), 109-119. Binder, S. R. (2001). The Homeless Court Program: Taking the Court to the streets. Federal Probation, 65(1), 14-18. Boydell, K., Goering, P., & Morrell-Bellai, T. L. (2000). Narratives of identity: Re- presentation of self in people who are homeless. Qualitative Health Research, 10(1), 26-38. Bresson, M. (1997). Les S.D.F. et le nouveau contrat social. Paris: L?Harmattan. Brooks, R. A., Milburn, N. G., Rotheram, M. J., & Witkin, A. (2004). The System- of-care for homeless youth: Perceptions of service providers. Evaluation and Program Planning, 27(4), 443-451. Carri?re, S ., Hurtubise, R., & Lauzon, ? . (2003). Les d?fis de l?intervention en CLSC aupr?s des personnes itin?rantes: S ynth?se d?une journ?e d??changes tenue le 15 novembre 200 2 . Montreal: CLSC des Faubourgs, CRI. Carter, T. (2005). Delivery models: National consultation on developing a Canadian housing. Framework prepared for expert roundtable #5. Hull, QC: National Secretariat on Homelessness [ Secr?tariat national des sans abri]. Charest, R., & Lamarre, M.-J. (2000). Structures et repr?sentations des servi ces: Le cas de dernier recours Montr?al. In D. Laberge (Ed.) L?errance urbaine (pp. 377-389). Sainte-Foy, QC: ?ditions Mu ltiMondes. Conseil de l?Europe. (1994). Le logement social, les sans-abri et les mal-log?s en Euro- pe. Strasbourg, France: Author. Culhane, D. P., & Kuhn, R. (1998). Patterns and determinants of public shelter utilization among homeless adults in New York City and Philadelphia. Journal of Policy Analysis and Management, 17(1), 23-43. Culhane, D. P., Metraux, S., & Hadley, T. (2002). The impact of supportive hous- ing for homeless people with severe mental illness on the utilization of the public health, corrections, and emergency shelter systems: The New York- New York initiative. Housing Policy Debate, 13(1), 107-163. Dalton, M. M., & Pakenham, K. I. (2002). Adjustment of homeless adolescents to a crisis shelter: Application of a stress and coping model. Journal of Youth and Adolescence, 31(1), 79-89. Damon, J. (2002). La question SDF: Critique d?une action publique. Paris: Presses Universitaires de France. Dattalo, P. (1991). Moving beyond emergency shelter: Who should fund low- income housing?. Social Work, 36(4), 297-301. de Fleurieu, A., & Chambaud, L. (2006). L?H?bergement des personnes sans abri ? Paris et en Ile de France. Paris: Minist?re de la coh?sion sociale et de la p arit?. De Rosa, C. J., Montgomery, S. B., Kipke, M. D., Iverson, E., Ma, J. L., & Unger, J. B. (1999). Service utilization among homeless and runaway youth in Los Angeles, California: Rates and reasons. Journal of Adolescent Health, 24(3), 190-200. Denoncourt, H., Bouchard, M.-C., & Keays, N. (2007). Cent fois sur le m?tier... Le renouvellement de la pratique infirmi?re aupr?s des per sonnes itin?rantes. In S. Roy & R. Hurtubise (Eds.) L?Itin?rance en question . Sainte-Foy, QC: PUQ. Desai, M. M., & Rosenheck, R. A. (2005). Unmet need for medical care among homeless adults with serious mental illness. General Hospital Psychiatry, 27(6), 418-425. DiMarco, M. A. (2000). Faculty practice at a homeless shelter for women and children. Holistic Nursing Practice, 14(2), 29-37. Distasio, J., Sylvestre, G., & Mulligan, S. (2005). Home is where the heart is and right now that is nowhere: An examination of hidden homelessness among aboriginal peoples in prairie cities. Winnipeg, MB: University of Winnipeg, Institute of Urban Studies. Dordick, G. (1996). More than refuge: The social world of a homeless shelter. Journal of Contemporary Ethnography, 24(4), 373-404. Dorvil, H., & Morin, P. (Eds.). (2001). Dossier: Le logement social et l?h?bergement. Nouvelles Pratiques Sociales, 14(2), 20-27. Dorvil, H., Morin, P., Beaulieu, A., & Robert, D. (2002). Le logement comme fac- teur d?int?gration sociale pour les perso nnes class?es malades mentales. D?viance et Soci?t?, 26(4), 497 -515. Dunlap, K. M., & Sondra J. F. (1998). A preliminary analysis of research on recov- ery from homelessness. Journal of Social Distress and the Homeless, 7(3), 175- 188. Elias, C., & Inui, T. S. (1993). When a house is not a home: Exploring the meaning of shelter among chronically homeless older men. The Gerontologist, 33(3), 396-402. Feins, J. D., & Fosburg, L. B. (1999). Emergency shelter and services: Opening the front door to the continuum of care. In L. B. Fosburg & D. L. Dennis (Eds.) Practical Lessons: The 1998 National Symposium on Homelessness Research. Delmar, NY: National Resource Center on Homelessness and Mental Ill- ness. Fiedler, R., Schurman, N., & Hyndman, J. (2006). Hidden homelessness: An indi- cator-based approach for examining the geographies of recent immigrants at risk of homelessness in Greater Vancouver. Cities, 23(3), 205-216. Firdion, J. M., & Marpsat, M. (1998). Caract?ristiques sociod?mographiques et modes de vie. Les personnes sans domicile ? Paris (pp. 3-15). Paris: CREDES. Fontaine, M. (2000). Les besoins et les services: Les paradoxes. In D. Laberge (Ed.) L?errance urbaine (pp. 275-290). Sainte-Foy, QC: ?ditions Mult iMondes. Fournier, L., & Mercier, C. (Eds.). (1996). Sans domicile fixe: Au-del? du st?r?o type. Montreal: ?ditions du M?ridien. Frankish, J. C., Hwang, S. W., & Quantz, D. (2005). Homelessness and health in Canada: Research lessons and priorities. Revue Canadienne de Sant? Publ i- que/Canadian Journal of Public Health, 96(Suppl. 2), 23-29. Friedman, B. D. (1994). No place like home: A study of two homeless shelters. Journal of Social Distress & the Homeless , 3(4), 321-339. Fuller-Thomson, E., Hulchanski, J. D., & Hwang, S. (2000). The housing/health relationship: What do we know?. Review on Environmental Health, 15(1-2), 109-133. Gaetz, S. (2004). Understanding research on homelessness in Toronto: A literature re- view. Toronto: York University and Wellesley Central Health Foundation. Geddes, J., Newton, R., Young, G., Bailey, S., Freeman, C., & Priest, R. (1994). Comparison of prevalence of schizophrenia among residents of hostels for homeless people in 1966 and 1992. British Medical Journal (International ed.), 308(6932), 816-819. Glisson, G. M., Thyer, B. A., & Fischer, R. L. (2001). Serving the homeless: Eva- luating the effectiveness of homeless shelter services. Journal of Sociology and Social Welfare, 28(4), 89-97. Goldberg, J. E. (1999). A short term approach to intervention with homeless mothers: A role for clinicians in homeless shelters. Families in Society, 80(2), 161-168. Goldberg, M. (2005). On our streets and in our shelters: Results of the 200 5 Greater Vancouver homeless count. Vancouver, B.C.: Social Planning and Research Council of BC. Gondolf, E. W., Fisher, E., & McFerron, J. R. (1988). Racial differences among shelter residents: A comparison of Anglo, Black, and Hispanic battered women. Journal of Family Violence, 3(1), 39-51. Goodfellow, M. (1999). Rural homeless shelters: A comparative analysis. Journal of Social Distress and the Homeless, 8(1), 21-35. Gounis, K., & Susser, E. (1990). Shelterization and its implications for mental health services. In N. Cohen (Ed.) Psychiatry takes to the streets: Outreach and crisis intervention for the mentally ill (pp. 231-255). New York: The Guilford Press. Grella, C. (1994). Contrasting a shelter and day center for homeless mentally ill women: Four patterns of service use. Community Mental Health Journal, 30(1), 3-16. Grigsby, K. R. (1992). Mental health consultation at a youth shelter: An ethno- graphic approach. Child and Youth Care Forum , 21(4), 247-261. Grunberg, J., & Eagle, P. (1990). Shelterization: How the homeless adapt to shel- ter living. Hospital and Community Psychiatry, 41(5), 521-525. Hall, L. K. (1991). Homelessness: A model for mental health intervention. Admin- istration and Policy in Mental Health, 18(6), 451-454. Harris, S. N., Mowbray, C. T., & Solarz, A. (1994). Physical health, mental health, and substance abuse problems of shelter users. Health & Social Work, 19(1), 37-45. Hartnett, H. P., & Harding, S. (2005). Geography and shelter: Implications for community practice with people experiencing homelessness. Journal of Pro- gressive Human Services, 16(2), 25-46. Hecht, L., & Coyle, B. (2001). Elderly homeless: A comparison of older and younger adult emergency shelter seekers in Bakersfield, California. Ameri- can Behavioral Scientist, 45(1), 66-79. Herzberg, G., & Finlayson, M. (2001). Development of occupational therapy in a homeless shelter. Occupational Therapy in Health Care, 13(3-4), 133-149. Hopper, K. (1998). Du droit ? l?h?bergement au droit au log ement: Quinze ans de mobilisation en faveur des sans-domicile aux ?tats -Unis. Soci?t?s Contemp o- raines, 30, 67-94. Hopper, K. (2003). Reckoning with homelessness. Ithaca, NY & London: Cornell University Press. Hopper, K. (2004). Shelters. In D. Levinson (Ed.) Encyclopaedia of homelessness (pp. 498-503). Thousand Oaks, California: SAGE Publications. Hopper, K., Jost, J., Hay, T., & Welber, S. (1997). Homelessness, severe mental illness, and the institutional circuit. Psychiatric Services, 48(5), 659-665. Hurtubise, R. (2000). Aide alimentaire et pauvret?: V ers de nouvelles formes de priorisation des client?les et des interventions. In D. Laberge (Ed.) L?errance urbaine (pp. 349-359). Sainte-Foy, QC: ?ditions Mu tliMondes. Hurtubise, R., Roy, S. (2008). ? Le r?cit de s?ropositivit? comme production de sens de sa propre vuln?rabilit?. VIH/sida et vie ? la rue?, dans Penser la vuln?rabilit?. Visages de la fragilisation du social, Ch?tel V. & Roy, S. (Eds.) : PUQ, p. 165-180. Hurtubise, R., Roy, S. et Bellot. C. (2003). ?Youth Homelessness: The Street and Work-From Exclusion to Integration?. in Youth and Work in the Post - IndustrialCity of North America and Europe, Ed. Laurence Roulleau- Berger, Boston: Brill Leiden, p. 395-407. Hurtubise, R., Roy, S., Rozier, M. & Morin, D. (2007). Agir sur sa sant? en situ a- tion d?itin?rance. In S. Roy & R. Hurtubise (Eds.) L?itin?rance en question . Sainte-Foy, QC: PUQ. Hwang, S. W. (2000). Mortality among men using homeless shelters in Toronto, Ontario. Journal of the American Medical Association, 283(16), 2152-2157. Johnsen, S., Cloke, P., & May, J. (2005). Day centers for homeless people: Spaces of care or fear?. Social and Cultural Geography, 6(6), 787-810. Kozol, J. (1988). Rachel and her children: Homeless families in America. New York: Crown Publishers, Inc. Kushel, M. B., Vittinghoff, E., & Haas, J. S. (2001). Factors associated with the health care utilization of homeless persons. Journal of the American Medical Association, 285(2), 200-206. La Gory, M., Fitzpatrick, K., & Ritchey, F. (1990). Homeless persons: Differences between those living on the street and in shelters. Sociology & Social R e- search, 74(3), 162-167. Laberge, D. (Ed.). (2000). Urban wandering . Sainte-Foy, QC: Les ?ditions Mult i- Mondes. Laberge, D., & Roy, S. (2003). Continuit? identitaire et survie?. In V. Ch?tel & M. - H. Soulet (Eds.) Agir en situation de vuln?rabilit? (pp. 143-156). Saint-Nicolas, QC: Presses de l?Universit? Laval . Layton, J. (2000). Homelessness: The making and unmaking of Canada?s housing crisis. Toronto: Penguin. Levac, C., & Labelle, F. (2007). La rue, un chemin trac? d?avance?. Montreal: Refuge des jeunes de Montr?al. Levinson, D. E. (2004). Encyclopaedia of homelessness. Thousand Oaks, California: SAGE Publications. Liebow, E. (1993). Tell them who I am: The lives of homeless women. New York: Free Press. Marcus, A. (2003). Shelterization revisited: Some methodological dangers of insti- tutional studies of the homeless. Human Organization, 62(2), 134-142. Marks, S. M., Taylor, Z., Burrows, N. R., Qayad, M. G., & Miller, B. (2000). Hospi- talization of homeless persons with tuberculosis in the United States. Amer- ican Journal of Public Health, 90(3), 435-438. McDonald, R., Jouriles, E. N., & Skopp, N. A. (2006). Reducing conduct problems among children brought to women?s shelters: Intervention effects 24 months following termination of services. Journal of Family Psychology, 20(1), 127-136. Metraux, S., Marcus, S., & Culhane, D. (2003). The New York-New York housing initiative and use of public shelters by persons with severe mental illness. Psychiatric Services, 54(1), 67-71. Min, S., Wong, Y., & Rothbard, A. (2004). Outcomes of shelter use among home- less persons with serious mental illness. Psychiatric Services, 55(3), 284-289. Mosher-Ashley, P., & Henrikson, N. (1997). Shelter-based services in Massachu- setts for homeless adult individuals. Social Work in Health Care, 26(2), 15-32. Neale, J., Burrows, R., Pleace, N., & Quilgars, D. (1997). Hostels. In R. Burrows, N. Pleace, & D. Quilgars (Eds.) Homelessness & social policy (pp. 203-215). London: Routledge. Novac, S., & J. Brown. (2004). Transitional housing: Objectives, indicators of success, and outcomes. Ottawa: SCHL/CMHC. Novac, S., Brown, J., & Bourbonnais, C. (1996). Elles ont besoin de toits: Analyse documentaire sur les femmes sans-abri. Ottawa: Soci?t ? canadienne d?hypoth?ques et de logement. Novac, S., Serge, L., Eberle, M., & Brown, J. (2002). On her own: Young w omen and homelessness in Canada. Ottawa: Status of Women Canada. Peled, E., Spiro, S., & Dekel, R. (2005). My home is not my castle: Follow-up of residents of shelters for homeless youth. Child and Adolescent Social Work Journal, 22(3-4), 257-279. Pel?ge, P. (2004 ). H?bergement et r?insertion sociale: Les CHRS, dispositif, usagers, intervenants. Paris: Dunod. Pichon, P. (2002). Au nom d?une exp?rience commune de sans domicile fixe. M o- bilisation collective et entraide. In V. Ch?tel & M. -H. Soulet (Eds.) Faire face et s?en sortir. Vol. 2 (pp. 15-22). Fribourg, Switzerland: ?ditions Universita i- res de Fribourg. Podymow, T., Turnbull, J., Tadic, V., & Muckle, W. (2006). Shelter-based conva- lescence for homeless adults. Canadian Journal of Public Health, 97(5), 379- 383. Poirier, M., Chanteau, O., Marcil, F., & Guay, J. (2007). La pr?vention de l?itin?rance et l?autonomisation des jeunes plac?s en Centre jeun esse. In S. Roy & R. Hurtubise (Eds.) L?itin?rance en question . Sainte-Foy, QC: PUQ. Poirier, M., Hachey, R., & Lecomte, Y. (Eds.). (2000). L?inqui?tante ?tranget ? de l?itin?rance. Sant? Mentale Au Qu?bec , 25(2), 9-20. Pollio, D. E., Thompson, S. J., Tobias, L., Reid, D. & Spitznagel, E. (2006). Longi- tudinal outcomes for youth receiving runaway/homeless shelter services. Journal of Youth and Adolescence , 35(5), 852-859. Poole, D., & Zugazaga, C. B. (2003). Conceptualizing prevention as the first line of offense against homelessness: Implications for the federal continuum of care model. Journal of Primary Prevention, 23(4), 409-424. Power, R., French, R., Connelly, J., George, S., Hawes, D., Hinton, T., Klee, H., Robinson, D., Senior, J., Timms, P., & Warner, D. (1999). Health, health promotion, and homelessness. British Medical Journal, 318(7183), 590-592. Racine, G. (1993). L?intervention en sant? mentale: L e mandat inattendu des mai- sons d?h?b ergement pour femmes sans abri. Sant? Mentale Au Qu?bec , 18(1), 251-268. Roman, N., & Berg, S. (2004). Legislation, programs and policies. In D. Levinson (Ed.) Encyclopaedia of homelessness (pp. 340-346). Thousand Oaks, California: SAGE Publications. Roy, S., & Hurtubise, R. (Eds.). (2007). L?Itin?rance en question . Sainte-Foy, QC: PUQ. Roy, S. et Hurtubise, R. (2004). ? Homelesness in Montr?al ?, in Encyclopedia of Homelesness, Levinson, D. (Ed.) New Haven: Sage Publications. Roy, S., Morin, D., Lem?tayer, F., & Grimard, C. (2006). Itin?rance et acc?s aux ser- vices: Probl?mes et enjeux . Rapport de recherche soumis au RHDCC, dans le cadre de l?IPAC. Montreal: CRI. Roy, S., Rh?aume, J., Rozier, M., & H?tu , P. (2000). L?h?bergement des jeunes mineurs en difficult?: Une solution?. In D. Laberge (Ed.) L?errance urbaine (pp. 405-416). Sainte-Foy, QC: ?ditions Mult iMondes. Roy, S., Yanick, N., & Thomas, G. (2003). Le support communautaire en logement social. Montreal: CRI, RAPSIM. Serge, L., Eberle, M., Goldberg, M., Sullivan, S., & Dudding, P. (2002). Organismes de protection de la jeunesse et itin?rance chez les jeunes au Canada . Ottawa: Ini- tiative nationale pour les sans-abri. S?vigny, O., & Racine, G. (2002). Liens sociaux et pratiques en maisons d?h?bergement pour fem mes en difficult? et sans abri. Intervention, 115, 75- 85. Simard, M. (2000). Entre l?asile et le refuge: Rep?res d?un par cours. Sant? mentale au Qu?bec , 25(2), 132-152. Simard, M. (2005). Le syst?me d?urgence sociale du pavillon Webster de la Mi ssion Old Brewery (Research Report). Trois-Rivi?res , QC: la Mission Old Brewery de Montr?al. Snyder, M. D., & Weyer, M. E. (2002). Educational innovations: Facilitating a collaborative partnership with a homeless shelter. Journal of Nursing Educa- tion, 41(12), 547-549. Soulie, C. (1997). Le classement des sans-abri [Classifying the h omeless]. Actes de la recherche en sciences socials, 118, 69-80; 92-95. Stark, L. R. (1994). The shelter as total institution. American Behavioral Scientist, 37(4), 553-562. Stein, J. A., Lu, M. C., & Gelberg, L. (2000). Severity of homelessness and adverse birth outcomes. Health Psychology, 19(6), 524-534. Susser, E., Valencia, E., Conover, S., Felix, A., Tsai, W.-Y., & Wyatt, R. J. (1997). Preventing recurrent homelessness among mentally ill men: A ?critical time? intervention after discharge from a shelter. American Journal of Public Health, 87(2), 256-272. Teare, J. F., & Peterson, R. W. (1994). Treatment implementation in a short-term emergency shelter program. Child Welfare, 73(3), 271-281. Thibaudeau, M.-F. (2000). L?acc? s aux services de sant? et leur utilisation par les personnes itin?rantes. In D. Laberge (Ed.) L?errance urbaine (pp. 313-328). Sainte-Foy, QC: ?ditions Multimondes. Tolomiczenko, G., & Goering, P. (2001). Gender differences in legal involvement among homeless shelter users. International Journal of Law and Psychiatry, 24(6), 583-593. United States Census Bureau. (2001). Emergency and Transitional Shelter Popula- tion: 2000 . Washington, DC: Author. Wadd, S. L., Hutchinson, S. J., Taylor, A., Ahmed, S., & Goldberg, D. J. (2006). High risk injecting behaviour in hostel accommodation for the homeless in Glasgow 2001-02: A study combining quantitative and qualitative metho- dology. Journal of Substance Use, 11(5), 333-341. Webb, E. (1998). Children and the inverse care law. British Medical Journal, 316(7144), 1588-1591. Zeneidi-Henry, D. (2002). Les SDF et la ville: G?ographie du savoir -survivre. Rosny- sous-Bois, France: Bor?al. Chapter 1.3 One in Five?Housing as a Factor in the Admission of Children to Care SHIRLEY CHAU, ANN F IT ZPATRICK , J. DAVID HULCHANSKI, BRUCE LESLIE, AND DEBBIE SCHATIA The Children?s Aid Society of Toronto (CAST) is mandated to protect children under the age of sixteen in the community under Ontario?s Child and Family Services Act. It is the largest board-operated child wel- fare organization in North America and has been serving children and families for more than 100 years. CAST has a legal responsibility to protect children at risk of abuse and/or neglect. One form of intervention involves removing children from their parent?s home and providing substitute care. A child comes into substitute care of a child welfare agency by one of two methods. Under a voluntary agreement, the parents agree to tem- porarily place a child into CAST?s care. Otherwise, the children are placed in CAST?s care through an apprehension by the Family Division of Provincial Court. Apprehensions occur when there is serious and immediate danger to a child?s well-being. The decision to place a child in care is made on the basis of input from a variety of professional assessments by commu- nity and child welfare workers. The assessment is coordinated by family service workers. Family service workers are professional social workers who visit families in their home as a routine practice, and have first-hand informa- tion of a family?s housing circumstances. Therefore, these social workers are particularly knowledgeable about the major factors affecting the fam- ilies and children they work with. This report summarizes the findings of a research project designed to determine the extent to which housing is a factor in the decision to place children in care and the decision to return them to their homes. The survey asked family service workers two key questions and several follow-up questions about the housing conditions of CAST clients: ? In your opinion, was the family?s housing situation one of the factors that resulted in admission of a child or children into care? ? In your opinion, was there any delay of the return home of the child from care due to housing-related problems? This research replicates a study carried out in 1992 (Cohen- Schlanger et al., 1995), the results of which were widely reported and discussed. In its May 1993 report on Canada, for example, the United Nations Committee on Economic, Social and Cultural Rights in Geneva referred to the 1992 study: ?Paragraph 14. The Committee received in- formation from non-governmental organizations about families being forced to relinquish their children to foster care because of inability to provide adequate housing or other necessities.? The Committee asked Canada to explain why this was occurring and made recommendations to encourage progress on this issue. The questionnaires provided opportunities for respondents to write in additional comments. Researchers also conducted a follow-up inter- view with some of the family service workers. A premise of this research is that access to adequate and affordable housing will not necessarily prevent child admissions to care. However, adequate housing may: (a) reduce the number of admissions by stabiliz- ing families? living situations in ways that promote children?s well-being; and (b) reduce the delay in the return of children to their homes because of housing problems. This study raises the broader question that no one study on its own can answer: Could the incidence of child abuse and neglect be reduced if more families had access to affordable, adequate, and appropriate hous- ing? This is a critical question for all child welfare organizations, all le- vels of government, and the community in general. Research method The design of this project was similar to the 1992 study (Cohen Schlanger et al., 1995). All family service workers employed by CAST were asked to complete a questionnaire on two of their case files. The CAST?s Execu- tive Director gave approval to carry out the study using the proposed method and the University of Toronto?s ethics review office also ap- proved the method. CAST?s staff researcher reviewed the questionnaire and helped address practical issues such as obtaining an appropriate sample and maintaining confidentiality. The draft questionnaire and proposed method was circulated to a number of housing and child wel- fare experts for comment. The questionnaire was pretested and modifica- tions were made as a result. A case file is opened for every child who is admitted to care. To en- sure that there was no selection bias by the family service workers, the two case files for each worker were selected on a random basis by the research team. The aim of this process was to achieve a sample that would accurately represent the population of children admitted to care. The random selection process was adjusted to avoid selection of more than one child per family. The sample was selected from children?s case files that were open from September to December 2000 (a few months before the question- naires were distributed). About 950 case files were open in each of these months, and, adjusting for continuing cases, a total in-care sample of 1,331 distinct cases was obtained. A randomized selection of two cases for each family service worker resulted in a final sample of 271 cases. This is a 32% increase in sample size from the 1992 study. Response rate Allowing for vacations, illness, and turnover among family service workers, a good response rate was obtained. At the time of the study there were 128 family service workers at the agency, of which 106 re- turned the questionnaires (an 83% participation rate). Of the 271 ques- tionnaires distributed, 191 were returned (70% ). In the 1992 study, 108 family service workers were surveyed, and 69 returned their completed questionnaires (a 63.8% participation rate). Of the 205 children?s cases in the total sample, questionnaires were returned for 128 (62.4% ). This relatively high response rate and the random sampling tech- nique allows us to be confident that the findings from this sample can be generalized to other CAST admissions of children to care. Sample characteristics The characteristics of the families and children in the random sample are summarized in Table 1. There was an increase in the admissions of child- ren to care through apprehensions: 74% of the cases in 2000 involved apprehensions versus 68% of the cases in 1992. Apprehensions involve the most serious cases, in which a child is deemed to face an immediate risk of abuse, neglect, or abandonment. Table 1. Comparison of 1992 and 2000 Sample Characteristics 1992 2000 Children from single-parent families 68% 74% Estimated monthly family income (median) $1,225 $1,500 Families receiving welfare/family benefits 70% 56% Families receiving unemployment insurance 5% 1.6% Families living in public housing (MTHA) 23% 30% Median age of the child placed into care 6.5 yrs old 11 yrs old The child?s gender 53% M 47% F 48% M 52% F Survey findings (A) Housing as a factor in admissions to care In one out of five cases (20.7% ) the family?s housing situation was a fac- tor that resulted in temporary placement of a child into care. This is a significant increase from the 18.4% in the 1992 study. According to Ontario?s Child and Family Services Act, inadequate housing or housing problems are not sufficient grounds to consider a child in need of protection. Even homelessness, in the absence of other concerns, is not sufficient legal grounds for placement of a child into CAST care. Social workers use community services such as shelters and legal clinics to deal with housing problems and homelessness. Within these legal restrictions, it is significant that the family service workers identified 20.7% of their child admissions as cases in which they considered housing a factor in the decision to place the child in care. This represents 39 cases out of 191. Of these 39 cases, 10 cases (25.6% ) were admitted by voluntary agreement and 29 cases (74.4% ) cases involved apprehensions. These percentages are similar to those in the 1992 study. In this study, the ratio between the number of apprehensions com- pared to voluntary agreements is higher in cases in which housing is a factor. Housing was a factor in 26% of the voluntary agreement cases, and a factor in 74% of apprehensions. This is in contrast to the total ad- missions to care in our survey (whether or not housing was a factor), where 32% were voluntary agreements and 68% were apprehensions. This finding suggests that where housing is a factor, there is greater risk of the abuse or neglect of a child. In many of the cases in which housing was a factor, the family service workers had serious concerns about the child?s welfare, including the risk of physical abuse, emotional abuse, or abandonment. This substantiates other findings in the litera- ture that link the family housing situation with child welfare (Trocm? et al., 1994; Courtney et al., 2004). (B) An affordable, safe, and appropriate fam ily housing situation In 8.6% of the cases, families ?did not have housing that would be af- fordable now,? and in 20.1% of the cases families did not have housing considered ?safe and appropriate to meet their physical housing needs.? Compared to the 1992 study, these results indicate an increase in housing problems. Family service workers were asked to assess whether the family had affordable, safe, and appropriate housing. These two ques- tions were asked of all the cases, not just the 21% of cases in which hous- ing was known to be a factor. In 29% of the cases, families ?did not have housing that would be affordable now? (compared to 23% in 1992), and in 21% of the cases, families did not have housing considered ?safe and appropriate to meet their physical housing needs? (compared to 14% in 1992). Compared to the 1992 study, therefore, these results indicate an increase in housing problems related to affordability and to the safety of the child. These results indicate some potential for further admissions to care if the affordability problem worsens for the family or if the safety or appropriateness (e.g., overcrowding) of the family?s home deteriorates. (C) Housing situation a factor in delaying the return home In 11.5% of the cases, the return home of a child from care was delayed due to a housing-related problem (an increase from 8.6% in the 1992 study). Family service workers were asked if there was any delay in re- turning the child to the family because of housing-related problems. Whether or not housing is a factor in the initial decision to place a child in care, the family may subsequently develop a severe housing problem. CAST will postpone the return of a child until an assessment is made that the family has secured adequate housing. Out of the 134 responses to this question, family service workers re- ported that they had delayed the return home of a child due to housing- related problems in 11.5% of cases. This is a significant increase from the 8.6% of the cases reported in the 1992 study. In the cases in which the return was delayed, the respondent was asked to estimate the length of the delay in months. The delay was reported for 14 out of 15 cases. Com- pared to the 1992 study, the length of the delay is three times as long. Table 2. Reported Delay of a Child Due to Housing-Related Problems Delay No. of cases 1 month 1 3 months 2 5 months 1 6 months 1 12 months 3 More than 12 months 6 (D) Nature of the housing problem when there is a delay in re turn home In the 11.5% of these cases (n=15) in which the return home was delayed due to housing-related factors, family service workers were asked about the nature of the housing problem. More than one answer was possible. The two most common reasons noted by workers were ?No permanent housing for the family? and ?Inadequate income.? Table 3. Nature of the Housing Factors, 2000 In cases where the family?s housing situation was a factor in keeping the child in care (n=15 in 2000) Important or very important No permanent home for the family 75% Inadequate income 73% Inadequate health standards 71% Inadequate amount of living space 67% No affordable housing for the family 67% No first and last months? rent 54% Conclusion and discussion This survey indicates that Toronto?s housing situation is having a detri- mental effect on the well-being of many families with children. The situ- ation was worse in 2000 than it was in 1992. The families and children who are clients of the Children?s Aid Society of Toronto are among the most economically disadvantaged in Ontario. They face substantial ob- stacles to obtaining adequate and appropriate housing, and for some this affects their ability to care for their children. The finding that in 20.7% of the cases surveyed the family?s housing situation was one of the factors that resulted in the temporary placement of a child or children into care indicates how serious the situation is for many families. This is a significant increase since the 1992 survey. Given that there were 2,250 CAST cases during 2000, this finding means that about 450 children were in care that year at least partly because of their families? housing situation. The financial cost of a child in care is very high, averaging about $40,761 pe r child. (This figure is based on the CAST estimate for an aver- age month in care of $1,941 per month per child, and 21 months as the average length of time in care, during 2000.) This means that the cases in which housing was a factor in the admission assessment cost about $18 million a year. This survey also found that housing problems are delaying the re- turn of children to their families in 11.5% of cases. This is a significant increase from the 8.6% reported in the 1992 survey. During 2000, housing factors delayed the return of children to their families in about 250 cases. Even a one-month delay for the 250 cases is very expensive, costing CAST almost $500,000 (that is, $1,941 per month per child). Assessing the financial costs of child admissions to care does not in- clude the social and emotional costs, both short and long-term, of a child being placed in out-of-home care. Placement in out-of-home care is an intervention of last resort for CAST because extensive research has dem- onstrated the negative consequences of removing children from their parents. The method and scope of this survey is too limited to state more precisely the degree to which housing was a factor and the precise na- ture of the housing factors involved. The aim was to identify the extent to which housing is a factor in child admissions to care. Although this study cannot state that housing-related factors caused the admissions of children to care, the family service workers identify housing as a factor in one out of five of their cases during 2000. Clearly, there is a significant connection between a family?s housing situation and child admissions into care. The finding further suggests that a significant proportion of CAST?s budget for in-care cost is associated with the inability of some families in Toronto to obtain adequate housing. Addressing the housing needs of low-income families may be important in reducing child admissions and in facilitating a quicker return of children to their families. Access to safe and affordable housing will not necessarily prevent child admissions to CAST care, but housing support may reduce the number of admissions, stabilize the family?s living situation in ways that promote children?s well-being, and reduce housing-related delays in the return of children to their homes. Unfortunately, this study demonstrates that progress on this front has not been made in the eight years since the 1992 study. Shirley Chau is assistant professor in the School of Social Work at the Univer- sity of British Columbia. Ann Fitzpatrick works for the Children?s Aid Society of Toronto, Scarborough Branch. J. David H ulchanski is a professor in the Faculty of Social Work, University of Toronto and holds the endowed Dr. Chow Yei Ching Chair in Housing . Bruce Leslie is research and quality assurance analyst at the Children?s Aid Society of Toronto. Debbie Schatia is supervisor of family service workers, Children Aid Society of Toronto. References Cohen-Schlanger, M., Fitzpatrick, A., Hulchanski, J.D., & Raphael, D. (1995). Housing as a factor in admissions of children to temporary care: A survey. Child Welfare 74(3):547?562. Courtney, M.E., McMurtry, S.L., & Zinn, A. (2004). Housing problems experienced by recipients of child welfare services. Child Welfare 83(5):393?422. Trocm? , N., McPhee, D., Tam, K.K., & Hay, T. (1994). Ontario incidence study of reported child abuse and neglect. Toronto, Ontario: The Institute for the Prevention of Child Abuse. Chapter 1.4 The Toronto Shelter Zoning By- law : Municipal Limits in Addressing Homelessness PRASHAN RANASINGHE AND MARIANA VALVERDE If you are allowed to put a shelter anywhere you want in the city, it takes away a fundamental right of the public to have meaningful input into what occurs in their city... [Public input] is fundamental to local demo cra- cy. Paul Sutherland, Toronto City Councillor, April 2002 (quoted in La- key, 2002b, p. B5). The 1990s witnessed a dramatic rise in the number of homeless people in many North American cities, including Toronto. Their presence and vi- sibility was so pronounced that it garnered attention among the public, which in turn provoked strong, and rather divergent, responses from both sides of the political spectrum. One type of response to the problem of homelessness was the right- wing, law-and-order, ban them from ?our? city type approach. Ontario Premier Mike Harris?s Safe Streets Act, 1999, which banned aggressive panhandling and squeegeeing, was one notorious example (see, for ex- ample, Graser, 2000; Hermer & Mosher, 2002). On the opposite end of the political spectrum, responses were equally strong. In Toronto, the death of three homeless men in January 1996 prompted outrage not only about the deaths but also about the cuts in welfare and housing subsi- dies, which had, it was felt, indirectly led to these tragedies. The forma- tion of the Toronto Disaster Relief Committee led by anti-poverty activ- ists Cathy Crowe and Michael Shapcott, with a strong emphasis on hous- ing policy and homelessness, was one response to the crisis. In this paper, we focus on the City of Toronto?s efforts in the late 1990s to address the problem of homelessness by building more home- less shelters and spreading them across the city. We argue that attempts by municipal governments to address homelessness?and more broadly, matters of social justice?are likely to be thwarted, significantly delayed or deviate drastically from their original intentions on the one hand, or at worst, fail miserably. When municipalities are left to their own devices to battle the problem of homelessness, the result, our case study shows, fails to provide meaningful solutions in a timely and systematic way. Municipalities are ill-equipped to address homelessness for two main reasons. First, cities are fundamentally limited in the means they command to deal with social problems. Given the subordinate status of municipalities in Canadian law and politics, cities have very few legal tools to attend to local matters?this is still the case, despite the highly touted ?new deal for cities? in Canada (see, for example, Valverde & Le- vi, 2005). Municipalities therefore rely heavily on zoning, one of the few legal tools they have at their disposal. This means that matters that might be better suited to other types of legal solutions, if brought before muni- cipalities, end up funnelled into zoning and planning mechanisms. Land use law (of which zoning is the most important component) has never been about substantive democracy, equality or social justice (see, for example, Blomley, 2004; Fischler, 1998; Frug, 1999; Gerecke, 1976; Gunton, 1979). Rather, land use law, since its inception, has worked primarily to protect property values, segregate certain ?undesirable? uses of land, and generally, to constitute an urban space that is highly differentiated not only by class, but also along other lines as well (for example, single versus multi-family dwellings). In other words, land-use law, and in particular, zoning, allows the segregation and compartmen- talization of spaces according to uses. It governs spaces and uses, not persons; this, by extension, also means that land uses, in and of them- selves, have no rights. Thus land uses that provide solace to the very poor?for example, shelters or supportive housing?have no rights. Moreover, since homeless people have no (immobile) property to call their own, they are excluded from relying strictly on a rights-based ap- proach (for example, the right to shelter), because rights, in land use law, are tied to uses and not to persons. Thus, land-use law in particular, and municipal politics by extension, can do very little to provide meaningful solutions for many homeless people. Second, municipalities must follow procedures for public participa- tion in local policy formation, especially in cases of land use and (re)zoning matters. Here, a problem arises, because land-use law resists democratization. We do not mean that municipal politics does not facili- tate a forum for interested parties to voice their opinions and concerns; nor that particular groups are excluded from participation in local policy formation?indeed, both those who opposed the spreading of shelters and those who favoured it relied on their right to participate in the de- bates surrounding the by-law so as to influence and shape its content. What we mean is that, given that rights in land-uses are tied to property, it is usually the case that those groups who end up influencing particular land uses are those who have legal occupancy in relation to a particular property (that is, residents, ratepayers, and tenants). John Se- well, the former mayor of Toronto, in his book The Shape of the City (1993), shows that before the 1960s all planning issues were undertaken and put into practice by experts, without any public input (Sewell, 1993). All this changed however, in the 1960s, when citizens, particularly resi- dents? groups, began opposing planners, often because they were deeply dissatisfied with the vision that planners had for their neighbourhoods (for more recent trends, see Hume, 2005b; for similar trends in the U.S., see for example, Arnstein, 2003). And although neighbourhood groups often failed to halt proposed developments, residents? and ratepayers? groups?that is, the propertied and those who have legal occupancy? were heavily embroiled in fights about development projects. While public input is, theoretically, open to all concerned citizens, it is often the propertied who have most to say about development proposals. Advo- cates for the homeless and for homeless shelters cannot construct an ar- gument based solely on rights. That is why we argue that land-use law resists democratization. While the process of public input is open to all concerned parties, the nature of municipal politics renders those without property unable to rely on the notion of rights to make a claim to shelters. Thus, even though the by-law eventually passed, it did so after considerable delay and haggles over its content, primarily because the input into its content came via property owners, and by extension, those who did not want shelters in their ?backyards.? The comment of councillor Paul Suther- land quoted at the beginning of this chapter is typical. While Sutherland lauds the idea of public input into local policy formation, he assumes that all concerned parties have equal status to voice their opinions. The Mayor?s Homelessness Action Task Force In November 1998, when the mayors of Canada?s major cities convened at the ?Big City Mayor?s Meetings? in Winnipeg, Manitoba, homeless- ness was labeled as a ?national disaster? which was deemed to require immediate political attention (Layton, 2000). Even before the meeting, in January 1998, the mayor of Toronto, MeI Lastman, had formed the Mayor?s Homelessness Action Task Force in an effort to provide a systematic study of, and solutions to, the prob- lem of homelessness. The Task Force was made up of four members and chaired by Dr. Anne Golden, who, at that time, was President of the United Way of Greater Toronto. In July 1998 the Task Force released its interim report, Breaking the Cycle of Homelessness and in January 1999, the final report, Taking Responsibility: An Action Plan for Toronto. Taking Responsibility outlined 105 recommendations for action.1 With respect to housing, the Task Force recommended three distinct initiatives: affordable housing programs, supportive housing programs (that is, housing plus support services), and shelters. The first two were Two key themes emerge from these recommendations: first, that prevention and long-term approaches ought to replace reactive and emergency-type responses to homelessness, and second, that all three levels of govern- ment must take responsibility for solving it. With respect to long-term solutions, the Task Force recommended a ?housing first policy:? that is, the undertaking of long-term rather than short-term solutions which seek to house rather than merely shelter homeless people. Taking Respon- sibility clearly and repeatedly noted that homelessness was a problem of housing, or to be more precise, a lack of affordable housing. meant to be long-term solutions while the third was merely a short-term solution until the first two could be implemented. The Task Force rec- ommended that more shelters be built and that they be spread equitably across the city to ensure that homeless people would be able to easily access services and in so doing, would maintain their ties with the com- munity (this holds true especially for homeless children, who would oth- erwise have to be removed from their schools).2 Spreading homeless shelters across the city was much more difficult than originally anticipated. Under the zoning provisions then in force, shelters were allowed only in the municipalities of Toronto and North York; Scarborough had also facilitated the housing of homeless families in motels on Kingston Road to help families find a temporary roof over their heads. Homeless people in Etobicoke, East York, and York did not have access to shelters in their areas. As well, where shelters were per- mitted, they were regulated by spatial constraints. For example, zoning provisions prohibited the location of two ?crisis care facilities? within 250 metres of each other (a homeless shelter was defined as a crisis care facility). The Task Force, aware of these restrictions, recommended a process of inclusionary zoning, whereby the city would be permitted, as of right, to locate homeless shelters where it pleased, as long as the shelter met zoning criteria for height and density. This provision would also include rooming houses, affordable housing units, and supportive housing units. However, the Task Force clearly noted that this process must be opened up to the public for their input. The Toronto City Council accepted the recommendation of inclusionary zoning to locate shelters. However, be- fore discussing how this process led to a protracted and heated debate, it is worth explaining the idea behind inclusionary zoning. In the early 20th century, zoning was developed to demarcate land- uses within a particular geographical area; it operated under the prin- ciple of excluding ?inappropriate? land-uses from a particular space. Part of the appeal of exclusionary zoning was that it boosted property values, but it had a negative consequence?exclusionary land use trans- lated into the practice of excluding certain people from particular places: What was good for business was the right kind of people: the right cus- tomers downtown, the right neighbors in the new street car suburbs... [F]ar from being a device to spread the transition of the immigrant poor from the tenements to the street car suburbs, zoning in practice became a way of keeping them where they were (Hall, 1989, p. 278). In calling for a process of inclusionary zoning then, the Task Force understood the effects of exclusionary zoning, and was attempting to manoeuvre around legal and traditional planning mechanisms to facili- tate the creation of shelters. Drafting an enacting by -law Immediately following the release of Taking Responsibility in January 1999, Toronto City Council authorized various sub-committees to advise Council on the implementation of the Task Force?s recommendations. On February 17, 1999, the Chief Administrative Officer?s Office released its Response to the Mayor?s Homelessness Action Task Force Final Report. Based on the conclusions of this report, City Council, on March 2, 1999, en- dorsed, in principle, the 105 recommendations made by the Task Force, including the recommendation to locate homeless shelters in various parts of Toronto. To this end, the report noted: The City of Toronto is charged [with] taking the lead with planning and managing local homeless programs. In addition, the City is called upon to use the existing urban planning tools at its disposal and to seek additional powers to provide a framework for the development and preservation of affordable housing. The ?urban planning tools? were the zoning provisions then in force. A report prepared by the Commissioner of Urban Planning and Development Services, April 15, 1999, defined inclusionary zoning in this way: Inclusionary zoning for affordable housing is a land development control measure, enacted by way of municipal by-law, which generally requires a certain portion of the units within any new residential development to be set aside for low and/or moderate income households at below market prices or rents. A similar principle was to govern the spreading of homeless shel- ters: in other words, homeless shelters were to be included in, rather than excluded from, residential and industrial sites. Council drafted a by-law which would allow homeless shelters to be located in any part of the city, as of right. On May 11, 1999, Council- lors Joe Pantalone and Chris Korwin-Kuczynski moved that council ?adopt policies necessary to override existing zoning by-laws ... across the amalgamated city ... to ensure that new emergency shelter[s] can be opened as needed.? It appeared that the stage was set for the creation of homeless shelters throughout the city. This however, was not to be the case; at least not for another five years. Between this time and the actual passing of the by-law, efforts to open shelters in various parts of the city brought negative attention to Council?s actions; the attention turned into a powerful force that delayed the passing of the by-law. Two examples warrant discussion because they illustrate why the proposed by-law took so long to pass, and the particular concerns and issues that had to be dealt with. Resistance in Scarborough In summer 1999, there was a proposal to build a senior men?s hostel at 1673 Kingston Road in Scarborough. When the proposal was put for- ward, councillors Gerry Altobello and Brian Ashton raised the concerns of their constituents and asked that council not authorize the proposal for the following reasons: [The] use of an emergency shelter or an hostel is not permitted use under the Zoning by-law for this property; and our office has been inundated with calls from local residents against this proposal; and the community and the Principal from the Birch Cliff Public School located across the street are concerned about the impact on the safety of the children. Even though the motion failed, both Councillors gave notice that they would request permission to consider this matter in subsequent Council meetings. A public meeting scheduled for October 6, 1999, con- cerning the proposed shelter would, according to the Councillors, give sufficient grounds to halt the proposal. During the City Council debates on October 26, 1999, Altobello and Ashton introduced several pieces of evidence against the proposed shelter, including petitions signed by 1,350 concerned residents as well as numerous letters they had received. While they were unsuccessful in halting the building of the proposed shelter, they were successful in implementing several restrictions. The number of beds would be capped at 60, rather than the proposed 70 spaces. Potential clients would be ?screened? and occupants would be well known to staff before taking up occupancy. The Commissioner of Community Services had to report, by the end of 1999, on the effects of the hostel on the community. Finally, a Community Reference Board of 12-15 persons, made up of local residents, local business persons, local schools, the Toronto Police Services, and community organizations, would ?review profiles of individuals as they come to the building.? These add-ons were no doubt an effort to appease the residents of the Scarborough community and it appears that they did just that. A staff report released on May 30, 2000, gives a preliminary status of the hostel, which was named Birchmount Residence, by noting that ?the community has become actively engaged in the day to day life of the res- idence? and that ?to date, there have been no complaints.? A proposed shelter on the Danforth In July 2001, the Thunder Night Club located on Danforth Avenue and Dawes Road in Toronto was slated to be demolished and turned into a homeless shelter. Concerned citizens of the area took to the streets in protest saying that the shelter was ?sprung on them unannounced and [that] the community should have been involved in the decision making process? (Royce-Roll, 2001). The citizens feared that they would find no solution to the violent and raucous behaviour that often ?spilled? into the streets after the nightclub closed for the night, and believed they would find little reprieve once the homeless shelter opened; in particu- lar, they felt that the early (7 a.m.) discharge protocols of homeless shel- ters would result in many homeless people lying around the streets near their residences. Part of the problem was that the scheduled shelter would be located within 250 metres of an existing shelter, in violation of zoning provi- sions. In response, the Director of the Toronto Hostel Services, John Jagt, argued that the new facility would not be considered a crisis care facility, and therefore, could lawfully operate. This plan was foiled in the courts however, when the Ontario Superior Court, in March 2002, ruled that the Danforth Night Club project could not proceed because the proposed shelter did fit the description of a crisis care facility and therefore, did violate existing zoning regulations (Lakey, 2002a). The by- law battle The fact that many citizens thought that the shelter proposal had been ?sprung? on them without notice was of concern; this was the case with the proposed by-law as well, where many residents felt that their voices were being excluded during the drafting of the by-law. The right to voice one?s concerns and thereby shape public policy was relied on by both those who wished to see the by-law pass, and those who vehemently opposed it, though in different ways. On the one hand, politicians such as Paul Sutherland used the right of the public to participate in policy formation as a way to forestall the enactment of the shelter by-law. Other politicians who were keen on see- ing the by-law pass, wanted to circumvent the public?s right to partici- pate in policy formation to ensure that homeless people were provided with some sort of reprieve. For example, Councillor Jack Layton realized that community concerns would simply translate into free-for-all NIM- BYism, was quoted as saying, ?Zoning, by definition, is an exclusionary process... we can?t be exclusionary when it comes to services of the homeless, in my view? (Lakey, 2002b). Similar sentiments were proc- laimed by councillor Joe Pantalone, who said: Regretfully, a lot of people disguise their feelings that somehow the homeless people have only themselves to blame by bringing in extrane- ous arguments or simply succumb to constituents who are afraid. The problem is, we have to do what?s right and not play to the fears of our constituents (Lakey, 2002b). The manoeuvring around democratic participation captures the complexities involved in the passage of a contentious piece of legislation. On the one hand, politicians had to, and indeed wanted to, find mea- ningful ways to tackle the problem of homelessness. On the other hand, they also had a duty to listen to what the public had to say. MeI Lastman was well aware of the pressures in the situation: Look, I want this [the problems over the by -law] like I want a second head. I know people don?t want it in their backyard, but you can?t just have them in downtown Toronto... I would rather have voted and ended it, one way or another. But I felt it was going to create a major problem. I like the idea of building a consensus across the city because I know what we?re in for, in the future. I know the people are going to come yelling and screaming that we know nothing about this and you?re putting a homeless shelter in my backyard (Lakey, 2002c). These political complications delayed the passing of the by-law for three years. It was not until April 17, 2002 that council finally began de- bates concerning the by-law. The very next day, council voted 27-16 to refer the bylaw to Mayor Lastman?s office for further study and public consultation, and following this, to proceed to the Department of Plan- ning and Transportation Committee for further debate. On April 18, 2002, council set a date for October 2002 for all reviews, consultations and studies to be completed, so that council could vote. In September 2002, the six municipalities that make up the ?megacity? of Toronto held public consultation meetings on the by-law (Gillespie et al., 2002). Public concerns were studied by the Planning and Transportation Committee between September and December 2002. On January 28, 2003, the matter came back to Council for final debates. Particular councillors made concerted efforts to voice the opinions of their constituents, and impose restrictions on the by-law. For example, Councillor David Soknacki moved that Council develop appropriate ways to select shelters. That is, Council was to consider community safe- ty especially where public schools are concerned. Soknacki also wanted a system of notification for community members who would be kept ab- reast of what was taking place with respect to locating shelters. Three issues came to dominate the last efforts to impose some re- strictions on the by-law. First, the minimum distance of 250 metres sepa- rate one crisis care facility from another was proposed to be maintained. However, even this distance did not satisfy all councillors; Councillor Sutherland argued for a minimum distance of 1,000 metres. Second, some councillors called for locating shelters only on arterial roads, rather than on residential streets. Third, there were proposals to limit the num- ber of shelter beds per facility. On February 11, 2003, the municipal shelter by-law 138-2003 passed, with several modifications, apparently the result of last-minute efforts on the part of Councillors to impose some restrictions. The by-law allowed the city, as-of-right, to locate homeless shelters anywhere in the city, as long as it complied with applicable zoning provisions of the zone or dis- trict (that is, height and density requirements). However, the by-law re- quired that shelters be located only on major or minor arterial roads (the ?arterial road requirement?); that a minimum distance of 250 metres separate one shelter from another (the ?separation distance require- ment?); and that council approve each and every location of a homeless shelter. What began as an effort to allow shelters as of right in any part of the city resulted in a by-law fraught with restrictions, making it difficult and expensive to create shelters even on existing city properties. The last hurdle: The ruling of the OMB The by-law was subsequently appealed to the Ontario Municipal Board (OMB).3 Initially, the appeals concerned certain site-specific exemptions; that is, the appeals were geared towards ensuring that particular loca- tions were ?outside? the requirements of the by-law: there were 15 such appeals. Fourteen related to a site at 101 Ontario Street, home to Sojourn House, an emergency shelter (OMB Decision No. 0923, p. 1). The other was brought by a concerned resident whose property abutted a seniors? residence, at 717 Broadview Avenue; this site, which the city had pur- chased, was slated to be turned into an emergency shelter in the near future. The resident wanted to ensure that the property be subject to, not exempt from, the requirements of the by-law4 During the pre-hearings (hearings held to determine if sufficient evidence exists for a formal hearing) on July 8 and 9, 2003, the Advocacy Centre for Tenants Ontario (ACTO) and the Confederation of Residents and Ratepayers Association (CRRA) sought party status in the proceed- ings to voice particular concerns outside these specific issues related to site exemption. The ACTO (and Sojourn House as well) argued that the restrictions imposed on the by-law ?had no legitimate planning basis.? In addition, the ACTO argued that the requirements of the by-law vi- olated section 15 of the Charter of Rights and Freedoms, and hence, ought to be ruled as unconstitutional (OMB Decision No. 0923; OMB Decision Number 0569). 5 (OMB Decision No. 0569; see also Gillespie, 2003). The CRRA wanted more stringent requirements and sought relief to argue for a minimum separation distance of 1,000 metres between shelters; the CRRA also sought to have the size of these shelters capped at fifty beds (Gillespie, 2003). Both the ACTO and the CRRA were granted party status by the OMB6 The hearings, which began on September 29, 2003, occupied 21 days spread over two and a half months; considerably more than the 15 days that were originally set aside (OMB Decision No. 0569). The Board began by noting that the municipal shelter by-law ?represents a compromise of various community and business positions? (OMB Decision No. 0923) and that it is an ?interesting aspect to this matter that all parties wish ... to see the by-law approved, albeit in different forms? (OMB Decision No. 0569). (OMB Decision No. 0923). In reaching its decision, the Board acted more as a mediator than an arbitrator, seeking to appease all parties concerned. The Board began by acknowledging the fact that the intention of the by-law was to ensure that an adequate supply of homeless shelters in various parts of the city would become a reality, so that homeless people in various parts of the city would not be denied a temporary roof over their heads (OMB Deci- sion No. 0569, p. 17-19). The Board ruled that the ?separation distance requirement? was based on sound planning principles, because it would ensure that shel- ters were not concentrated in one particular area. Hence, the Board upheld this requirement of the by-law (OMB Decision No. 0569, p. 22- 23). The Board also ruled that the ?arterial road requirement? was based on sound planning principles and dismissed the view that the purpose of the ?arterial road requirement? was to ensure that shelters would not be located in residential neighbourhoods. The Board rather disingenuously noted that both major and minor arterial roads abut and even cut across residential neighbourhoods, so that this requirement was not geared to- wards keeping shelters away from residential neighbourhoods, but was an attempt to locate them within particular communities, with the specif- ic purpose of ensuring that homeless persons do not lose ties with their communities (OMB Decision No. 0569, p. 22, 20). This creative interpretation allowed the Board to replace the ?arteri- al road requirement? with the ?arterial road corridor requirement.? This new requirement allowed a shelter to be located either on an arterial road, or within 80 metres of a flanking street which abutted an arterial road: ?The Board finds that the arterial road corridor location should include any lot, the whole or part of which, is located on a flanking street to an arterial road to a distance of 80-metres from the corner of the ar- terial road and flanking street? (OMB Decision No. 0569, p. 25-26). This modified approach, the Board concluded, ?makes ... shelters] more a c- cessible for the users... improves accessibility to the required services by the users, and... encourages the distribution of the shelters on a wider basis across the City? OMB Decision No. 0569, p. 21). It is not entirely clear what led to the modification of the ?arterial road requirement.? However, it seems quite plausible that this was a concerted effort on the part of the Board to appease both sides concerned; and this modified approach seems to have done just that. The Board removed the requirement that Council approve every lo- cation, because the section ?compromises the integrity of the by-law as a zoning mechanism, is redundant, and without any land use purpose, creates uncertainty, and should not be included in the by-law? (OMB Decision No. 0569, p. 29). The Board concluded thus that: By-law 138-2003, as modified by this Board, conforms to the principles of good planning, and all applicable planning policy documents, and is supported by sound planning rationale... [T]he by -law will facilitate the achievement of the City?s program and service delivery objectives with respect to homelessness. The by-law will increase the number of sites across the City available for use as an emergency shelter, and properly di- rects the emergency shelter use to locations, which will meet the needs of the users, while minimizing the possible impacts of the use on neigh- bourhoods (OMB Decision No. 0569, p. 8). This approach served to preserve, to a small degree, Council?s orig- inal intentions of making shelters more accessible and at the same time, appease concerned parties to the appeal. With these modifications, To- ronto?s municipal shelter by-law finally passed. Although the passage of the by-law, theoretically at least, represented a victory for homeless people and those who advocate on their behalf, the victory came with a large price tag, which included not only several modifications to the proposed bylaw, but more importantly, the five years required to resolve the matter. Canadian law does not have the strong protection against segrega- tion and discrimination from zoning practices as afforded in the Ameri- can Fair Housing Act (which has been successfully used by public hous- ing and supportive housing providers, as well as by victims of racial segregation). However, it is nevertheless a principle of Canadian munic- ipal law that zoning powers cannot be used to discriminate against dis- advantaged groups; hence, what is commonly referred to as ?people zoning,? while not completely forbidden, is legally suspect and subject to constitutional challenge, given that municipalities are supposed to go- vern uses and not persons. In Canadian law, the protection afforded to residents of group homes and other non-standard households from discriminatory zoning is weak. The leading case on this issue is Re. Alcoholism Foundation of Ma- nitoba et al. and City of Winnipeg (1990) in which the Manitoba Court of Appeal struck down a Winnipeg by-law which named disabled and sub- stance-dependent individuals in its zoning provisions for group and re- habilitation homes. Monin C.J.M. (the then-Chief Justice of the Manitoba Court of Appeal) even went as far as stating that, as far as he was con- cerned, the exclusionary logic of zoning was by no means problematic, as long as particular disadvantaged groups, such as the ?disabled,? were not explicitly named. In other words, for a by-law to meet constitutional muster, it ought not name specific groups. Monin noted: Ratepayers building $150,000 or 200,000 single -family homes are entitled to expect that only similar homes will be built in their vicinity, and that the integrity of that particular zoned area in the community will not be in- terfered with... That was and should still be an entirely legitimate concern of the city councillors. Likewise, they should be free to protect those of lesser means from infiltration in their areas by businesses, manufactures, or other commercial ventures not in conformity with their legitimate aspi- rations for a modest residential area (Re. Alcoholism Foundation of Manitoba et al. and City of Winnipeg, 1990, p. 709). The efforts of the Toronto Task Force and City Council to rely on in- clusionary zoning to circumvent the problems associated with exclusio- nary zoning, and thereby create a ?space? from which to launch a cam- paign for shelters, were not only commendable but also rather ingenious, given that the Task Force was aware of the limited legal tools available to municipalities to address homelessness. And in that light, the ruling of the OMB was merely an extension of this vision. Conclusion The story of the municipal shelter by-law illustrates the point that at- tempts to implement programs to deal with homelessness are more cumbersome and daunting in practice than would appear at first glance. When compassionate approaches are promoted as the solution to a com- plex problem such as homelessness, they run into roadblocks which de- lay their implementation or lead to their demise. It is useful to examine what has transpired since the by-law was upheld by the OMB. Since this time, only one emergency shelter has opened, despite the fact that homelessness was considered to be in a state of crisis. On December 22, 2004, a temporary emergency shelter was opened at 110 Edward Street, in downtown Toronto; the shelter includes both a 80-bed co-ed and couples shelter and an Assessment and Referral Centre which operates between 8 p.m. and 8 a.m. The shelter originally operated on private property that was leased to the government and was only scheduled to operate (that is, funding was only guaranteed) till May 31, 2005, when the original lease was scheduled to expire; thereafter however, the government negotiated a month-by-month leasing option with the owner of the property so that the shelter would remain open, at least, till the end of 2005 (City of Toronto, Community Services Commit- tee, 2005, p. 3). More recently however, Council approved a proposal to purchase the land in question, in October 2005, so as to allow the shelter at 110 Edward Street continued operation (City of Toronto, 2005, p. 1). Yet, even after the purchase of the land in question, the shelter is only scheduled to be in operation until April 30, 2006; whether it will continue to operate is still uncertain. Thus even after a protracted and heated debate regarding the loca- tion of (more) homeless shelters, very little has actually materialized, and even where a new shelter has been opened, how long it will continue to be in operation is not at all clear. As well, it is important to point out here that this new shelter is located in downtown Toronto amongst other shelters in the area, and therefore, does little to spread shelters across the city as originally intended, first by the Mayor?s Task Force, and then by city council. Our case study then, leads to two important conclusions. First, it appears that homelessness?and other matters of social justice more generally?cannot be meaningfully addressed and resolved by munici- palities alone; it certainly requires the cooperation and active involve- ment of all three levels of government. Second, it seems that homeless persons bear the brunt of rather punitive sanctions from both the right and left of the political spectrum?though with respect to the latter, these effects are unintended to say the least. They are subjected to restrictions through various laws regulating their movements (for example, anti- panhandling or anti-squeegeeing laws). The many structural constraints evident in municipal politics renders the effectiveness of the left in trying to address homelessness in a compassionate way limited, so that these policies are often so diluted that they cease to be able to provide an effec- tive alternative to conservative politics. Thus, the result, though in a dif- ferent way, is the ?annihilation of spaces? of homeless people (Mitchell, 1997). Prashan Ranasinghe worked on this paper while a Ph.D. candidate at the Cen- tre of Criminology, University of Toronto. His doctoral dissertation examined the refashioning of vagrancy-type legislation and how this legal mechanism is used to (re)order publ ic spaces and interactions within these spaces He is cur- rently teaching at the University of Ottawa. Mariana Valverde is a Professor at the Centre of Criminology, University of Toronto, and is currently engaged in a socio-legal research project on urban/municipal law and bylaw enforcement. References Arnstein, S. (2003). A ladder of citizen participation. In R. T. LeGates & F. Stout (Eds.) The city reader (3rd ed., pp. 244-255). New York: Routledge. Blomley, N. (2004). Unsettling the city: Urban land and the politics of property. Lon- don: Routledge. City of Toronto (2000, May 30). Staff report from the Commissioner of Community and Neighbourhood Services to the Community Services Committee, regarding the status of the new hostel at 1673 Kingston Road: Birchmount r esidence. Toronto: Author. City of Toronto (2005, November 3). Staff report from the General Manager, Shelter and Support and Housing Administration regarding 110 Edward Street: Exten- sion of emergency shelter and referral centre programs. Toronto: Author. City of Toronto By-Law No. 138-2003 (2003). Municipal shelter by-law. Toronto: The City of Toronto. City of Toronto By-Law 438-86 (1986). The general zoning by-Law. Toronto: The City of Toronto. City of Toronto, Chief Administrative Officer?s Office (1999, February 17). Re- sponse to the Mayor?s Homelessness Action Task Force final report. Toronto: Au- thor. City of Toronto, Commissioner of Urban Planning and Development Services (1999, May 7). Process for by-law amendments to permit affordable housing, emergency shelters and rooming houses across the city. Toronto: Author. City of Toronto, Community Services Committee (2005, April 27). 110 Edward Street: Extension of emergency shelter and referral centre programs. Toronto: Au- thor. City of Toronto, Mayor?s Homelessness Action Task Force (1998, July). Breaking the cycle of homelessness: Interim report of the Mayor?s Homelessness Action Task Force. Toronto: Author. City of Toronto, Mayor?s Homelessness Action Task Force (1999, April 15). Final report: Recommendations and policy directions related to housing policies of the Official Plan. Toronto: The Commissioner of Urban Planning and Develop- ment Services. City of Toronto, Mayor?s Homelessness Action Task Force (1999, January). Tak- ing responsibility for homelessness: An action plan for Toronto. Toronto: Author. City of Toronto, Shelter Housing and Support (2002, June). Multi-year shelter strategy for the City of Toronto. Toronto: Author. Chipman, J. G. (2002). A law unto itself: How the Ontario Municipal Board has devel- oped and applied land use planning policy. Toronto: University of Toronto Press. Dear, M. J., & Wolch, J. R. (1987). Landscapes of despair: From deinstitutionalization to homelessness. Cambridge: Polity Press. Fischler, R. (1998). Toward a genealogy of planning: Zoning and the welfare state. Planning Perspectives, 13(4), 389-410. Frug, G. (1999). City making: Building communities without building walls. Princeton, New Jersey: Princeton University Press. Gerecke, K. (1976). The history of Canadian city planning. The City Magazine, 2(4), 12-23. Gillespie, K., Demara, B., & Moloney, P. (2002, September 18). Shelter bylaw fo- rums raise a tempest: Resident?s fear homeless influx into neighbourhoods. The Toronto Star, p. B1. Gillespie, K. (2003, October 14). Both sides oppose shelter bylaw: Too restrictive, tenant group says but ratepayers want more limits. The Toronto Star, p. B1. Graser, D. (2000). Panhandling for change in Canadian law. Journal of Law and Social Policy, 15, 45-91. Gunton, T. I. (1979). The ideas and policies of the Canadian planning profession, 1909-1931. In A. F. J Artibise & G. A. Stelter (Eds.) The usable past: Planning and politics in the modem Canadian city (pp. 177-195). Toronto: Macmillan. Gunton, T. (1983). The origins of Canadian urban planning. The City Magazine, 6, 27-36. Hall, P. (1989). The turbulent eighth decade: Challenges to American city plan- ning. Journal of the American Planning Association, 55(3), 275-282. Hermer, J., & Mosher J. (Eds.). (2002). Disorderly people: Law and the politics of ex- clusion in Ontario. Halifax: Fernwood Publishing. Hodge, G. (1985). The roots of Canadian planning. Journal of the American Plan- ning Association, 51(1), 8-22. Hume, C. (2005a, December 13). Are cities ready to grow up?. The Toronto Star, p. B1. Hume, C. (2005b, December 17). OMB: Reviled Board losing some of its clout. The Toronto Star, p. B5. Lakey, J. (2002a, March 12). Court sidelines shelter plan. The Toronto Star, p. B5. Lakey, J. (2002b, April 15). Shelter bylaw raises strong reaction. The Toronto Star, p. B5. Lakey, J. (2002c, April 19). Shelter bylaw pushed on to back burner: Issue must return to City Council before October. The Toronto Star, p. B3. Layton, J. (2000). Homelessness: The making and unmaking of a crisis. Toronto: Pen- guin Books. Lu, V. (2005, December 13). Mayors see hope in OMB reforms: Proposal hands power back to Council: Developers worry it?s a license to say no. The Toron- to Star, p. B2. Mitchell, D. (1997). The annihilation of space by law: The roots and implications of anti-homeless laws in the United States. Antipode, 29(3), 303-335. Ontario Municipal Board (2003, July 9). Decision No. 0923, Case No. PL030313. Ontario Municipal Board (2004, March 15). Decision No. 0569, Case No. PL030313. Perks, W. T., & Jamieson, W. (1991). Planning development in Canadian cities. In T. Bunting & Pierre F. (Eds.) Canadian cities in transition (pp. 487-518). To- ronto: Oxford University Press. Re Alcoholism Foundation of Manitoba et al. and City of Winnipeg (1990). 69 D.L.R. [4th], 697. Royce-Roll, H. (2001, July 22). Danforth residents oppose homeless shelter. The Toronto Star, p. A4. Sewell, J. (1993). The shape of the city: Toronto struggles with modem planning. Toron- to: University of Toronto Press. Smith, P. J. (1979). The principles of utility and the origins of planning legislation in Alberta, 1912-1975. In A. F. J. Artibise & G. A. Stelter (Eds.) The usable past: Planning and politics in the modern Canadian city (pp. 196-225). Toronto: Macmillan. Valverde, M., & Levi, R. (2005). Still the creatures of Provinces? Canadian cities? quest for governmental status (report submitted under the LCC/SSHRC joint initia- tive on freedom of choice. Ottawa: The Law Commission of Canada. Van Nus, W. (1979). Towards the city efficient: The theory and practice of zoning, 1919-1939. In A. F. J. Artibise & G. A. Stelter (Eds.) The usable past: Planning and politics in the modern Canadian city (pp. 226-246). Toronto: Macmillan. Verma, S. (1999, July 22). Police target squeegee kids. The Toronto Star, pp. B1, B5. Waldron, J. (2000). Homelessness and community. University of Toronto Law Jour- nal, 50(4), 371-406. Waldron, J. (1991). Homelessness and the issue of freedom. U.C.LA Law Review, 39(2), 295-324. Notes 1 These recommendations were wide-ranging and dealt with matters such as mental health issues, Aboriginal homelessness, homeless families and children; as well, the report focused not only on those who are homeless but also those at ?risk? of becoming homeless. 2 The Task Force however, was explicit in noting that the shelter system was only to be relied on as a short-term solution while long-term housing solutions (af- fordable and supportive housing programmes) were implemented. Hence, the Task Force actually called for a reduction in the number of shelter spaces by 10 percent each year until the overall number was reduced to half its base size; this however, was only to take place as long as, and only as long as, the number of affordable and supportive housing units was concomitantly increased. 3 The Ontario Municipal Board (OMB) is an independent adjudicative tribunal that hears appeals and applications from concerned parties on land-use disputes. 4 The Board ruled against the citizen in this matter arguing that because the city had already invested substantial money and time into the project, including this location as an exempted site made sense, because it ensured that if another shel- ter was to be located within 250 meters of the property in question prior to the property in question being turned into a shelter, the city would not lose the time and money it had already invested (OMB Decision No. 0569, p. 29-30). 5 In an interesting twist, the Board (correctly) noted that it had no jurisdiction to rule on whether a particular by-law meets the test of constitutionality; however, the Board then went on to spend considerable time arguing that the require- ments of the by-law did not violate the provisions of section 15 of the Charter (OMB Decision No. 0569, p. 34-52). 6 In yet another interesting twist, the CRRA, at the outset of the hearings, had given notice of its withdrawal from the proceedings because it could not muster sufficient resources to allow for full attendance and/or participation in the hear- ings (OMB Decision No. 0569, , p. 3). Chapter 1.5 Toronto?s Streets to Homes Program NICK F ALV O This chapter evaluates the effectiveness of the Housing First model of providing permanent housing to long-term or chronically homeless sin- gles, of which Toronto?s Streets to Homes (S2H) program is arguably the most popular model today. The chapter begins by examining the ?treat- ment first? approach to housing homeless persons, as well as the emer- gence of the Housing First model, followed by a case study of Toronto?s Streets to Homes program. The program?s origin, successes, and short- comings are discussed and recommendations on how to improve the program are offered. While the general view of interview subjects ? all of whom have been assured of anonymity ? is that S2H has been effective, most believe there is room for improvement. Method Toronto, Ontario, was chosen as the study area for this research because (a) its homeless population is larger than that of any other Canadian municipality and (b) its Housing First model is by far the largest and most developed example of the approach of any Canadian municipality. Moreover, the study area is well known to the researcher as he has worked as a front-line community worker with Toronto?s homeless pop- ulation for more than a decade. Semi-structured in-depth interviews were undertaken with 34 key informants from March 2008 until October 2008.1 All key informants were selected based on the researcher?s previous knowledge gained both as a front-line community worker in Toronto for the past decade and as a researcher over the past six years. While all of the above interviews informed the policy recommendations suggested by the research, not all interview correspondence is cited here. There were seven groups of informants. The first group consisted of four City of Toronto officials familiar with the S2H program. These interviews involved ques- tions about the province?s role in Toronto?s emergency shelter system, S2H?s main components and operations, and S2H?s Street Outreach Steering Committee. The second group consisted of two individuals, one from the United States and one from Canada, who were asked for infor- mation on academic resources on the Housing First model. The third group, consisting of five individuals ? four in the academic community and one in the activist community ? were asked about criticism of the Housing First model. The fourth group consisted of six policy experts in Toronto who were asked about the pre-S2H environment in Toronto, specifically, what efforts had been made in Toronto prior to S2H to pro- vide permanent housing to homeless persons. Members of a fifth group, consisting of six experts on poverty and health, were asked about the effects of low income on health ? particularly disposable income after shelter costs. A sixth group, consisting of six executive directors of To- ronto community agencies, was asked about the shortcomings of S2H. In particular, these executive directors were asked to what extent they felt that S2H was not meeting its program goals. Finally, a seventh group, consisting of three Canadian experts on affordable housing policy, was asked to what extent a Housing First program such as S2H can function in a context of relatively low vacancy rates. Time constraints precluded client interviews. However, the research drew on S2H?s 2007 post-occupancy research study, which involved in- 1 The research focuses on S2H before the enhancements that were implemented in May 2008. Thus, it will not explore the recent decision by Toronto City Council to use the S2H approach to address panhandling in Toronto, though this en- hancement will be touched on in the paper?s conclusion. terviews with 88 S2H clients. Data from the program?s post-occupancy research are the only data available on S2H clients and therefore have to be considered in any assessment of the program. There were, however, clear limitations to the data, and they should be interpreted with caution. First, the preoccupancy data used were taken at the same time as the post-occupancy data. Indeed, tenants were asked at the time of the sur- vey how their situation compared in many regards before and after te- nancy, but they were asked this retrospectively. This raises the research methodology question of reliability. Second, many of the outcomes were self-reported rather than externally verified. Third, the survey was done ?in-house? by City of Toronto staff, raising a methodological question of bias. (For more on epidemiological research methods, see Galea and Vla- hov, 2005). U.S. Models for Providing Housing to the Homeless The standard model of providing housing to chronically homeless adults in the United States is the ?treatment first? approach, also known as the ?continuum of care? model. In this model, a provider ? or team of pro- viders ? of homeless services determines when and if a homeless person is ready to be housed. The assessment process continues as the partici- pant progresses from emergency shelter, then graduates to transitional housing and then moves on to the final stage of the continuum: perma- nent housing with few if any supports. To complete the process, a home- less person must generally abstain from drugs and alcohol, and may be required to take psychotropic medication as prescribed by a physician. In short, the ?treatment first? approach requires one long ?audition? of sorts. Non-compliance with any of the conditions can result in either a delay in the transition or expulsion altogether (Greenwood et al., 2005; Tsemberis and Eisenberg, 2000). The goal is to see that the client is ?housing ready,? and the continuum is seen as one lengthy preparation process for independent living. For many homeless people ? the chronically homeless in particular ? the conditions involved in this process are onerous, if not completely unrealistic. Moreover, it is highly debatable as to whether the conditions required in such a process represent a good test for housing readiness. The new model of providing housing to the chronically homeless is the ?Housing First?2 The program has only two requirements of its participants: approach. Housing First does not require homeless people to go through the steps described above. Instead, it provides them with almost immediate access to permanent housing. Though staff periodically visit the participants at their units, the housing does not fea- ture 24-hour, on-site staffing (Padgett, et al., 2006: 75). The model is often believed to have developed first in New York City in 1992 with the founding of a non-profit agency called Pathways to Housing Inc., led by Dr. Sam Tsemberis, a clinical psychologist (Padgett, 2007: 1928). All of the Pathways participants are initially homeless and have a psychiatric diagnosis. Almost all also have problems with drugs and/or alcohol (McCarroll, 2002). Furthermore, the program will not refuse a client with a history of violence and/or incarceration (Padgett et al., 2006: 77). 1. They must agree to participate in a money management program with staff that takes 30 percent of their income and directs it toward rent (Greenwood et al., 2005: 225). The other 70 percent of each par- ticipant?s rent comes from grants from city, state and federal gov- ernments, as well as from Section 8 vouchers (Tsemberis and Eisen- berg, 2000: 489).3 1.2. They must agree to at least two visits to their apartment by staff per month. The client has access to an Assertive Community Treatment (ACT) team. The ACT team in question provides multidisciplinary clinical sup- port; its staff are led by a psychiatrist and include a social worker, a ?vo- cational trainer,? an addictions worker, a nurse practitioner and a hous- ing worker. The team is available to clients 24 hours a day, seven days a week (Greenwood et al., 2005: 225; Padgett et al., 2006: 77). While absti- nence is neither a program requirement nor an expectation, Pathways staff provide support from a ?harm reduction? perspective. Counselling 2 This should not be confused with the City of Toronto?s ?Housing First Policy? whereby surplus municipal land has to be used for housing. 3 With reference to these two requirements, Gulcur et al. (2003: 174) note: ?These criteria are... applied flexibly such that prospective clients are not denied housing on the basis of their refusal to comply.? Formatted: Bullets and Numbering on substance use is provided; Pathways even has its own harm reduction support groups. Clients who wish to enrol in residential treatment pro- grams are assisted by Pathways staff in doing so. Moreover, if the client chooses this option, a Pathways apartment unit is guaranteed upon her or his return from treatment (Padgett et al., 2006: 77). In comparing the two models, the academic literature on Housing First is overwhelmingly positive. It indicates that between 85 percent and 90 percent of those who participate in the Pathways program are still housed when followed up five years later (Tsemberis and Eisenberg, 2000). Also, compared with their ?treatment first? counterparts, Housing First participants remain housed longer, spend fewer days in hospital (Gulcur et al., 2003: 181), and are no more likely to use drugs or alcohol (Padgett et al., 2006: 74). Finally, it is cheaper to support a client through the Housing First model than through the ?treatment first? approach, due largely to the reduced days required for psychiatric hospitalization (Gulcur et al., 2003: 182).4 Although the ?treatment first? approach remains the dominant ser- vice delivery model in the United States (Padgett et al., 2006: 81), by 1996, Housing First programs had helped more than 100,000 participants (McCarroll, 2002). The Housing First approach is increasingly popular among policy-makers, politicians, business leaders, and the media. Toronto Context Toronto and most other Canadian cities have a smaller proportion of economically marginalized people than do most American cities (see Myles, 1996). Moreover, social housing provision has been considerably more significant in Canada than in the United States. In the 1965-1995 period, social housing (including both public housing and Section 8 housing) accounted for roughly 3 percent of housing stock in the United States. In Canada, the corresponding figure was 6 percent (but is now down to 5 percent). In 1982, there were an estimated 3,440 homeless persons in Metro- politan Toronto, of whom 1,600 were in hostels and another 1,800 had no 4 It is considerably cheaper to provide individuals with government-assisted housing (supportive or not) than to supply them with a shelter bed every night. fixed address (Metropolitan Toronto, 1983: ii).?5 Average daily hostel occupancy [in Toronto] increased overall for single adults by 63 percent from September 1992 to September 1998. In the same six-year period, the increase in shelter use by population groups was 80 percent for youth, 78 percent for single women, 55 percent for single men, and a shocking 123 percent for families (Golden et al., 1999: 14). By 1983, individuals under 25 years old, families and single women represented subgroups on the rise within Toronto?s homeless population (Metropolitan Toronto, 1983: vii). By 1988, roughly 20,000 people were using Toronto?s emergen- cy shelter system on an annual basis. Of those people, roughly 4,000 were single women, 6,000 came to the shelters in families and 10,000 were single men (Ontario, 1988: 36). As pointed out in 1999 in the final report of the Mayor?s Homelessness Action Task Force: In 1990, a total of 26,529 individuals used a Toronto emergency shel- ter at least once during the year. By 2002, this figure had risen 21 percent to roughly 31,985 (City of Toronto, 2003: 38). Thirty-one percent of formerly homeless people recently surveyed stated that, prior to being housed, they never stayed in shelters, 11 usual- ly electing to sleep outside. Another 40 percent said that they did so ?rarely? (City of Toronto, 2007: 79).6 Toronto?s homeless population has a smaller proportion of visible minorities than its general population, in contrast to the United States homeless population. Indeed, whereas 37 percent of homeless respon- dents in the 2007 Street Health survey identified themselves as ?non- Caucasian,? 44 percent of Toronto?s general population consisted of visi- ble minorities. However, the same survey also found that 15 percent of homeless people in Toronto identified themselves as Aboriginal, com- 5 ?The figure of 3,440 persons is still a minimum estimate of the number of home- less in Metropolitan Toronto, as anyone who was not a client of the agencies sur- veyed or who did not stay at a hostel was excluded? (City of Toronto, 1983: 7). 6 Respondents who said ?rarely? meant that ?they stayed in shelter less than a few days each month, or ?only when I had to ,? or ?only when it was very cold?? (City of Toronto, 2007: 12). Of those who ?never? used shelters, slightly more than half never even used Out of the Cold beds, while just under half did use Out of the Cold Beds (City of Toronto, 2007: 79). pared with 0.5 percent in the general population of Toronto (Khandor and Mason, 2007: 7-8). Thus, as is the case in the United States, Aborigin- al individuals are overrepresented in Toronto?s homeless population ? in fact, considerably more so than in the United States. The City of Toronto undertook a needs assessment of all those it identified as homeless on the night of April 19, 2006. The survey sample was ?representative of the demographic composition of homeless people encountered outdoors during the Street Needs Assessment in April 2006? (City of Toronto, 2007: 8-11). The findings suggest that homeless persons sleeping outside are, by far, the most ?chronically homeless? of all the groups surveyed. Table 1 outlines this situation clearly, showing that those sleeping outside on the night of the assessment reported hav- ing been homeless an average of six years. Table 1 : Length of Homelessness Location Average Number of Years Homeless Outdoor 6.0 Family Shelters 0.6 Youth Shelters 1.2 Mixed Adult Shelters 3.8 Men?s Shelters 4.1 Women?s Shelters 2.1 All Shelters 3.0 Corrections 4.5 Health and Treatment 4.2 All Survey Respondents 3.4 Source: City of Toronto (2006: 14). Furthermore, the outdoor homeless population is more inclined to have used a detox than those sleeping in shelters (23.5 percent vs. 16.0 percent) and less inclined to have participated in employment/job train- ing (17.5 percent vs. 27.0 percent [City of Toronto , 2006: 15]). In summary, Toronto?s homeless population has increased by 400 percent between 1980 and 2000. Within the homeless population, the numbers of couples, children and single-parent households have grown the fastest. Toronto?s current homeless population experiences chronic physical health conditions, as well as mental health conditions, at much higher rates than the general population. Moreover, Aboriginal persons are very much overrepresented in Toronto?s homeless population. Final- ly, those living outside have been homeless considerably longer than those living in shelters. Toronto?s Policy Responses Funding for homelessness relief programs in Toronto comes from all three levels of government, as well as the charitable sector, whose main players are the United Way of Greater Toronto, the Trillium Foundation, and churches (Dowling, 1998: 12). Some types of support serve many homeless people but are geared to a wider group that includes housed individuals. For example, the Ontario Ministry of Health funds mental health case management through agencies such as COTA Health and Street Health. It also funds drop-ins such as the Parkdale Activity- Recreation Centre (PARC), the Meeting Place (run by St. Christopher House) and Sistering (Dowling, 1998: 1-7). Toronto has had a municipally managed shelter system from the 1960s onward. In the 1980s, it expanded significantly. By 1988, Toronto had roughly 2,100 shelter beds open each night, generally at or near ca- pacity. That figure grew steadily and was roughly 3,500 by 1996 (Sprin- ger et al., 1998: 9). Metro Toronto?s budget for ?services to the homeless? grew from $38 million in 1992 to $56 million by 1997 (Main , 1997: 23). That said, the expansion of Toronto?s shelter system was not as dramatic as that experienced in American cities. While the number of emergency shelter beds in the United States grew sixfold between 1984 and 1996, Toronto?s capacity doubled. Government-assisted housing, both in the United States and in Canada, traditionally was not directed primarily at those who were ?homeless.? Before 1986, homeless people in Canada were ineligible for social housing unless they were diagnosed with a disability (Dowling, 1998: 2-3). Beginning in the 1980s, a sizeable percentage of government- assisted housing units began to be directed at the homeless. In Ontario, eligibility for government-assisted (i.e. rent-geared-to income [RGI]) housing was originally for low-income families with children and low- income seniors. ?Supportive housing? units were introduced in the 1980s as a provincial program, largely as a delayed response to the deinstitu- tionalization of individuals with mental health issues. Many of the reci- pients were homeless when they received the housing. From the mid- 1980s until the mid 1990s, roughly 300 new supportive housing units per year were made available to homeless singles in Toronto, mostly from the shelter system. Roughly 10 percent of the 100,000 social housing units in Toronto are supportive housing units. And in the mid-1980s, the Habi- tat boarding homes (jointly funded by the Province and the City on an 80:20 basis) began operations. The Toronto-based Homes First Society was especially innovative in pushing the envelope on providing housing to the homeless (both the sheltered homeless and rough sleepers) in the 1980s. In 1984, it opened its 90 Shuter Street complex, Toronto?s first government-assisted housing dedicated to homeless single people (Dowling, 1998: 2-3). In 1994, homeless people became designated as a priority target population for new vacancies arising in all non-profit housing units in Ontario (Dowling 1998: 3). In 2006, this meant that 825 homeless persons obtained housing in Toronto Community Housing Corporation (TCHC) units (Housing Connections, 2006: 13). For 2007, the figure was 941 (Housing Connections, 2007: 11). In 2000, in response to nationwide advocacy, the federal govern- ment introduced the Supporting Communities Partnership Initiative (SCPI),7 Until recently, no level of government made a concerted effort to move rough sleepers (i.e. those living outside the shelter system most nights) directly into permanent housing. A major reason was a bureau- cratic one: community agencies liked working with non-profit housing providing $135 million per year across Canada for homeles sness services and support programs. In spite of the federal government?s in- sistence that this was funding not be used for permanent housing, some communities succeeded in creating long-term ?transitional housing? units for homeless persons. There are now roughly 2,500 such units na- tionally, roughly 750 of which are in Toronto. 7 In December 2006, the Harper government modified the SCPI program and renamed it the Homelessness Partnering Initiative (HPI). As of October 2008, the HPI was extended beyond March 2009, but details on this extension are not yet clear. providers, largely because non-profit landlords charged rents that were geared to a tenant?s income. There were always waiting lists for govern- ment-assisted housing. Establishing a connection with a rough sleeper was hard enough. But completing an application with one, and then lo- cating the person months or years later after his or her application had made its way to the top of the waiting list was nearly impossible. However, some agencies did help rough sleepers move directly into non-profit housing. For example, at the Corner Drop-In, run by St. Ste- phen?s Community House, outreach workers helped some rough slee- pers move directly into rooming houses. Moreover, as part of a pilot project in the late 1990s, staff at PARC, Community Resource Connec- tions of Toronto and Sistering all helped rough sleepers access perma- nent housing at Houselink Community Homes (whose mandate was to house people with serious mental health problems). As for the model used, some Toronto housing providers followed the ?treatment first? approach, but many did not. For example, neither Houselink Community Homes nor Mainstay Housing insisted on medi- cation compliance for tenants who had serious mental health problems. Nor did Houselink or Mainstay require that a tenant with addictions is- sues complete an abstinence-based treatment program before receiving the keys to a housing unit. Indeed, the harm reduction approach, which does not require abstinence, has been ?commonly followed in supportive housing in Toronto? for many years. One of the Ontario government?s responses to the 1999 final report of the Mayor?s Homelessness Action Task Force was to initiate a Toronto program called Off the Streets Into Shelters, a program that featured four or five outreach workers who encouraged rough sleepers to enter emer- gency shelters. Moreover, the 1999-2000 period saw a major expansion in homeless services in Toronto, in part due to the provincial government?s response to the final report of the Mayor?s Homelessness Action Task Force and in part due to the advent of the SCPI. Increased services from both of these initiatives came in the form of a rent bank, eviction preven- tion programs, more housing of workers in shelters and the province?s Off the Streets Into Shelters street outreach program. This period also saw an increase in the number of all-day shelters and the revamping of Seaton House, Toronto?s largest men?s shelter. In 2000, the provincial government expanded its supportive hous- ing system. Over the next five years, the number of supportive housing units in the City of Toronto rose from 2,400 to 4,200 (including the ex- pansion of Habitat boarding homes, whose stock grew from 600 to 1,000 during this time). Also in 2000, the Ontario Ministry of Municipal Affairs and Housing started a rent supplement program that, by 2005, had re- sulted in 3,000 rent supplements. According to a 2003 City of Toronto report, Toronto?s approach to serving rough sleepers changed in 2001: Previous to 2001 the majority of street outreach funding was for programs that provided survival support. While the survival work continues, since 2001 the main focus of street outreach has shifted to ?high support street outreach.? This approach uses a case management approach where out- reach workers do comprehensive work with people to help them get off the street and into shelter, housing or other suitable programs and servic- es... In many situations workers were successful in helping someone find shelter or housing (City of Toronto, 2003: 49-50). Though the above shift in Toronto?s approach to rough sleepers by no means meant a full shift to a Housing First approach, it did represent the continuation of the aforementioned shift to supportive housing. In 2002, with funding from a City of Toronto grant program, the Fred Victor Centre began running a very effective program moving ?long-term homeless persons? (i.e. people who had been homeless for over a year) from shelter into permanent housing, and then providing follow-up services. But since then, the City of Toronto stopped encourag- ing community agencies to develop new programs. Indeed, that was the last year the City put out a request for proposals (RFP) to community agencies to come up with new service delivery models. Through many of these efforts, roughly 6,500 homeless persons per year were being moved from Toronto?s emergency shelter system into permanent housing. This is not a well-known fact, but it ought to be. To be sure, and contrary to the general perception, the City of Toronto?s shelter system and its many services has been very effective at moving its clients into permanent housing. Unfortunately, funding has been tight. For example, annual funding for emergency shelters not directly run by the City of Toronto has en- dured several years of flat-lined budgets. From the late 1990s until 2003, for instance, the per diems (e.g. the amount of money provided per filled shelter bed on a nightly basis) to non-City shelters did not increase (not even adjustments for inflation). And City ?grants programs,? which fund some drop-ins, help centres, food programs and housing support pro- grams, have received virtually no funding increases since 2000. And in spite of the Province?s expansion of program initiatives, it has been shortchanging the City of Toronto with respect to the funding of shelter beds. Under the Ontario Works Act, the Province is supposed to pay 80 percent of the cost of each shelter bed in Toronto?s emergency shelter system. The City is supposed to pay the remaining 20 percent. But the Province has capped the total dollar amount it will pay for each bed for each Ontario municipality. Assuming the City pays the addition- al 20 percent, this would bring the total ?per diem? per shelter bed to a total of $42. While $42 per night per occupied shelter bed might be suff i- cient to run a shelter in a small Ontario municipality, it is grossly inade- quate for Toronto. The actual cost involved in running an occupied shel- ter bed in Toronto is more like $57. Thus , in addition to paying the initial 20 percent required under the Ontario Works Act, the City of Toronto has been paying 100 percent of the difference between the actual cost of an occupied shelter bed and the Province?s capped amount. Thus, the City of Toronto is now the majority funder of shelter beds in Toronto, paying 52 percent of the actual costs versus the Province?s 48 percent. For the Province to honour the 80:20 split for Toronto alone, it would have to start paying an additional $20 to $30 million ann ually. Not surprisingly, Toronto?s current shelter system is far from ade- quate. The 2007 Street Health Report found that 55 percent of all home- less people surveyed reported that they were unable to get a shelter bed at least once in the previous year ? 20 times on average. The Streets to Homes Program (S2H) Toronto?s Streets to Homes (S2H) program originated in February 2005 with an annual budget of $4 mi llion. Prior to the May 2008 enhancement, the program?s annual budget stood at roughly $8.7 mi llion. The program emerged out of a unique context. First, in 2003- 2004, Toronto City Coun- cil had a series of debates on homelessness, during which time concern was raised about the fact that large sums of money were being spent on homelessness, yet the number of homeless people was continuing to grow. Second, almost 100 people a night had been sleeping rough at Na- than Phillips Square (Toronto City Hall). Third, beginning in 2002, the City of Toronto had undertaken a successful relocation of the Tent City squatters, whereby roughly 100 squatters had been given immediate access to private market housing, a deep rent supplement and staff sup- port (see Gallant, Brown and Tremblay, 2004). Finally, in 2004, roughly 20 to 30 people had been evicted from underneath the Bathurst Street Bridge when a nearby building was being demolished. There was a great deal of media coverage of this event. Several squatters interviewed by the media said that they had not been offered housing when they were evicted (Falvo, 2008: 33). S2H?s goal is to ?end street homelessness.? The program?s official mandate is to ?serve homeless people who live outdoors, which includes individuals living in parks, ravines, under bridges, on sidewalks, lane- ways, alleys, stairwells, building alcoves, squats and living in vehicles? (City of Toronto, 2007: 61). The program finds permanent housing for these people. For the program?s first 18 months of operation, staff planned to work only with clients who were believed to have stayed outside for at least seven consecutive nights. These narrow criteria proved difficult to establish. Now, S2H staff8 Like Housing First, S2H strives to provide homeless people with immediate access to housing. Abstinence from drugs or alcohol is not a prerequisite, nor is compliance with psychiatric medication. Nor does a participant have to prove to be ?housing ready? (Falvo, 2008: 33). work with clients who appear to be spending most nights outside and are not already receiving the services of a hous- ing worker (Falvo, 2008: 33). There are seven steps involved in a rough sleeper?s acquiring hous- ing through S2H. These are outlined in Table 2. 8 The term ?S2H staff? is used broadly here to include staff directly employed by the City of Toronto and staff employed by S2H-funded partner agencies. Table 2 : Steps Involved in an Individual?s Acquiring Housing through S2H Step 1 S2H staff approach the rough sleeper and attempt to have a discussion about housing, explaining to the client that provision of permanent hous- ing is the program?s prime focus. Other matters important to the client?s well-being (i.e. health care, ID replacement, social support, etc.) can be taken care of afterwards. Step 2 If the client shows interest, an intake assessment is done, during which time the client is asked about basic demographic characteristics, how long he or she has been homeless, the last time he or she was housed, how he or she can be contacted, what kind of income support ? if any ? he or she is currently receiving, the part of the city where he or she wishes to be housed and the type of building he or she wishes prefers. The client is also told how to contact S2H staff. Step 3 S2H staff develop housing options for the client. Step 4 S2H staff help the client take care of issues such as income support arrangements and outstanding work orders on the housing unit. Step 5 S2H staff accompany the client to see housing units. Step 6 Once an appropriate housing unit is found that the client likes, the lease is signed. Step 7 Finally, a joint meeting takes place involving the client, the ?street out- reach counsellor? who has been working with the client thus far and the new ?follow-up support worker? who will be providing follow-up support to the client. Source: Key informant interview with source close to the S2H program. The above process happens very quickly. From the third contact with the client, it takes an average of only 16 days for that client to re- ceive keys to the unit. When the S2H program acquires a new housing unit, there is a two-step process involved. First, if the unit has outstand- ing work orders identified in the Multiple Listing Service (MLS), staff immediately reject the unit. If there are no outstanding work orders iden- tified by the MLS, staff assess the unit themselves, checking electricity, heating systems, and safety. Although a client can move into a unit that still has outstanding work orders identified during this process, S2H staff advocate with the landlord to address them as soon as possible. If a problem develops after the client has been moved into his or her unit, S2H staff help the person move (Falvo, 2008: 33). Thirty-two percent of those interviewed in the program?s post- occupancy survey reported having moved at least once since being housed.9 The four components of the S2H program are outlined in Table 3. In fact, the rate is 50 percent for those who have been housed for longer than a year (City of Toronto, 2007: 33-34). The reasons for moves vary. Often the move oc- curs because a person has been initially housed in a non-subsidized unit, and then a (subsidized) TCHC unit has become available. Other times, it happens because S2H clients have become reunited with ? and regained full custody of ? their children after being housed. Other times, S2H clients obtain a job after being housed and then have to relocate to be closer to the job site. Other times, the client may not be getting along with the landlord. Or, the client changes his or her mind about the loca- tion. Still other times, the landlord wants to change the initially agreed- upon arrangement or is being difficult in other ways. Table 3 : Streets to Homes Program Components Elizabeth Street Component The Elizabeth Street component works primarily out of 112 Elizabeth Street. It consists of over 20 full-time staff, including over a dozen front-line workers, six management and administrative staff, two full-time research analysts and an in-house lawyer. This office serves as the central administration and coordination of the program. Most of the landlord recruitment, for example, happens out of this office. Funded Partner Agencies S2H funds 29 non-profit partner agencies to assist in the delivery of its services.10 MDOT ? This program is run by Toronto North Support Services, in partnership with St. Michael?s Hospital, the Centre for Addiction and Mental Health and the Fred Victor Centre. It features a multidisciplinary support team whose goal is to work with clients with ?the most complex needs,? usually of a mental health nature. Many have had previously existing programs ?realigned? in order to better meet S2H objectives. Programs run by funded partner agencies include, but are not limited to: Post-Incarceration Housing ? Run by both the John Howard Society of Toronto and the Elizabeth Fry Society of Toronto, this program provides post-incarceration housing and support services 9 The post-occupancy survey being referred to in this paper interviewed 88 S2H clients between November 2006 and April 2007. The results of the survey can be found online at www.toronto.ca/housing/pdf/results07postocc.pdf. 10 A full list of all partner agencies can be found at www.toronto.ca/housing/about-streets-homes-partners.htm. to people who have been street homeless and then become incarcerated. Housing assessments take place while the individual is incarcerated. This service is provided on a limited basis in all Toronto-area detention centres. Rapid-Access Housing ? Ten ?rapid access housing? units provided by the Fred Victor Centre are the only transitional housing units offered by S2H. Intended for clients with serious substance use problems, participants in this stream of the program receive at least three months of intensive case management. Psycho-Vocational Assessments ? In partnership with Toronto Social Services, JVS Toronto conducts psycho-vocational assessments with roughly 75 S2H participants per year. Sometimes these result in identifying disabilities that result in successful ODSP applications. Other times, they result in the identification of literacy issues. Non-Funded Partner Agencies Eight partner agencies do not receive S2H funding but have signed formal service agreements. One such partner agency is the Toronto Community Housing Corporation. Volunteer Component11 Volunteers (often from the faith community, many of whom used to volunteer with the Out of the Cold 12 Source: Key informant interviews with source close to the program (I. 14 and ). program) provide ?non- professional? assistance to both S2H and non-S2H clients, by engaging in community development. This includes such things as hosting bingo nights and spaghetti dinners. No formal service agreements are signed for this component of the program. S2H clients are housed in three types of housing. Sixty-two percent are in privately owned units, which include small and large residential units, secondary suites, privately owned rooming houses and entire houses (shared). Only one-quarter of the 62 percent of S2H clients in pri- vately owned units receive a shelter allowance from an external funding program. This arrangement takes place through the Housing Allowance Program (HAP) and offers a shelter allowance of $350 per month per participant, for a total of five years. HAP participants represent 15 per- cent of all S2H clients. 11 This is a very small component of S2H. One person close to S2H interviewed for this paper had never even heard of the volunteer component of the program. 12 ?Out of the Cold is a faith-based volunteer program which provides meals and shelter at locations throughout the city during winter months? (City of Toronto, 2007: 12). Another 20 percent of S2H clients are in social housing units owned and operated by a non-profit agency, which charge a rent calculated in line with 30 percent of a tenant?s income (City of Toronto, 2007: 48). Finally, 18 percent of S2H clients are in alternative or supportive housing units, meaning that the housing in question is owned and oper- ated by a non-profit organization such as Ecuhome, CRC Self-Help, the Fred Victor Centre or St. Clare?s Multifaith Housing (City of Toronto, 2007: 76). Alternative and supportive units usually have ?some form of on-site staff support and were often rent-geared-to-income units (City of Toronto, 2007: 13).? Some providers charge rent calculated at 30 percent of the tenant?s income. Others charge rent equivalent to the shelter por- tion of each tenant?s social assistance cheque ($325 in the case of O ntario Works and $436 in the case of the ODSP) City of T oronto, 2007: 48). Of the clients interviewed for the post-occupancy survey, 61 percent were living in independent housing. This includes a single person living in a bachelor apartment (30 percent) or a single in a one-bedroom apart- ment (24 percent) or a couple/family living in a two- bedroom apartment (8 percent) (City of Toronto, 2007: 82). The other 39 percent live in shared accommodation, which, in the context of S2H, includes individuals shar- ing a two- or three-bedroom private market apartment with non-related roommates (8 percent), shared accommodations in alterna- tive/supportive housing (generally individual rooms with shared com- mon areas such as kitchens and washrooms) (26 percent), or a rooming house (5 percent) (City of Toronto, 2007: 14). When the program began, most S2H participants doubled up with a roommate, due largely to a lack of program funding. S2H clients are often given ?housing incentives? of various types, especially in the first three months of tenancy. These include gift certifi- cates from various grocery stores and retail outlets, which are especially helpful to clients who are ineligible for a community start-up allowance or those in deep arrears with a landlord. Once a client has been given housing, follow-up support is offered by S2H staff, for up to one year. This includes informal counselling, as well as help with Ontario Works or the ODSP, finding furniture, con- necting to resources in the community, dealing with the landlord, gro- cery shopping, transportation, accessing health services, and acquiring clothing (City of Toronto, 2007: 84). According to the program?s post- occupancy follow-up survey: Follow-up supports are for approximately a one year period, and through intensive goal setting the frequency of visits decreases over time. At the end of the year, the individual is expected to be able to live independently without ongoing support or are [sic] transitioned to more appropriate o n- going case management services (City of Toronto, 2007: 62). That said, S2H staff sometimes do make exceptions and continue providing support to clients after 12 months. S2H is run directly by the City. Relative to most programs for the homeless run by community agencies, it serves a large number of clients and has a large budget. This gives it clout, and it has used this to its ad- vantage by creating special arrangements with key actors (Falvo, 2008: 33). Some examples follow. ODSP ? The Ontario Disability Support Program processes ODSP applications by S2H clients are processed remarkably quickly. Whereas an ODSP application would normally take 6 to 12 months to be approved, in 2006 S2H clients began having their applications approved in as little as 48 hours, helping them to increase their monthly income much more quickly than non-S2H clients (Falvo, 2008: 33).13 As stated in the City?s post-occupancy survey of S2H clients: Income assistance programs now offer fast-tracked access to benefits (usually on the same day), are willing to maximize discretion when is- suing benefits, have meeting space within their offices for housing work- ers, and now send income assistance staff to Streets to Homes offices once per month (City of Toronto, 2007: 63). 13 Only 31 percent of S2H survey respondents reported being on ODSP. Another 64 percent reported receiving Ontario Works (City of Toronto, 2007: 89). With Ontario Works (i.e. welfare), intakes for S2H clients can now be arranged within 24 hours. Moreover, S2H clients receive faster approval and receive more discre- tionary benefits than non-S2H clients. Toronto Community Housing Toronto Community Housing is an arm?s-length, non-profit corporation accountable to ? and owned by ? the City of Toronto. It has made a few hundred of its subsidized housing units available to S2H clients without requiring that they spend the typical multi-year stint on its waiting list. In other words, some S2H clients have bypassed the social housing wait- ing list. The only units offered via this arrangement are ones that have already been turned down by at least three Toronto Community Hous- ing applicants (or by current tenants seeking a transfer). Private Landlords Several large, private landlords give special concessions to the S2H pro- gram.14 In addition to making some units available to the program, they may reduce the rent by modest amounts. (In exchange, the landlord knows that S2H staff do follow-up with the tenant, ensure that tenants initially agree to a pay-direct arrangement for their rent [Falvo , 2008: 33] and even have special S2H program money to fund some maintenance costs for the unit.)15 Non-Profit Housing Providers Several non-profit housing providers ? including Mainstay Housing, Ecuhome Corporation, Homes First Society and the Fred Victor Centre ? allow S2H clients to bypass their waiting lists and then offer them high levels of support once housed. In exchange, the S2H program gives them funding over and above what the tenant pays them for rent (Falvo, 2008: 33). The non-profit housing providers apply for this via an RFP process. 14 Relationships with many of these landlords were developed in the years prior to S2H though some of the other efforts that resulted in 6,500 homeless persons per year being moved into permanent housing. Some of the shelter staff who had developed these relationships worked for S2H in the early days of the program and ?brought their contacts with them.? 15 As one policy expert put it: ?With incentives such as these, the perceived unde- sirable tenant all of a sudden becomes a desirable one.? Newly Built Housing S2H clients get priority access to 30 yet-to-be-completed housing units created through the City of Toronto?s Affordable Housing Office. S2H Successes Roughly 600 people have been housed each year through the program since February 2005, and 87 percent of the tenants it has housed remain housed. Of the 13 percent of clients who are not still housed, 2 to 3 per- cent have since died and another 2 to 3 percent have moved to another city. In 2007, contacts were made with almost 3,900 potential clients.16 Judging from results of the program?s post-occupancy survey, S2H appears to be doing a very good job of reaching its target group, namely rough sleepers. According to results of the survey, 31 percent of the people S2H housed had ?never? used the shelter system prior to being housed through S2H, and another 40 percent of them had ?rarely? used the shelter system (City of Toronto, 2007: 79). 17 representative of the demographic composition of homeless people en- countered outdoors during the Street Needs Assessment in April 2006. This indicates that the clients being housed through Streets to Homes are reflective of the composition of the outdoor homeless population (City of Toronto, 2007: 8). Furthermore, according to a City of Toronto report, the sample of S2H clients interviewed in their post-occupancy survey was: Post-occupancy survey results also show that, once housed, most S2H clients report improvements in their health, the amount of food they are eating, the quality of food they are eating, their stress levels, their sleep, their personal safety and their mental health (City of Toronto, 16 The exact number was 3,896. This is the total number of people that S2H ?en- gaged.? Not all of these people met S2H?s criteria. However, one well-placed source told the author that this figure ?is a roll-up of 10 organizations, and there- fore contains a lot of duplication.? 17 Only 29 percent of respondents stated that, prior to being housed with S2H, they stayed in the shelter system ?more often,? meaning that ?they stayed for several nights a week, or would stay for several months at a time off and on? (City of Toronto, 2007: 12, 79). 2007: 86-88). Roughly half of all S2H clients report reduced drinking, and roughly three-quarters report reduced drug use (City of Toronto, 2007: 86-88). In fact, 17 percent of respondents reported quitting drinking alto- gether (City of Toronto, 2007: 44), and one-third reported quitting drugs altogether (City of Toronto, 2007: 88). S2H clients, once housed, reported making fewer calls to 911, get- ting arrested less often, spending less time in jail (City of Toronto, 2007: 89-91) and less use of hospital emergency rooms (City of Toronto, 2007: 50). For S2H clients who continued to use the above emergency services, they used them much less often (City of Toronto, 2007: 51). Once housed through S2H, the number of people reporting income from panhandling dropped by 57 percent (City of Toronto, 2007: 49). S2H clients, once housed, also reported increased use of family doctors, optometrists, and specialists (City of Toronto, 2007: 50). Of those housed by S2H, roughly 60 percent more are now receiving ODSP benefits than before (City of Toronto, 2007: 46). City officials claim that the overall numbers of homeless people in Toronto have decreased since the onset of S2H, and point to the fact that they have been able to close several shelters in the last year.18 S2H Shortcomings Unlike New York City?s Pathways program, there is no stipulation with S2H that participants pay no more than 30 percent of their income on rent.19 18 This information was provided to Toronto City Council on May 26, 2008, by Phil Brown, General Manager of Shelter, Support and Housing Administration. He also stated that the shelter closures had no serious impact on occupancy le- vels of the remaining shelters. In fact, S2H participants pay an average of 41 percent of their in- come on rent. Some S2H clients receiving ODSP benefits have as much as $600 per month to live on once rent is paid. But most have considerably 19 New York?s program is by no means the only Housing First program with strong affordability stipulations. Calgary?s Housing First program has an iden- tical stipulation: no participant pays more than 30 percent of his or her income on rent. Likewise, Ottawa?s Housing First program (run by CMHA-Ottawa) stipu- lates that no participant pays more than the shelter portion of their monthly in- come support cheque. less than this. Some have as little as $100 per month to li ve on once rent is paid (Falvo, 2008: 34). With 64 percent of clients receiving Ontario Works benefits (i.e. basic welfare), perhaps it should come as no surprise that a similar percentage (68 percent) reported that, once rent was paid, they did not have enough money to live on (City of Toronto, 2007: 46-48). The affordability problems experienced by S2H clients have impor- tant implications for their general well-being. For example, due largely to housing affordability problems, fewer than 10 percent of S2H partici- pants have a telephone.20 S2H post-occupancy research does not track the extent to which participants are meeting their nutritional needs. However, roughly two- thirds of respondents reported that they ?regularly ran out of money to buy food? (City of Toronto, 2007: 47). And not surprisingly, S2H clients report that, of all the services they have used once housed, food banks are by far the ones that they use the most (City of Toronto, 2007: 90). This may explain ? at least in part ? why only 40 percent of respondents to the post-occupancy survey reported that their social interaction had improved since being housed. In fact, 26 per- cent of respondents reported that their social interaction had ?gotten worse? (City of Toronto, 2007: 88). Research demonstrates a direct relationship between a household?s income level and its purchase of foods from all groups, particularly fruit, vegetables and milk. This relationship is especially strong when a household?s annual income is below $15 ,000 (Ricciuto et al., 2006). One recent study even shows an inverse relationship between the percentage of household income allocated to housing and the adequacy of food spending. Again, this relationship is especially strong among lower in- come households (Kirkpatrick and Tarasuk, 2007; Friendly, 2008). When asked if they felt that they had a choice in the type of housing they were offered through the program, 29 percent of survey participants responded with an outright ?no.? Likewise, when asked if they felt that they had a choice in the location of their housing, 30 percent said ?no? (City of Toronto, 2007: 81). 20 Not surprisingly, those receiving ODSP benefits are far more likely to have a telephone than those receiving Ontario Works benefits (Falvo, 2008: 34). The post-occupancy survey also identified problems with clients in shared accommodation, representing 39 percent of all S2H clients. Ac- cording to the City?s post-occupancy survey report: Those in shared accommodation are less likely to feel secure about their housing, are far more likely to move,21 and need more help from their fol- low-up workers to relocate. People in shared accommodation frequently reported issues with roommates/housemates that made it difficult to keep their housing. Most quality of life indicators also showed less improve- ment for people in shared accommodation (City of Toronto, 2007: 2)... Those in shared accommodation were more likely to say that the amount of food they ate had stayed the same or gotten worse... This was most of- ten attributed to a lack of secure food storage areas, as several people commented on the fact that they had problems with roommates stealing their food, or that they lacked adequate, secure food storage spaces (City of Toronto, 2007: 38)... Those in shared accommodations are less likely to have reductions in the use of emergency services, and are more likely to have been arrested since being housed (25% compared to 12%) and to have used an ambulance (28% compared to 14%) (City of Toronto, 2007: 52).22 Not surprisingly, most of the S2H clients who are not still housed with the program (and yet are still alive and in Toronto) were in shared accommodation. The program?s reliance on shared accommodation for such a substantial percentage of its units is mostly due to a lack of fund- ing and the lack of supply of affordable housing in Toronto. If the pro- gram had sufficient funding to provide shelter allowances (i.e. ?portable 21 The survey revealed that ?46% of those who were originally in shared accom- modation had moved at least once, compared to 17% of those in independent units. Of those who moved while in shared accommodation, 38% said it was be- cause of problems with their roommates? (City of Toronto, 2007: 34). 22 Paradoxically, those in shared accommodation fared better in one category: they were more likely to report that they had reduced their drinking (58 percent compared with 44 percent). But not surprisingly, they were less likely to report that they had quit drinking (12 percent compared with 20 percent); less likely to report that they had decreased their use of other drugs (63 percent compared with 84 percent); and less likely to report that they had quit using other drugs altogether (12 percent compared with 44 percent) (City of Toronto, 2007: 45). rent supplements?) for all its tenants, few if any would live in shared accommodation arrangements. Post-occupancy research also shows that Aboriginal program par- ticipants ? who made up 26 percent of those surveyed ? fared significant- ly worse in several areas, as illustrated in Table 4.23 Table 4 : Aboriginal Clients Aboriginal Non-Aboriginal Improved health 61 % 74% Improved food 43% 73% Reduced stress 48% 65% Improved sleeping 52% 75% Improved personal safety 52% 80% Source: City of Toronto (2007: 43). Concern is also warranted about the long-term well-being of S2H clients, especially after their 12-month follow-up support period has ex- pired. For example, the post-occupancy research study was done in the relatively early stages after each client?s placement into permanent hous- ing. Indeed, 100 percent of all S2H clients surveyed were still in contact with the S2H program staff at the time of the survey, and many were still receiving regular support. Since homeless people housed in supportive housing typically need many years of support after receiving their hous- ing, it would be naive to believe that S2H clients need only 12 months of follow-up support. Interagency Relations Most of the representatives from community agencies interviewed for this paper told the writer that officials with the Shelter Support and Housing Administration Division of the City of Toronto have become less flexible and conciliatory with S2H than they have been with past programs. There is a sense that the input of community agencies is less fully accepted now than was the case with program planning prior to 23 The Aboriginal clients surveyed had been homeless longer and were more like- ly to have been in shared accommodation than the non-Aboriginal clients. S2H. Moreover, major changes are made to S2H without sufficient con- sultation with community agencies. According to interviewees, the clearest manifestation of this new approach is with the Street Outreach Steering Committee. The commit- tee?s role is to provide advice to the General Manager of Shelter, Support and Housing Administration (who chairs all committee meetings) on the direction of the S2H program. Indeed, a wide range of community part- ners are full-fledged members of the committee. This typically means, among other things, that the Executive Directors of various S2H partner agencies attend meetings. To the City?s credit, this includes strong voices who were known in advance as being blunt and not always agreeable. One interviewee who is well-informed on the workings of the committee noted that the General Manager has shared information with committee members that he might not have shared with other stakeholders. As a result, the interviewee noted that the committee has had important and frank discussions that have informed S2H?s direction. However, key interviewees for this paper consistently expressed concern about the committee?s insufficient involvement of its members. One interviewee pointed out that minutes of committee meetings were not even kept for the first year. The perceived lack of flexibility shown in this committee?s opera- tions has, according to some interviewees, alienated representatives of community agencies with long established track records in serving To- ronto?s homeless population. One interviewee went further, noting that: the City sets the agenda and poses specific questions of the group. [How- ever,] input is not sought on the direction of the committee?s work , and certainly not on the direction of S2H initiatives in general. I think there may be a point to be made that in other areas (Ottawa, York Region for sure) the municipality is at the planning table, but is not driving the process quite like Toronto. Toronto is headed for a situation in which they are doing all of the work themselves because they have alienated the community. And that would be very expensive for taxpayers? Transferability A useful ? albeit unscientific ? indicator of the amount of interest throughout Canada in the Housing First model is the dissemination work of S2H staff. Between mid-2007 and mid-2008, S2H staff travelled to 23 different Canadian municipalities to discuss S2H with local offi- cials. Moreover, Regina, Ottawa, Grand Prairie, Lethbridge, Calgary, and Edmonton have sent contingents of staff to Toronto to learn and train with S2H officials, usually for four days at a time. Interestingly, there is no Canadian equivalent of the United States Interagency Council on Homelessness, which, among other responsibilities, typically carries out this mentoring and training role for municipal officials in the United States. In Canada, S2H program officials have been playing this role by default. Lethbridge, Calgary, Sudbury, Ottawa and London, already have Housing First programs in place, and Edmonton and Victoria are planning their own programs. But, as successful as S2H has been in Toronto, there are important considerations to bear in mind for other jurisdictions wanting to imple- ment Housing First programs of their own. These considerations fall into four broad categories: leadership, market dynamics, institutional capaci- ty and regulatory systems. Leadership Canadian municipalities that have successfully implemented Housing First programs typically have one key person each who has pushed the model forward ? usually either a city councillor or a bureaucrat. Other Canadian municipalities ought to do the same when trying to implement the model. Market Dynamics One of the reasons for the program?s success has been the fact that va- cancy rates have been relatively high in Toronto since the program?s in- ception. The February 2005 report to Toronto City Council that paved the way for S2H noted the following: ?There are [now] increased opportuni- ties in the private rental market. In 1999 the reported vacancy rate in pri- vate rental housing was a mere 0.9 percent, while today it has risen to 4.3 percent? (City of Toronto, 2005: 22). Calgary, for instance, has lower vacancy rates than Toronto. Moreo- ver, its rental housing stock is newer and more expensive than Toronto?s. Not surprisingly, officials with Calgary?s Housing First program have not been able to find landlords as easily as the counterparts in Toronto. Though Calgary?s program has recruited landlords, it has only done so by offering them very deep rent supplements in the order of $700 to $800 per unit per month. By comparison, when Toronto used rent supple- ments to recruit some of its landlords, the rent supplements in question were roughly half that amount. Thus, municipalities should seek to implement or expand this mod- el during times of relatively high vacancy rates.24 Institutional Capacity Not all municipalities have the same institutional capacity to design and implement a program for homeless people. Toronto, with its large home- less population and years of programming in the area, is exceptional among Canadian municipalities. Toronto officials have many years of expertise and knowledge in designing and delivering homeless pro- grams, which is a relatively new area for most municipalities. Thus, oth- er municipalities should seek guidance from Toronto in implementing their programs. Regulatory Systems Throughout Canada, there has been a general tendency toward rental market deregulation in the past decade. Relative to several other Cana- dian provinces, Ontario has a significant degree of rent control, to which most landlords have grown accustomed. Alberta, by comparison, is a province with very basic tenant protection; it has much less regulation, meaning that landlords are not as used to co-operating and remaining at a given rent. Therefore, municipalities with less rental market regulation should be cautious in moving forward on an S2H-type framework and expect more challenges in finding landlords who will co-operate. All of the above considerations need bearing in mind when officials contemplate transferring the Housing First model to other jurisdictions. Of course, the model can be replicated in any jurisdiction, but the ques- 24 For a consideration of which policy options are appropriate for which contexts, see Falvo (2007). tion is one of scale. Will the replicated program in another jurisdiction house 600 new people per year (as is the case with Toronto), or will it house 20 people per year? Conclusion Canada?s supply of affordable housing is limited, and a disturbing num- ber of Canadian households are in core housing need. Thus, a well- funded national housing strategy aimed at the most destitute ? in partic- ular, the homeless ? may be more important now than ever. Toronto?s application of the Housing First model does not replace the need for a broader, national housing strategy. Rather, S2H is a program that helps a limited number of those in Toronto who experience housing affordability problems. Seen in that way, it appears to be an effective model of helping rough sleepers access Toronto?s limited supply of low-cost rental hous- ing. While the data on the program?s clientele suffer from the methodo- logical shortcomings outlined at the outset of this paper, these data do indicate that S2H has done a good job of moving its target population into permanent housing. On the basis of both these data and key informant interviews, To- ronto?s S2H program should not only continue to operate but also be seen as a model for other Canadian municipalities to emulate. But sever- al ingredients would improve Toronto?s S2H program and facilitate the model?s transferability to other municipalities. First, the federal govern- ment ought to make permanent the Homelessness Partnership Initiative (HPI). Second, provinces have to help municipalities both bridge the af- fordability gap for Housing First clients and ensure that long-term case management is available to those clients who need it. Third, municipali- ties need to both work effectively with their community partners and plan for solid research at the outset of Housing First program develop- ment. As S2H evolves ? as it did in May 2008 to address the broader is- sue of panhandling in downtown Toronto ? city officials have the oppor- tunity to improve it. To be sure, it may be that some of the above recommendations have already begun to inform both S2H and compara- ble programs in other Canadian municipalities. Policy Recommendations The poor results for clients who have had to settle for shared accommo- dation ? as well as the real possibility of a drop in vacancy rates ? speaks to the ongoing need of senior levels of government to fund a long-term affordable housing strategy, complete with supply-side measures. In the interim, each level of government can help make S2H an even more ef- fective program. Government of Canada Service Canada, Human Resources and Social Development Canada, and Canada Mortgage and Housing Corporation have all been actively en- gaged with S2H officials through discussions with city officials and funding for S2H. But the federal government should go further by mak- ing permanent the Homelessness Partnership Initiative (HPI), a program of Human Resources and Social Development Canada. The HPI provides a substantial amount of the S2H budget. Thus, not only could this make the budgets of S2H and similar programs in other municipalities more secure, but it would also allow municipal staff to engage in long-term planning. Government of Ontario In its final report of January 1999, the most exhaustive task force study undertaken on homelessness in Canadian history had the following to say about which level of government should pay for shelter allowances: Shelter allowances, because they are income transfers, should be a pro- vincial responsibility. This is the case in the four Canadian provinces that have shelter allowances today. Shelter allowances fit with the declared priorities of the provincial government (Golden et al., 1999: 85). It is therefore astonishing that the Province has not been more en- gaged with S2H. Though overtures have been made by city officials to discuss the program with the Ministries of Community and Social Ser- vices, Health and Long-Term Care, and Municipal Affairs and Housing, the Province has shown little interest. This ought to change, especially given the Province?s responsibility for assisting low-income Ontario te- nants with housing affordability. A good start would be for the Ministry of Health to involve S2H officials with the Local Health Integration Net- work process. More importantly, however, the Ministry of Municipal Affairs and Housing should provide sufficient funding so that each S2H participant can have a portable shelter allowance (rent supplement) for use in the private-sector units. The portability would be important be- cause many S2H participants transfer at least once after being housed. Likewise, the Province should provide similar rent supplement funding to all municipalities in Ontario that fund Housing First programs. If S2H clients each had a portable rent supplement, they would be less likely to have affordability problems and less likely to have to settle for shared accommodation. A deep rent supplement program providing 400 rent supplements per year in Toronto in the range of $400 per unit per month would cost roughly $2 million annually. Also, the Mental Health Branch of the Ontario Ministry of Health and Long-Term Care should commit to addressing the long-term case management needs of S2H clients beyond their first 12 months in the program. At present, Ontario?s mental health system consists of an ad hoc, uncoordinated support system. Given this reality, S2H clients could soon become homeless after their first 12 months if the Province does not commit to assisting them after this point. City of Toronto The general manager of the Shelter, Support and Housing Administra- tion (City of Toronto) could build greater trust and confidence among community agencies by inviting a member of this sector to co-chair the Street Outreach Steering Committee. If a new co-chair from the commu- nity sector had a role in setting the committee?s agenda on a regular ba- sis, representatives of community agencies might feel less alienated. Canadian Municipalities Other Canadian municipalities wanting to transfer the Housing First model into their jurisdiction should plan for solid evaluation from the outset. This should involve the following three evaluation components: 1. Plan for evaluation while developing the program, not after. 1.2. Collaborate with people who have expertise in evaluation and/or research. 1.3. Ensure that the evaluators have a reasonable degree of independence from those who have a vested interest in the findings. This chapter is based on Homelessness, Program Responses, and an Assess- ment of Toronto?s Streets to Homes Program, published in February 2009 by the Canadian Policy Research Networks. Nick Falvo is a PhD Candidate at Carleton University?s School of Public Poli- cy and Administration. A frequent guest lecturer at Canadian universities, his research interests include homelessness, affordable housing, child benefits, social assistance and program evaluation. He is currently the main researcher on a three-year study exploring homelessness and affordable housing in Canada?s North. Prior to his doctoral studies, Mr. Falvo was a Parliamentary Intern in Ottawa and worked for 10 years as a front -line community worker with home- less persons in Toronto. Acknowledgements The author wishes to acknowledge the funding support of the Canadian Policy Research Networks and the Social Housing Services Corporation Housing Internship Program. The author also wishes to thank the fol- lowing individuals for invaluable assistance at different stages in the preparation of this article: Mike Buzzelli, George Fallis, David Hul- chanski, Judith McKenzie, Sylvia Novac, Louis-Philippe Rochon, Beth Shinn, Greg Suttor and Suzanne Swanton. Dozens of others were also very helpful, but their anonymity is being preserved. The usual disclai- mer applies. References City of Toronto (2003). Toronto report card on housing & homelessness. Toronto. City of Toronto (2005). From the street into homes: A strategy to assist homeless per- sons find permanent housing. Toronto. City of Toronto (2006). Street needs assessment: Results and key findings. Toronto. City of Toronto (2007). What housing first means for people: Results of streets to homes 2007 post -occupancy research. Toronto: Shelter and Housing Administration. Formatted: Bullets and Numbering Dowling, P. (1998, September). Analysis of funding for homelessness initiatives in Toronto. Background paper for the Mayor?s Homelessness Action Task Force. Toronto: City of Toronto. Falvo, N. (2007, June). Addressing Canada?s lack of affordable housing. Paper pre- sented at Annual meetings of the Canadian Economics Association, Halifax, NS. Available from Street Health Web Site: www.streethealth.ca/Down- loads/NickCEA-0507.pdf. Falvo, N. (2008). The ?housing first? model: Immediate access to permanent hou sing. Canadian Housing [Special ed.], 32 -35. Friendly, A. (2008). Towards food security policy for Canada?s social housing sector. Ottawa: Canadian Policy Research Networks. Galea, S., & Vlahov, D. (Eds.). (2005). Handbook of urban health: Populations, me- thods, and practice. New York: Springer Science+Business M edia, Inc. Gallant, G., Brown, J., & Tremblay, J. (2004). From tent city to housing: An evalua- tion of the City of Toronto?s Emergency Homelessness Pilot Project. Toronto: City of Toronto. Golden, A., Currie, W. H., Greaves, E., & Latimer, E. J. (1999). Taking responsibility for homelessness: An action plan for Toronto. Report of the Mayor?s Homeless- ness Action Task Force [Anne Golden Task Force Report]. Toro nto: City of Toronto. Greenwood, R. M., Schaefer-McDaniel, N. J., Winkel, G., & Tsemberis, S. J. (2005). Decreasing psychiatric symptoms by increasing choice in services for adults with histories of homelessness. American Journal of Community Psychology, 36(3-4), 223-238. Gulcur, L., Stefancic, A., Shinn, M., Tsemberis, S., & Fischer, S. N. (2003). Hous- ing, hospitalization, and cost outcomes for homeless individuals with psy- chiatric disabilities participating in continuum of care and housing first programmes. Journal of Community & Applied Social Psychology , 13(2), 171- 186. Housing Connections (2006). Annual statistical report. Toronto: City of Toronto. Housing Connections (2007). Annual statistical report. Toronto: City of Toronto. Khandor, E., & Mason, K. (2007). The Street Health report. Toronto: Street Health. Kirkpatrick, S. I., & Tarasuk, V. (2007). Adequacy of food spending is related to housing expenditures among lower-income Canadian households. Public Health Nutrition, 10(12), 1-10. Main, T. (1997, November 3). A tale of two cities: Managing homelessness in New York City and Toronto. In Mental illness and pathways into homelessness: Findings and implications [Conference Proceedings] . Toronto: Mental Health Policy Research Group. McCarroll, C. (2002, May 1). Pathways to housing the homeless. Christian Science Monitor, 94(110), 11. Metropolitan Toronto (1983). No place to go: A study of homelessness in Metropolitan Toronto: Characteristics, trends and potential solutions. Prepared for the Metro- politan Toronto Assisted Housing Study. Toronto. Myles, J. (1996). When markets fail: Social welfare in Canada and the United States. In G. Esping-Andersen (Ed.), Welfare states in transition (pp. 116-140). London: Sage Publications. Ontario Ministry of Housing (1988). More than just a roof: Action to end homeless- ness in Ontario. Final report of the Minister?s Advisory Committee on the In- ternational Year of Shelter for the Homeless. Toronto. Padgett, D. K. (2007). There?s no place like (a) home: Ontological security among persons with serious mental illness in the United States. Social Science and Medicine, 64(9), 1925-1936. Padgett, D. K., Gulcur, L., & Tsemberis, S. (2006). Housing First services for people who are homeless with co-occurring serious mental illness and sub- stance abuse. Research on Social Work Practice, 16(1), 74-83. Ricciuto, L., Tarasuk, V., & Yatchew, A. (2006). Socio-demographic influences on food purchasing among Canadian households. European Journal of Clinical Nutrition, 60(6), 778-790. Springer, J. H., Mars, J. H., & Dennison, M. (1998). A profile of the Toronto homeless population. Background report prepared for the Homelessness Action Task Force. Tsemberis, S., & Eisenberg, R. F. (2000). Pathways to housing: Supported housing for street-dwelling homeless individuals with psychiatric disabilities. Psy- chiatric Services, 51(4), 487-493. Chapter 2 .1 The Relationship Between Homelessness and Health: An Overview of Research in Canada C. JAMES F RANKISH, STEPHEN W. HWANG, AND DARRYL Q UANT Z Canada has long had an international reputation for high quality of life. For a growing number of Canadians, homelessness has become a grim reality and obtaining shelter part of a daily struggle (Begin et al., 1999). Research on homelessness is essential for policy-makers, program plan- ners, service providers, and community groups. This knowledge can play an important role in public education and awareness campaigns, policy decisions, resource allocation, program development, and pro- gram or policy evaluation (Quantz & Frankish, 2002). The identification of needs and priorities for research on homelessness is, therefore, a valu- able undertaking. The two primary goals of this article are to provide an overview of previous research on homelessness and the relationship between home- lessness and health (with a main focus on Canada), and to spur discus- sion regarding strategic directions for future research. The National Homelessness Initiative has called for a comprehensive Canadian re- search agenda to ?lay the foundation for understanding the root causes of homelessness, support policy development and serve as a resource for accountability and reporting.? Development of this agenda will require active engagement by a wide range of stakeholders, including homeless people, those at risk of becoming homeless, service providers and advo- cates for homeless people, government representatives, researchers and research funding agencies. Literature review A variety of strategies were used to identify literature on homelessness that reflected diversity in both geographical and topical focus. This was deemed essential considering that many important sources of informa- tion are found in reports from government and community agencies, in addition to the peer-reviewed academic literature. This article is not a comprehensive review of the literature on homelessness in Canada, but rather an effort to frame the different types and areas of research for the purpose of developing future work. An initial search strategy involved the use of electronic databases, including major social sciences, health and humanities databases. A second strategy sought out examples of literature from government, community, advocacy and service websites. Examples of homelessness research, program descriptions and policy documents were collected. Canadian literature was the primary target of these searches, but review papers from international sources were also included for comparison purposes and to provide additional examples of interventions. Only documents that identified homelessness as a major focus were collected. Papers and reports on housing policy and programs were only included if they focused on homelessness. General reports on housing policy and programs were excluded. Only literature and reports published since 1990 in English were reviewed. Collected documents were reviewed and categorized. Research was defined broadly to include the systematic generation of new knowledge through a variety of means, including descriptive reports. A more re- strictive definition (for example, one based on specific methods such as controlled trials) would have excluded a large proportion of the litera- ture on homelessness in Canada. Research within the following catego- ries were included: ? conceptual research (examining the definition/meaning of homeless- ness); ? environmental scans (documenting the extent of homelessness and health and social issues related to homelessness); ? methods research (focusing on the development of new tools for studying homelessness); ? needs assessments (focusing on the needs of the homeless as ex- pressed by the homeless and service providers); ? evaluation research (describing the process and outcomes of pro- grams and policies); ? intervention research (examining the effectiveness of programs and services). The scope of homelessness in Canada Many efforts have focused on obtaining a clearer understanding of the nature and extent of homelessness in Canada. Canada?s first efforts to provide an estimate of the homeless population began in 1987 through the work of the Canadian Council on Social Development (Begin et al., 1999). Further efforts at measuring homelessness have been undertaken by Statistics Canada. Data from the 2001 Census indicated that over 14,000 individuals were homeless in this country (Statistics Canada, 2002). Most advocates and researchers, however, believe that these num- bers vastly under-represent the problem, and new strategies are neces- sary to accurately capture usable information. Other strategies include the development of the Homeless Individuals and Families Information System (HIFIS) that focuses on capturing more complete information on shelter users in cities across Canada (Canada Mortgage and Housing Corporation, 1999). Specific cities in Canada have also initiated local homelessness counts in an attempt to measure the numbers of homeless and at-risk persons in their jurisdictions. Examples from large urban areas include a report on homeless and at-risk persons in the Greater Vancouver region (Woodward et al., 2002), the Toronto Report Card on Homelessness (City of Toronto, 2000), and the City of Calgary Homeless Count (Stroik, 2004). A number of smaller cities and regions have produced similar reports. The challenges associated with obtaining a clear picture of the scope of homelessness in Canada included the lack of a consistent definition of homelessness, difficulty in identifying homeless persons, the transient nature of homelessness, difficulty in communicating with homeless per- sons, and lack of participation by local agencies (Bentley, 1995). The de- finition of homelessness is particularly important. Homelessness can be viewed along a continuum, with those living outdoors and in other plac- es not intended for human habitation at the extreme, followed by those living in shelters. These individuals are referred to as being absolutely homeless. Homelessness also includes people who are staying with friends or family on a temporary basis, often referred to as ?couch surf- ing? or being ?doubled up.? Those at risk of being homeless include per- sons who are living in substandard or unsafe housing and persons who are spending a very large proportion of their monthly income on hous- ing. The definition of homelessness is not trivial. It can have profound consequences for policy, resource allocation, and parameters used to evaluate the success of homelessness initiatives. This article focuses on research and interventions related to absolute homelessness. Much of this information has implications for those who are at risk. Other important aspects of homelessness in Canada are the impact of urbanization, the heterogeneity of the homeless population, and the complexity of the causes of homelessness. Canada is experiencing a rap- id and continuing trend towards urbanization, as indicated by the fact that almost 80 percent of Canadians now live in cities with populations of 10,000 or more. Although homelessness is a problem in rural areas of Canada, it has become an obvious crisis in large urban areas, where availability of affordable housing is limited due to a loss of rental units and a shortage of social housing (Woodward et al., 2002). Heterogeneity within the homeless population is important to rec- ognize. Homelessness affects single men and women, street youth, fami- lies with children, people of all races and ethnicities, lifelong Canadians, immigrants and refugees, and these groups often face different health issues (Hwang, 2001). For most individuals, homelessness represents a transient one-time crisis or an episodic problem; for a distinctly different subgroup of individuals, homelessness is a chronic condition (Kuhn & Culhane, 1998). There is no single pathway to homelessness. Homelessness is the re- sult of a complex interaction of factors at the individual level such as ad- verse childhood experiences, low educational attainment, lack of job skills, family breakdown, mental illness and substance abuse (Herman et al., 1997; Koegel et al., 1995; Susser et al., 1993) and at the societal level, such as poverty, high housing costs, labour market conditions, decreased public benefits, and racism and discrimination (Jencks, 1994; O?Flaherty, 1996; Schwartz & Carpenter, 1999) (see Figure 1). Research on homelessness has often reflected disciplinary tradi- tions, with health researchers focusing on individual risk factors and so- cial scientists looking at marginalization, exclusion and economic forces. This is important because the formulation of the causes of homelessness can become highly politicized and can influence public perceptions and policies related to homelessness. The health status of homeless persons Causal Pathways Homelessness is clearly associated with poor health. In reviewing the research in this area, a schema of causal pathways underlying this asso- ciation may be useful. Homelessness has a direct adverse impact on health (Figure 1, arrow C). Crowded shelter conditions can result in ex- posure to tuberculosis or infestations with scabies and lice, and long pe- riods of walking and standing and prolonged exposure of the feet to moisture and cold can lead to cellulitis, venous stasis and fungal infec- tions (Stratigos & Katsambas, 2003). However, the relationship between homelessness and ill health is far more complex (Hwang, 2002). Many risk factors for homelessness, such as poverty and substance use (Figure 1, arrow A), are also strong independent risk factors for ill health (Figure 1, arrow D). Many people who are homeless remain at risk for poor health even if they obtain stable housing. In addition, certain health con- ditions (particularly mental illness) may contribute to the onset of home- lessness and then in turn be exacerbated by the homeless state (Figure 1, arrows C and E). Finally, improved health and adequate housing are means of achieving the ultimate goal of improved quality of life. Re- searchers are now recognizing the need to understand and measure the impact of interventions on quality of life, in addition to housing and health outcomes (Lehman et al., 1995). Specific Health Conditions Homeless people are at greatly increased risk of death. Mortality rates among street youths in Montreal are nine times higher for males and 31 times higher for females, compared to the general population (Roy et al., 1998a). Men using homeless shelters in Toronto are two to eight times more likely to die than their counterparts in the general population (Hwang, 2000, 2002). The prevalence of mental illness and substance abuse is much high- er among homeless adults than in the general population. Contrary to popular misconceptions, only a small proportion of the homeless popu- lation suffers from schizophrenia. The lifetime prevalence of schizophre- nia is only 6 percent among Toronto?s homeless (Canadian Mental Health Association, 1998). Affective disorders are more common, with lifetime prevalence rates of 20-40 percent (Fischer & Breakey, 1991; Suss- er et al., 1993). Alcohol use disorders are widespread, with lifetime pre- valence rates of about 60 percent in homeless men (Fischer & Breakey, 1991). Cocaine and marijuana are the illicit drugs most often used by homeless Canadians (Smart & Adlaf, 1991). Patterns of substance abuse and mental illness vary across subgroups of homeless people: single women are more likely to have mental illness and less likely to have sub- stance use disorders than single men (Fischer & Breakey, 1991). Female heads of homeless families have far lower rates of both substance abuse and mental illness than other homeless people (Shinn et al., 1998). Homeless people are at increased risk of tuberculosis (TB) due to al- coholism, poor nutritional status and AIDS (Advisory Council for the Elimination of Tuberculosis, 1992). In addition, the likelihood of expo- sure to TB is high in shelters due to crowding, large transient popula- tions, and inadequate ventilation (Nolan et al., 1991). Canadian data on the incidence and molecular epidemiology of TB among homeless people are lacking. In the United States, more than half of TB cases among homeless people represent clusters of primary tuberculosis, rather than reactivation of old disease (Barnes et al., 1996). Treatment of active TB in the homeless is complicated by loss to follow-up, non-adherence to ther- apy, prolonged infectivity and drug resistance (Pablos-Mendez et al., 1997). Directly observed therapy results in higher cure rates and fewer relapses (Advisory Council for the Elimination of Tuberculosis, 1992). Homeless persons with positive skin tests without active TB may be con- sidered for directly observed prophylaxis (Nazar-Stewart & Nolan, 1992). Among homeless youth in Canada, risk factors for HIV infection in- clude survival sex, multiple sexual partners, inconsistent use of condoms and injection drug use (Roy et al., 1999). Infection rates were 2.2 percent and 11.3 percent among homeless youths seeking HIV testing at two clinics in Vancouver in 1988 (Manzon et al., 1992). In contrast, the preva- lence of HIV infection was only 0.6 percent in a group of homeless youths surveyed in Toronto in 1990 (Wang et al., 1991). In a 1997 study of homeless adults in Toronto, the HIV infection rate was 1.8 percent, with increased risk observed among individuals with a history of using IV drugs or crack cocaine (Goering et al., 2002). A study of homeless adults and runaway youth in 14 US cities in 1989?92 found HIV infection rates ranging from 0 to 21 percent with a median of 3.3 percent (Allen et al., 1994). Sexual and reproductive health are major issues for street youth. Studies of street-involved youth in Montreal have documented high rates of involvement in survival sex, sexually transmitted diseases and unplanned pregnancy (Roy et al., 1998b, 1999, 2003). Anecdotal reports suggest that pregnancy is common among street youths in Canada; in the US, 10 percent of homeless female youths aged 14-17 years are cur- rently pregnant (Greene & Ringwalt, 1998). Injuries and assaults are a serious threat to the health of homeless people. In Toronto, 40 percent of homeless persons have been assaulted and 21 percent of homeless women have been raped in the past year (Crowe & Hardill, 1993). Unintentional injuries due to falls or being struck by a vehicle, as well as drug overdoses, are leading causes of mor- tality among homeless men in Toronto (Roy et al., 1998a). Homeless adults suffer from a wide range of chronic medical condi- tions, including seizures, chronic obstructive pulmonary disease and musculoskeletal disorders (Crowe & Hardill, 1993). Hypertension and diabetes are often inadequately controlled (Hwang & Bugeja, 2000; Kin- chen & Wright, 1991). Homeless people in their forties and fifties often develop health disabilities that are commonly seen in persons who are decades older (Gelberg et al., 1990). Oral and dental health is poor (Gib- son et al., 2003; Lee et al., 1994; Pizem et al., 1994). Homeless people face many barriers that impair their access to health care, even under the Canadian system of universal health insur- ance. Many homeless persons do not have a health card, are unable to make or keep appointments, or lack continuity of care due to their tran- sience (i.e., no permanent address or telephone). Homelessness entails a daily struggle for the essentials of life. Competing priorities may impede homeless people from obtaining needed health services (Gelberg et al., 1997). Access to mental health care and substance abuse treatment re- mains a crucial issue (Wasylenki et al., 1993). Obtaining prescription me- dications can be problematic and adhering to medical recommendations regarding rest or dietary modification is often impossible (Hwang & Bu- geja, 2000; Hwang & Gottlieb, 1999). Studies from the United States have shown that homeless adults have high levels of health-care utilization and often obtain care in emergency departments (Kushel et al., 2002; Ku- shel et al., 2001). Homeless people are hospitalized up to five times more often than the general public (Martell et al, 1992) and stay in the hospital longer than other low-income patients (Salit et al., 1998). Interventions to reduce homelessness and improve the health of homeless persons This section provides an overview of the wide array of interventions re- ported within the literature that have attempted to decrease the preva- lence of homelessness and improve the health of homeless people. We have classified these interventions into four clusters using a taxonomy derived from the literature, theory and past experience: ? biomedical and health care strategies; ? educational and behavioural strategies; ? environmental strategies; ? policy and legislative strategies. For each cluster, we provide a brief description, examples of inter- ventions of that type, and a summary of research gaps and opportunities within that cluster. These clusters are not mutually exclusive; some in- terventions may fit under more than one cluster. Biomedical and Health Care Strategies This cluster of strategies focuses on medical interventions to improve health status and includes primary health-care programs, clinical servic- es through outreach programs, psychiatric treatment teams and sub- stance abuse treatment. Interventions that are purely biomedical, how- ever, may improve the health of homeless people but fail to address their homelessness. Thus, interventions that combine health care with housing and other social services need to be considered. Only a small number of studies have examined the effectiveness of biomedical or health care interventions for homeless people using a ri- gorous controlled design. Most of these studies have focused on home- less persons with mental illness or substance abuse. For example, studies have confirmed the effectiveness of the Assertive Community Treatment (ACT) model for homeless people with severe mental illness. ACT in- volves a team of psychiatrists, nurses, and social workers that follows a small caseload of clients in the community and provides high-intensity treatment and case management (Lehman et al., 1997; Wasylenki et al., 1993). Compared to usual care, patients receiving ACT have fewer psy- chiatric in-patient days, more days in community housing, and greater symptom improvement. A recent example of a combined housing and health service pro- gram is the New York City Housing Initiative (Metraux et al., 2003). This program made resources available to create 3,300 housing units and so- cial services support for mentally ill homeless persons. Over two years, people in the program stayed in shelters an average of 128 days fewer than similar people in a control group. The treatment of substance abuse in homeless persons has been the subject of a number of studies; a recent review of the literature is available (Zerger, 2002). Gaps in this area include a lack of research on interventions for homeless youth or families with children, limited research on interven- tions to address health problems other than mental illness or substance abuse, and little or no data on the effectiveness of various models of pri- mary care delivery for the homeless. Opportunities for future research include a focus on ?harm-reduction? programs that seek to minimize adverse health impacts among homeless substance users rather than fo- cusing exclusively on abstinence. Examples include ?safe injection sites? for drug users and shelter-based controlled drinking programs in which residents are provided with alcohol on a metered schedule. Educational and Behavioural Strategies This cluster of strategies seeks to prevent homelessness or improve the health status of homeless persons through educational programs and behavioural change. Educational programs may focus on homeless people, individuals at risk of homelessness, or the general public. Efforts to promote behavioural change in the homeless include harm-reduction programs, counselling, and referral services. Education of health care workers, shelter workers, and service providers is included in these strategies. For example, the Streethealth Coalition in Ottawa provides prevention and education on infectious diseases and health conditions often found in the homeless (Canada Mortgage and Housing Corpora- tion, 1995). The Federation of Non-Profit Housing Organizations of Mon- treal promotes education on a range of basic life skills. Ontario?s Urban Aboriginal homelessness strategy includes culturally appropriate pro- grams, such as cultural counselling and programs, and employment ser- vices. Examples of programs targeting homeless or at-risk individuals in- clude tenants? rights organizations, eviction prevention services, and groups such as the Safe Homes for Youth in Ottawa, which provides education and support for high-risk youth (Canada Mortgage and Hous- ing Corporation, 1995). Alternatively, educational initiatives may focus on increasing public and government awareness of homelessness issues. Examples include a public awareness campaign in Ontario to aid the public in assisting homeless persons (Provincial Task Force on Home- lessness, 1998) and efforts by advocacy groups such as the Canadian Housing and Renewal Association, the Centre for Equality Rights in Ac- commodation and the Housing and Homelessness Network in Ontario to promote changes in government policy related to homelessness. Very little evaluation research has been undertaken on health edu- cation programs for the homeless (May & Evans, 1994). This constitutes a major research gap. Reports of educational and behavioural interven- tions have often been limited to basic program information. More in- depth descriptions of development and implementation processes are needed; such information could provide a valuable resource for service providers seeking to begin similar initiatives. Opportunities for future research include a need for conceptual research on educational and be- havioural interventions for homeless people, studies on how to make these interventions more accessible and appealing for the homeless pop- ulation, and rigorous studies to evaluate the outcomes of such programs. Such efforts could benefit from attention to three key factors: motivation of individuals toward change through altered knowledge, attitudes, be- liefs and values; enabling individuals to take action through skill build- ing and availability and accessibility of supportive resources; and reward or reinforcement of positive action (Green & Kreuter, 1999). Environmental Strategies Environmental strategies are attempts to alter the social, economic, or physical environment in a specific setting to create a supportive envi- ronment that enables and facilitates behaviour change. This approach recognizes that the environment or context in which homelessness oc- curs may be altered to enhance desired behaviours or limit undesirable actions. The environment or context may vary in scale from a single pro- gram (e.g., a supportive housing site or outreach program) to a specific neighbourhood to an entire city, province, or country. Examples of environmental strategies at the program level are Street City in Toronto, which provided services to homeless persons in an envi- ronment designed to engage individuals unaccustomed to living indoors (Canada Mortgage and Housing Corporation, 1995), and the Lookout Emergency Aid Society in Vancouver, which provided both short-term shelter as well as long-term supportive housing for adult men and wom- en who were unable independently to meet basic daily needs (Canada Mortgage and Housing Corporation, 1999). A macro-level example is the federal government?s Supporting Community Partnerships Initiative (SCPI), which seeks to promote cooperation and coordination at a local level and to provide ?communities with the tools and resources needed to set their own course of action? to respond to homelessness in their community. Research undertaken in environmental strategies has largely taken the form of environmental scans and needs assessments. Two reviews have documented and categorized a number of Canadian pro- grams/projects that included environmental strategies (Canada Mort- gage and Housing Corporation, 1995, 1999). A number of projects have provided examples of community development processes in the home- less population. Researchers have outlined lessons learned while con- ducting community-based research on homelessness in Toronto (Boydell et al., 2000). Others have looked at factors that restrict or facilitate com- munity participation by disadvantaged persons (Boyce, 2001). Opportun- ities for research include conceptual work to organize and frame these efforts, in-depth evaluations to ensure that programs have measurable outcomes, and translation of information into a form useful for planning (Quantz & Frankish, 2002). Policy and Legislative Strategies This cluster includes efforts to reduce homelessness through policies and legislation related to poverty and its amelioration, social housing, public health, immigration and law enforcement. Recognizing that a variety of policy, regulatory, legislative and political factors create a climate that has an enormous impact on homelessness and its management, these strategies focus on the creation of ?healthy public policies.? Examples of current initiatives include the government of Alberta?s framework outlining policy responses to homelessness with respect to housing and support services, local capacity development and govern- mental coordination (Alberta Community Development, 2000). The 1999 Vancouver Agreement is an example of collaboration at the federal, pro- vincial and municipal levels to focus on economic, social and community development in the Downtown Eastside neighbourhood, where home- lessness is a major issue. Examples of public health policies that have been implemented or considered include safe-injection sites, needle ex- change programs and other harm reduction policies. These strategies are foundational to all others, because the absence of a strong policy-legislative approach to homelessness will seriously limit and undermine efforts in other areas. There is a need for work to examine the impact of various health and social policies on the lives of homeless people. Particularly vital (Classer et al., 1999) areas include welfare policy as it affects adults and families with children, policies that impact young women (Novac et al., 2002), and practices in the child wel- fare system that may contribute to youth homelessness (Appathurai, 1991; Kufeldt, 1991). Comparing policies in different jurisdictions and their impact on homelessness can provide important insights (Classer et al., 1999; Eberle et al., 2001). Government frameworks on homelessness call for efforts to ensure accountability in reaching specific targets and goals. But, there has been relatively little work on policy evaluation in this area. Future research has the potential to provide essential informa- tion to guide future policy-making. Strategic Directions for Future Research on Homelessness Based our review, we conclude that Canadian research in the area of homelessness and health faces important challenges. First, the complexi- ty of the issue of homelessness requires the involvement of a wide range of stakeholders, including all levels of government, service providers, health professionals, biomedical/social science researchers, community groups and homeless people themselves. Both horizontal integration (across various sectors such as health, law, housing, social services) and vertical integration (across federal, provincial, territorial, and local gov- ernments, and within communities) are needed. Second, the diversity of values, beliefs and perspectives on home- lessness must be acknowledged, and public discourse is needed on the causes of homelessness in Canada and the appropriate response to this problem. Third, consensus needs to be reached on the definition of homeless- ness and the measures by which efforts to reduce homelessness or im- prove the quality of life of homeless people will be judged. Fourth, researchers need to design and conduct studies on home- lessness that are policy-relevant and develop strategies to translate their research into policy and practice. There has been little research evaluat- ing the effects of policy on homelessness or quality of life among the homeless and the vast majority of programs for homeless people have not been evaluated. Many of the evaluations that have been conducted are of modest quality, but at the present time, the resources and expertise that would allow for a robust evaluation are often not available at the local level. These challenges should not deter or diminish current interests and efforts around research on homelessness and health in Canada. Rather, they call for renewed commitment, strategic planning and wise invest- ment of human and fiscal resources. Within all six categories of research there is significant need for further development. Conceptual research on the definition and meaning of homelessness can provide greater clarity in ongoing discussions of homelessness among advocacy groups and policy-makers. Environmental scans that document the extent of home- lessness and the health problems of homeless people are useful, but they remain primarily descriptive in nature. There is a need to move from this understanding to outcome measures and interventions. Methods re- search could make significant contributions through the development of valid/precise measures of quality of life in homeless people and individ- uals at risk. Needs assessment research needs to be systematically linked to objectives and interventions. Finally, more high-quality evaluation and intervention research is urgently needed. Community involvement is vital in any work on homelessness and its conceptualization, measurement or change. While this may seem self- evident, the reality is that many groups often have limited capacity for engagement in these efforts. Concrete efforts are needed to ensure that communities are able to contribute to, and participate effectively in, the study of homelessness and use of research findings. The primary need is capacity-building to allow communities to initiate projects in equitable partnerships with government and academia. Resources must be made available to both promote research by various community groups and to teach research skills such as proposal writing and research design. Po- tential strategies include workshops, access to research courses at aca- demic institutions, the development of easy-to-use research information, and financial support to allow community members to participate in these activities. The issue of dissemination remains a key challenge in homelessness research. The question is how we can best communicate the lessons, ex- periences and best practices of dealing with homelessness. How can this information be communicated in a variety of forms and media that are appropriate to their target audiences? Significant barriers exist, including time, personnel, research capacity and resources. We suggest three strategic priorities towards a better understanding of homelessness and the implementation and evaluation of efforts to re- duce homelessness and improve the lives and quality of life among the homeless. The first priority is a nationwide effort to achieve a core, con- sensus definition and set of indicators related to the definition and extent of homelessness. Second, we need clear definitions and measures for a) the health status of homeless (and at-risk) groups; b) the use of the health and social services by homeless people; and c) relations between home- lessness and broader, non-medical determinants of health (e.g., income, education, employment, social support, gender, culture, etc.). This effort to create a common dataset would not preclude communities from col- lecting additional data of local interest and value. A third priority must be the development of research infrastructure. This effort would include the development of demonstration projects or surveillance systems that could reliably collect data on the indicators of homelessness. Government-funded projects that purport to address ei- ther the processes or outcomes of homelessness should be subjected to an ?evaluability? assessment. Groups such as the Canadian Consortium for Health Promotion Research could assist all levels of government in determining whether current projects/programs are in fact, evaluable. We suspect that many projects and programs presently lack the neces- sary and sufficient conditions to be fairly evaluated. This effort could move research toward a model of program evaluation that sets realistic expectations in terms of measurement of focused aspects of homeless- ness, and one that provides sufficient time and resources to allow for appropriate assessment of homelessness interventions and their effects. We encourage investment of the needed resources toward the science and application of research on homelessness. Building on its tra- ditions in health promotion and its strengths in population health re- search, Canada is well placed to be a world leader in intervention re- search on homelessness. This can be a vehicle for building community health. These efforts may generate additional benefits, including com- mitment to reducing health disparities, new partnerships across academ- ic disciplines, and intersectoral work on the determinants of health. C . James Frankish is with the Institute of Health Promotion Research, Univer- sity of British Columbia. Stephen W. Hwang and Darryl Quantz are at the Centre for Research on Inner City Health, St. Michael?s Hospital; Faculty of Medicine, University of Toronto. References Advisory Council for the Elimination of Tuberculosis (1992). Prevention and con- trol of tuberculosis among homeless persons: Recommendations of the Ad- visory Council for the Elimination of Tuberculosis. Morbidity and Mortality Weekly Report, 4l, 13-23. Alberta Community Development (2000). Moving forward? homelessness policy framework: Implementation strategy. Edmonton: Author. Allen, D. M., Lehman, J. S., Green, T. A., Lindegren, M. L., Onorato, I. M., & Forrester, W. (1994). HIV infection among homeless adults and runaway youth, United States, 1989-1992. AIDS, 8(11), 1593-1598. Appathurai, C. (1991). Developing policies for runaways: Insights from the litera- ture. Journal of Health & Social Policy, 2(4), 51-64. Barnes, P. F., El-Hajj, H., Preston-Martin, S., Cave, M. D., Jones, B. E., Otaya, M., Pogoda, J., & Eisenach, K. D. (1996). Transmission of tuberculosis among the urban homeless. The Journal of the American Medical Association, 275(4), 305-307. Begin, P., Casavant, L., Chenier, N. M., & Dupuis, J. (1999). Homelessness. Ottawa: Parliamentary Research Branch. Bentley, D. (1995). Measuring homelessness: A review of recent research. Winnipeg: Institute of Urban Studies, University of Winnipeg. Boyce, W. F. (2001). Disadvantaged persons? participation in health promotion projects: Some structural dimensions. Social Science & Medicine, 52(10), 1551- 1564. Boydell, K. M., Goering, P., & Morell-Bellai, T. L. (2000). Narratives of identity: Re-presentation of self in people who are homeless. Qualitative Health Re- search, 10(1), 26-38. Canada Mortgage and Housing Corporation (1995). Inventory of projects and pro- grams addressing homelessness. Ottawa: Author. Canada Mortgage and Housing Corporation (1999). Documentation of best practices addressing homelessness. Ottawa: Author. Canadian Mental Health Association (1998). Mental illness and pathways into home- lessness: Proceedings and recommendations. Toronto: Author. City of Toronto (2000). The Toronto report card on homelessness. Toronto: Author. Classer, I., Fournier, L., & Costopoulos, A. (1999). Homelessness in Quebec City, Quebec and Hartford, Connecticut: A cross-national and cross-cultural analysis. Urban Anthropology, 28(2), 141-159. Crowe, C., & Hardill, K. (1993). Nursing research and political change: The street health report. Canadian Nurse, 89(1), 21-24. Eberle, M., Kraus, D., & Serge, L. (2001). Homelessness: Causes & e ffects - Back- ground report: A profile and policy review of homelessness in the provinces of On- tario, Quebec and Alberta (Vol. 4). Vancouver: Ministry of Social Develop- ment and Economic Security. Fischer, P. J., & Breakey, W. R. (1991). The epidemiology of alcohol, drug, and mental disorders among homeless persons. American Psychologist, 46(11), 1115-1128. Gelberg, L., Gallagher, T. C., Andersen, R. M., & Koegel, P. (1997). Competing priorities as a barrier to medical care among homeless adults in Los An- geles. American Journal of Public Health, 87(2), 217-220. Gelberg, L., & Leake, B. D. (1993). Substance use among impoverished medical patients: The effect of housing status and other factors. Medical Care, 31(9), 757-766. Gelberg, L., Linn, L. S., & Mayer-Oakes, S. A. (1990). Differences in health status between older and younger homeless adults. Journal of the American Geria- trics Society, 38(11), 1220-1229. Gibson, G., Rosenheck, R., Tullner, J. B., Grimes, R. M., Seibyl, C. L., Rivera- Torres, A., Goodman, H. S., & Nunn, M. E. (2003). A national survey of the oral health status of homeless veterans. Journal of Public Health Dentistry, 63(1), 30-37. Goering P., Tolomiczenko, G., Sheldon, T., Boydell, K., & Wasylenki, D. (2002). Characteristics of persons who are homeless for the first time. Psychiatric Services, 53(11), 1472-1474. Green, L., & Kreuter, M. (1999). Health promotion planning: An educational and eco- logical approach. Mountain View, CA: Mayfield. Greene, J. M., & Ringwalt, C. L. (1998). Pregnancy among three national samples of runaway and homeless youth. Journal of Adolescent Health, 23(6), 370-377. Herman, D. B., Susser, E. S., Struening, E. L., & Link, B. L. (1997). Adverse child- hood experiences: Are they risk factors for adult homelessness?. American Journal of Public Health, 87(2), 249-255. Hwang, S. W. (2000). Mortality among men using homeless shelters in Toronto, Ontario. The Journal of the American Medical Association, 283(16), 2152-2157. Hwang, S. W. (2001). Homelessness and health. Canadian Medical Association Journal, 164(2), 229-33. Hwang, S. W. (2002). Is homelessness hazardous to your health?: Obstacles to the demonstration of a causal relationship. Canadian Journal of Public Health, 93(6), 407-410. Hwang, S. W., & Bugeja, A. L. (2000). Barriers to appropriate diabetes manage- ment among homeless people in Toronto. Canadian Medical Association Jour- nal, 163(2), 161-165. Hwang, S. W., & Gottlieb, J. L. (1999). Drug access among homeless men in To- ronto. Canadian Medical Association Journal, 160(7), 1021. Jencks, C. (1994). The homeless. Cambridge, MA: Harvard University Press. Kinchen, K., & Wright, J. D. (1991). Hypertension management in health care for the homeless clinics: Results from a survey. American Journal of Public Health, 81(9), 1163-1165. Koegel, P., Melamid, E., & Burnam, A. (1995). Childhood risk factors for home- lessness among homeless adults. American Journal of Public Health, 85(12), 1642-1649. Kufeldt, K. (1991). Social policy and runaways. Journal of Health & Social Policy, 2(4), 37-49. Kuhn, R., & Culhane, D. P. (1998). Applying cluster analysis to test a typology of homelessness by pattern of shelter utilization: Results from the analysis of administrative data. American Journal of Community Psychology, 26(2), 207- 232. Kushel, M. B., Perry, S., Bangsberg, D., Clark, R., & Moss, A. R. (2002). Emergen- cy department use among the homeless and marginally housed: Results from a community-based study. American Journal of Public Health, 92(5), 778- 784. Kushel, M. B., Vittinghoff, E., & Haas, J. S. (2001). Factors associated with the health care utilization of homeless persons. The Journal of the American Medi- cal Association, 285(2), 200-206. Lee, J., Gaetz, S., & Goettler, F. (1994). The oral health of Toronto?s street youth. Journal of the Canadian Dental Association, 60(6), 545-548. Lehman, A. F., Dixon, L. B., Kernan, E., DeForge, B. R., & Postrado, L. T. (1997). A randomized trial of assertive community treatment for homeless persons with severe mental illness. Archives of General Psychiatry, 54(11), 1038-1043. Lehman, A. F., Kernan, E., DeForge, B. R., & Dixon L. (1995). Effects of homeless- ness on the quality of life of persons with severe mental illness. Psychiatric Services, 46(9), 922-926. Manzon, L., Rosario, M., & Rekart, M. L. (1992). HIV seroprevalence among street involved Canadians in Vancouver. AIDS Education and Prevention, (Suppl.), 86-89. Martell, J. V., Seitz, R. S., Harada, J. K., Kobayashi, J., Sasaki, V. K., & Wong, C. (1992). Hospitalization in an urban homeless population: The Honolulu Ur- ban Homeless Project. Annals of Internal Medicine, 116(4), 299-303. May, K. M., & Evans, G. G. (1994). Health education for homeless populations. Journal of Community Health Nursing, 11, 229-237. Metraux, S., Marcus, S. C., & Culhane, D. P. (2003). The New York-New York housing initiative and use of public shelters by persons with severe mental illness. Psychiatric Services, 54(1), 67-71. Nazar-Stewart, V., & Nolan, C. M. (1992). Results of a directly observed intermit- tent isoniazid preventive therapy program in a shelter for homeless men. American Review of Respiratory Disease, 146(1), 57-60. Nolan, C. M, Elarth, A. M, Barr, H., Saeed, A. M., & Risser, D. R. (1991). An out- break of tuberculosis in a shelter for homeless men: A description of its evo- lution and control. American Review of Respiratory Disease, 143(2), 257-261. Novac, S., Serge, L., Eberle, M., & Brown, J. (2002). On her own: Young w omen and homelessness in Canada. Ottawa: Status of Women Canada. O?Flaherty, B. (1996). Making room: The economics of homelessness. Cambridge, MA: Harvard University Press. Pablos-Mendez, A., Knirsch, C. A., Barr, R. G., Lerner, B. H., & Frieden, T. R. (1997). Nonadherence in tuberculosis treatment: Predictors and conse- quences in New York City. The American Journal of Medicine, 102(2), 164-170. Pizem, P., Massicotte, P., Vincent, J. R., & Barolet, R. Y. (1994). The state of oral and dental health of the homeless and vagrant population of Montreal. Journal of the Canadian Dental Association, 60(12), 1061-1065. Provincial Task Force on Homelessness (1998). Report of the Provincial Task Force on Homelessness. Ottawa: Author. Quantz, D., & Prankish, J. (2002). Homelessness research in the Greater Vancouver Regional District: An agenda for moving forward. Vancouver: University of British Columbia. Roy, E., Boivin, J. F., Haley, N., Lemire, N. (1998a). Mortality among street youth. The Lancet, 352(9121), 32. Roy, E., Haley, N., Eemire, N., Boivin, J. F, Leclerc, P., & Vincelette, J. (1999). He- patitis B virus infection among street youths in Montreal. Canadian Medical Association Journal, 161(6), 689-693. Roy, E., Haley, N., Leclerc, P., Cedras, L., Biais, L., & Boivin, J. F. (2003). Drug injection among street youths in Montreal: Predictors of initiation. Journal of Urban Health, 80(1), 92-105. Roy, E., Lemire, N., Haley, N., Boivin, J. F., Frappier, J. Y., & Claessens, C. (1998b). Injection drug use among street youth: A dynamic process. Cana- dian Journal of Public Health, 89(4), 239-240. Salit, S. A., Kuhn, E. M., Hartz, A. J., Vu, J. M., & Mosso, A. L. (1998). Hospitaliza- tion costs associated with homelessness in New York City. The New England Journal of Medicine, 338(24), 1734-1740. Schwartz, S., & Carpenter, K. M. (1999). The right answer for the wrong question: Consequences of type III error for public health research. American Journal of Public Health, 89(8), 1175-1180. Shinn, M., Weitzman, B. C., Stojanovic, D., Knickman, J. R., Jimenez, L., Duchon, L., James, S., & Krantz, D. H. (1998). Predictors of homelessness among families in New York City: From shelter request to housing stability. Ameri- can Journal of Public Health, 88(11), 1651-1657. Smart, R. G., & Adlaf, E. M. (1991). Substance use and problems among Toronto street youth. British Journal of Addiction, 86(8), 999-1010. Statistics Canada (2002). 2 001 census: Analysis series, collective dwellings. Ottawa: Author. Stratigos, A. J., & Katsambas, A. D. (2002). Medical and cutaneous disorders as- sociated with homelessness. Skinmed, 2(3), 168-172. Stroik, S. M. (2004). Biennial count of homeless persons in Calgary: Enumerated in emergency and transitional facilities. Calgary: City of Calgary, Service Agen- cies, On the Streets. Susser, E., Moore, R., & Link, B. (1993). Risk factors for homelessness. Epidemio- logical Reviews, 15(2), 546-556. Wang, E. E., King, S., Goldberg, E., Bock, B., Milner, R., & Read, S. (1991). Hepati- tis B and Human Immunodeficiency Virus infection in street youths in To- ronto, Canada. The Pediatric Infectious Disease Journal, 10(2), 130-133. Wasylenki, D. A, Goering, P. N., Lemire, D., & Lindsey, S. (1993). The hostel out- reach program: Assertive case management for homeless mentally ill per- sons. Hospital & Community Psychiatry , 44(9), 848-853. Woodward, J., Eberle, M., Kraus, D., & Graves, J. (2002). Research project on home- lessness in Greater Vancouver. Vancouver: Greater Vancouver Regional Dis- trict. Zerger, S. (2002). Substance abuse treatment: What works for homeless people? : A re- view of the literature. Nashville: National Health Care for the Homeless Council. Chapter 2.2 The Health of Toronto?s Homeless Population THE STREET HEALTH REP ORT , 2 0 0 7 Homelessness is a devastating social problem in Toronto. In 2002, about 32,000 different people slept in one of the city?s homeless shelters (City of Toronto, 2003). In 2006, about 6,500 individuals stayed in a shelter on any given night (Shapcott, 2006). In 1998, the City of Toronto endorsed a declaration acknowledging that homelessness is a national disaster. Homeless people have much poorer health and higher mortality rates than the general population, and often experience difficulties ob- taining the health care and social services they need. They are also largely excluded from broad-based government health and census sur- veys, which often depend on people having an address or telephone number In 1992, Street Health, a community-based health agency serving homeless people in Toronto, decided to conduct its own study to explore the health status of homeless people and their ability to access the health care system (Ambrosio et al., 1992). It was the first report of its kind in North America and continues to be used today. . Even when these surveys reach homeless people, they do not address the unique circumstances of this group. When the 1992 Street Health Report was published, Toronto was emerging from an economic downturn. Political and business leaders promised that economic good times would bring rewards for everyone. However, by 2007 the research wing of one of Canada?s largest banks was reporting that social inequity and poverty were on the rise in the city (TD Economics, 2007). During the 1990s, the richest 10 percent of Torontonians saw their family income increase by about 8 percent while the poorest 10 percent had a drop of almost the same amount. Between 2001 and 2005, the bottom 20 percent of Canadian families saw an out- right decline in their income (Shapcott, 2007). In the years since the 1992 Street Health Report was published, homelessness and housing insecurity have increased in Toronto. The nightly count of people sleeping in homeless shelters has more than tri- pled. This increase reflects funding and program cuts at the federal and provincial levels, coupled with the downloading of responsibility for social programs to the province and city. Social assistance rates were cut and have not been fully restored, rents have risen, and very little social housing has been built. At the same time, new health issues have emerged in the homeless community. Street Health decided it was time to conduct another comprehensive study to fill a gap in current knowl- edge about the health status of homeless people in Toronto, find out how the health of homeless people had changed in 15 years, and create new evidence on which to ground our advocacy efforts and those of other community groups. The 2007 Street Health Survey The second Street Health Survey was conducted between November 2006 and February 2007. We surveyed a random sample of 368 homeless adults at meal programs and shelters across downtown Toronto about their health and access to health care. Homelessness was defined as: hav- ing stayed in a shelter, outdoors or in a public space, or with a friend or relative for 10 or more days in the 30 days prior to being surveyed. Of those interviewed, 73 percent identified as male, 26 percent as female, and 1 percent as transgender/transsexual. The average age of people in- terviewed was 42 years, and participants ranged in age from 19 to 66 years. More than three-quarters (77 percent) were between the ages of 25 and 49; 3 percent were under 24 and 20 percent were over 50. Of the sample, 63 percent identified solely as Caucasian; 37 percent as non-Caucasian. Aboriginal people made up a much higher percentage of our sample (15 percent) compared with the percentage they represent in the general population of Toronto (about 0.5 percent in the 2001 Cen- sus). Thirty-two percent were born in Toronto and an additional 45 per- cent were born in Canada outside of Toronto. Immigrants were under- represented in our sample (23 percent) compared to the general population of Toronto, where 49 percent were born outside the country (Statistics Canada, 2005).1 Seventy-two percent of our sample had lived in Toronto for 10 years or longer. The people interviewed in our study had lived in Toronto for a long time. They were typically not newcomers to the city who had just arrived and were getting settled. Only 15 percent had lived in Toronto less than five years. Of those who were not born in Canada, 53 per- cent had Canadian citizenship, 36 percent were landed immigrants, 5 percent had refugee status and 5 percent had temporary or no status. Patterns of Homelessness You can?t get out of poverty, no matter how you try. Nothing works to- gether. They have systems but they don?t work together. Believe me, I have tried every possible way, but you can?t. For three years I?ve been go- ing around in a circle. And I can?t get out of it. I?m very resourceful, I?m intelligent, and I?m not lazy. I?m sure people give up, but I keep going. (Survey Respondent) For survey participants, homelessness was not, on average, a short-term crisis. People in our survey had been homeless an average of 4.7 years. The length of time that participants in our survey had been homeless throughout their lives ranged from two weeks to 50 years. Seventy-eight percent had been homeless for one year or longer and 34 percent had been homeless for five years or longer. 1 Our main source of information on the general population is the Canadian Community Health Survey (CCHS) Cycle 3.1, Public Use Microdata File (Sta- tistics Canada, 2005), which contains anonymized data. All computations on these microdata were prepared by Street Health and the responsibility for the use and interpretation of these data is entirely that of the authors. Main reasons respondents gave for becoming homeless % Economic reasons (cost of rent, low income, unemployment) 52% Unsafe or poor living conditions 31% Eviction or conflict with landlord 25% Respondent?s own drug or alcohol use 23% Relationship or family breakdown 20% Institutionalization (went to hospital, substance treatment program, jail) 13% Neighbourhood was inappropriate or too isolated 3% Lack of support to keep housing 2% Survey participants were also asked the two main reasons prevent- ing them from finding and maintaining housing right now: Main reasons respondents gave for remaining homeless % Economic reasons (cost of rent, low income, unemployment) 78% Mental and physical health conditions 33% Lack of suitable housing (unsafe or poor living conditions, bad landlords) 24% Waiting list too long 11% Lack of adequate support to find and keep housing 10% Discrimination (against welfare recipients, people with criminal records) 7% Lack of person identification 6% Our findings are consistent with other studies, which have also found that poverty is the leading cause of homelessness in Canada. Pov- erty is a concern for many Canadians, and 49 percent of the population feel they are always just one or two paycheques away from being poor (Canadian Centre for Policy Alternatives, 2006). Official estimates from 2001 suggest that 1.7 million Canadian households were at risk of home- lessness (Engeland, 2004). Key changes to social policies in recent years have a direct connec- tion to some of the most common reasons that people became homeless: ? A Shortage of Social Housing: From the 1960s until 1993, roughly 20,000 units of social housing were built each year with the help of government funding, most of which came from the federal govern- ment. In 1993 the federal government withdrew its funding of social housing, and in 1995 the province of Ontario did the same. As a re- sult, throughout most of the 1990s, very little affordable, supportive housing was built. ? Cuts to Social Assistance: Throughout the 1990s, the federal govern- ment made major cuts in social program funding for the provinces. In real dollars, Ontario welfare benefits are now roughly half what they were in 1995 and disability benefits are roughly 22 percent less (Task Force on Modernizing Income Security, 2006). It has been es- timated that the 21.6 percent cut to social welfare benefits in Ontario pushed 67,000 families out of their rental housing (Falvo, 2003). ? Easing of Rent Controls: Since 1998, the City of Toronto has lost 85 percent of its one-bedroom apartments renting at or below $700 a month (City of Toronto, Shelter, Support & Housing Administration, 2006). Rents have gone up at a faster rate than incomes. In the late 1990s to early 2000s, rent increases averaged 5 percent higher than wages. While average rents in Toronto grew by 30 percent between 1997 and 2002, from $751 to $976, r eal wages (adjusted for inflation) decreased for those earning minimum wage (City of Toronto, 2003). ? Decreased Tenant Protections: New laws were introduced in On- tario in the 1990s that decreased tenant protection and made eviction easier. The Ontario Rental Housing Tribunal received over 30,000 eviction applications by landlords to terminate tenancies in 2005; of these 86 percent were because of rental arrears (City of Toronto, Shelter, Support & Housing Administration, 2006). It is likely that many of the people in our survey were evicted because they couldn?t afford the rent. Many of the reasons homeless people in our survey gave for why they are unable to find or maintain housing point to the lack of adequate help to find housing and the lack of subsidized and supportive housing options that are available: 47 percent said they were not currently getting help to find housing and 44 percent were on the waiting list for social housing. As of December 31, 2006, there were 67,083 households in Toronto on the waiting list for social housing (Housing Connections, 2006). The length of time on the social housing waiting list for survey respondents ranged from 1 day to 20 years. Fifty-two percent have been on the social housing wait list for six months or less. This large portion of people who have been on the wait list a very short time is probably a reflection of stepped-up efforts by the City of Toronto?s housing workers, as well as its new plan to address homelessness, which requires every homeless person they work with to apply for social housing. Among the 48 percent who have been on the wait list for longer than 6 months, the average wait time was 4.6 years. Income, Benefits, and Money Management The homeless people we surveyed report extremely low incomes: 36 per- cent live on $200 a month or less. Formal employment was a source of income for 20 percent of survey respondents: 11 percent did casual or piece work; 5 percent did part-time work; and 4 percent did full-time work. Informal work includes panhandling, sex work, selling scrap metal or bottles and other street-involved work: 19 percent of respondents cited income from informal employment. Panhandling was the most common type, cited as a source of income by 9 percent of survey respon- dents. However, it is likely that informal work was underreported be- cause many forms are illegal or stigmatized. In terms of access to government income supports, 52 percent re- ceived no major government benefits, 27 percent received support through Ontario Works (OW)2, 16 percent through Ontario Disability Support Program (ODSP),3 Although 74 percent of the people we surveyed have at least one se- rious physical health condition, only 22 percent of those with serious illnesses are getting either ODSP or federal disability benefits. Thirty- and 5 percent received other government benefits such as government pensions, federal disability benefits, unem- ployment insurance, and workers? compensation benefits. 2 In Ontario, government welfare benefits are obtained through the Ontario Works (OW) program. OW benefits are for people who need money because they are unable to find work or are temporarily unable to work. 3 The Ontario Disability Support Program (ODSP) is the provincial government program that offers long-term disability benefits to people with serious disabilities who have lit- tle or no other way to support themselves. The basic rate for a single person on Ontario Works is approximately $548 per month. For ODSP, the rate is $979. eight percent of all respondents felt that they were eligible for ODSP, but were not receiving it, for various reasons: 50 percent had not applied; 19 percent had their application rejected; 17 percent could not complete the application process; and 12 percent had applications still in process. What we heard about ODSP is consistent with a separate study conducted by Street Health, which specifically examined the barriers homeless people face when attempting to secure disability benefits (Street Health, 2006). The study found that homeless people with dis- abilities cannot navigate the overall ODSP application process without help, due to its complexity. Certain disabilities such as mental illness, developmental disabilities and learning disabilities make this system even more difficult to navigate. In the study, participants were provided with intensive, one-on-one support with all aspects of the ODSP applica- tion process and their related income, housing and legal needs. As a re- sult, 93 percent of participants secured ODSP benefits and 100 percent of those were then able to get housing. Many survey respondents cited smaller streams of government benefits that provide important, but inadequate, supplemental support. Only 11 percent of respondents said they received the GST credit, de- spite the fact that the vast majority live on extremely low incomes and should therefore be eligible. This poor access to the GST credit is likely because of the barriers inherent in a tax return-based benefit, such as not having the resources to file your income tax.4 Thirty-four percent of respondents said they received Personal Needs Allowance (PNA) benefits. PNA is a stipend given to people stay- ing in shelters to help meet incidental needs other than those provided for by shelters and was worth $3.90 a day at the time of the survey, or $109 to $120 a month. People cannot receive other social assistance ben e- fits at the same time that they are receiving PNA. Most homeless people do not use bank services. Requiring multiple forms of personal identification to open a bank account is a significant barrier for people who are homeless. Therefore, 60 percent of those in 4 The amount of the GST credit is based on factors such as net income and number of children. A single adult earning $400 per month could expect to receive approxi mately $60, four times a year, in GST credits. our survey used cheque-cashing services. These services typically charge fees of $2.50 per cheque, plus an additional 3 percent of the total cheque amount. This means, on a welfare cheque of $548, the service takes about $19. Because these companies do not provide savings accounts, ind i- viduals have no choice but to receive the entire value of their cheque in cash. This makes money management difficult and leaves people at risk of being robbed. The Daily Lives of Homeless People It's hard to want to stay healthy wh en you have to walk round the streets in the cold, rain, snow, broke. It?s part of life. It?s something I live with. (Survey Respondent) Almost all (96 percent) of the homeless people in our survey reported using shelters at least once in the past year. Of those who use shelters, 55 percent said that they had been unable to get a shelter bed at least once in the past year, on average 20 times, and of those, 74 percent said that this had happened at least once in winter months. Places respondents stayed overnight in the past month % Shelter 88% Outside (e.g. parks, streets, bus stops) 32% Friend?s or relative?s place 26% Hotel, motel, rooming or boarding house 10% Hospital or treatment program 8% Car or abandoned building 7% Place of business (e.g. coffee shop, laundromat) 6% Jail 4% Shelters in Toronto range from approximately 25 to 750 beds. While many are doing their best with limited resources, a typical Toronto shel- ter is still a crowded place full of bunk beds, with a shared washroom for dozens of people, and few food choices. Some shelters have maximum lengths of stay, forcing people to be constantly on the move. Some re- quire people to leave early in the morning, leaving people with no place to rest during the day. Out of the Cold programs are meal and shelter services run by faith- based groups and community centres across Toronto during the winter (mid November to mid April). These programs are generally volunteer- run and often operate only one day a week. The shelter that these pro- grams provide is sometimes just a mat and a blanket on the floor of a church basement. Out of the Cold Programs provide accommodation for approximately 200 people per night. During the time that this study was conducted, three shelters in the downtown core closed. Some homeless people avoid the shelter system altogether. We asked people who had not stayed in a shelter in the last 10 days why they chose not to. In addition to providing a place to sleep, shelters are also an important source of food. Fewer shelter beds also mean fewer meals for homeless people. Common reasons respondents gave for avoiding shelters % Bed bugs 34% Crowded conditions 31% Fear of getting sick 23% Fear of violence 20% Fear of theft 15% Bed bugs have become a common problem for homeless people in Toronto. Bed bugs hide in cracks and crevices in beds, flooring and walls. Their bites can cause clusters of small but extremely itchy red bumps. Although bed bugs are not known to transmit infectious dis- eases, they cause physical discomfort and emotional distress (Hwang et al., 2005). Sleep You can?t go to sleep because you don?t know what?s going to happen from minute to minute. So you just keep staying up and staying up and staying up. And I noticed that physically ? I had clumps of hair coming out ? and memo ry loss. I don?t know if it?s just exhaustion or nerves. But that?s how it?s affected me. (Survey Respondent) Survey respondents reported low levels of sleep. Forty-six percent got an average of six hours or less per night. The most common problem was too much noise or light. Other reasons included being woken up by oth- ers, crowded conditions, the cold, bed bugs, and unclean conditions. Others cited nightmares, bad nerves, pain, and other physical health problems as reasons they could not get enough sleep. Lack of sleep can have many important impacts on physical health, psychological well-being, and energy levels. More than half of respon- dents (54 percent) said that in the past month, they had been so tired that they did not have the energy to walk one block or do light physical work. Sleep disturbances also contribute to the development, or increase the severity, of various medial and psychiatric conditions, including heart attacks and depression (Zee & Turek, 2006). Hygiene When survey respondents were asked about some essential daily hy- giene activities, 32 percent said they sometimes or usually had difficulty finding a place to use the washroom; 25 percent said they sometimes or usually had difficulty finding a place to bathe; and 41 percent said they sometimes or usually had difficulty getting their clothes washed. Hygiene is an important part of overall health and is particularly important for some health issues. Getting rid of bed bugs requires expos- ing them to very high or very low temperatures. People are usually ad- vised to put their clothes and bedding in the dryer at a high temperature if they have a bed bug problem. Almost half of the people we inter- viewed would have difficulty following this advice. Hunger and Nutrition I don?t eat at all, some days. Sometimes the food is not available, you know, especially on the weekends ? it?s hard. A lot of places ain?t open as frequently as they are through the week. So, I just do whatever I can. I see the health bus, I get vitamins. (Survey Respondent) Sixty-nine percent of homeless people in our survey had experienced hunger at least one day per week in the past three months because they could not get enough food to eat. Homeless people rely heavily on meal programs, because they do not have places to store or cook food, and 96 percent of respondents said that they regularly used meal programs at a shelter, drop-in, or other organization. Even so, homeless people are clearly not getting their food needs met by these programs: 58 percent reported that in the past three months they had still been hungry after going to a meal program. Not including Out of the Cold Programs (which operate only in the winter months), there are approximately 80 programs in Toronto that provide meals to homeless people outside of the shelter system. Of these programs, about two-thirds provide only a single meal a day, and more than three-quarters are closed on Saturday or Sunday. An analysis of the meals served in a sub-sample of 18 of these programs found that the av- erage energy content of meals served was half of what a healthy adult would require for minimal physical activity during the day (Tarasuk, 2007.) Many homeless people in our survey had special dietary needs, mainly for health reasons. Of the 33 percent of respondents who said they were supposed to be following a special diet, 53 percent said they were able to follow it less than once a week. Through the Ontario gov- ernment?s Special Diet Supplement, people receiving social assistance are eligible for additional income (up to a maximum of $250) if they can provide evidence that they have a medical condition that requires a spe- cial diet, but 70 percent of respondents who were supposed to follow a special diet were not receiving the Special Diet Supplement, for various reasons. Common reasons respondents gave for not receiving Special Diet supplement % had not applied (because they did not know about it, did not know how to apply, or could not navigate the application) 55% had been cut off from the supplement due to changing eligibility criteria 14% had not been able to get the form filled out by a health care provider 10% had applied, but had been turned down 9% could not get the supplement because they were not receiving social assistance 9% In 2006, new regulations by the provincial government made access to the Special Diet Supplement even more difficult, and reduced the amount that many people were already receiving. Everyone receiving the supplement was required to re-submit their applications on new, more restrictive eligibility forms. Additional diet income is now tied di- rectly to specific medical conditions and, in some cases, to how sick you are as a result of the condition. For example, if you have diabetes, you are eligible to receive an additional $42, while the amount someone re- ceives if they HIV/AIDS varies between $75 and $240; depending on how much weight they have lost. Of the 9 percent of respondents who were getting the Special Diet Supplement at the time of the interview, more than half reported that the amount they receive had been reduced in the 12 months prior to the sur- vey, by an average amount of $ 147. Social Isolation Social exclusion has a major negative impact on health, increasing one?s risk of disability, illness, and addiction (Marmot & Wilkinson, 2003). Homeless people in our survey experience low levels of social support and high levels of social isolation: 37 percent said they had no one to help them in an emotional crisis and 38 percent said that they often feel very lonely or remote from other people. Poverty creates social exclusion because it denies people access to decent housing, education, and other factors that are important to full and equal participation in society. Discrimination, hostility, unemploy- ment, and stigmatization also contribute to social exclusion and are part of the daily reality of the homeless people we interviewed. Injury and Violence Without their own private or safe spaces to go and stay, many homeless people are forced to live much of their lives in public, putting them at greater risk for injuries and accidents. Almost one in ten (9 percent) of survey respondents had been hit by a car, truck, public transit vehicle, or bicycle in the past year. Rates of physical violence are also very high. Thirty-five percent of homeless people in our survey had been physically assaulted in the past year. Of those, 68 percent were assaulted more than once, on average six times. This is much higher than among the general population of Toronto, where less than 1 percent reported a physical as- sault to the police in 2005 (Toronto Police Service, 2006). Who respondents reported being assaulted by the past year % Stranger 56% Acquaintance 38% Police 35% Another shelter resident 27% Partner or spouse 21% Shelter staff 6% Three-quarters of respondents in our survey who had been as- saulted by police said they did not lodge a formal complaint. The main reasons cited were related to fear of repercussions (46 percent), and feel- ing that it would not accomplish anything (46 percent). Respondents were also asked if they had been sexually assaulted or raped in the past year, defined in our survey as unwanted touching and/or sexual intercourse. Of the entire group, 7 percent said they had been sexually assaulted in the past year, but this statistic was higher for women, 21 percent of whom had been sexually assaulted in the past year. Even though these rates are extremely high, it is likely that sexual assault was under-reported in our survey, due to the personal nature of the issue and the stigma that surrounds it. General Health & Well Being It is widely recognized that homeless people have much poorer health than the general population. Other studies have found that people living in poverty are more likely to die from certain diseases, including cancer, diabetes and respiratory diseases, and particularly cardiovascular dis- ease (Raphael, 2002). Our findings on health and well-being speak over- whelmingly to the overall poor physical and mental health of homeless people. When asked to think about the amount of stress in their lives, 44 percent of respondents said that most days were quite a bit or extremely stressful. In comparison, 24% of people in Toronto reported the same (Statistics Canada, 2005). Stress has an important impact on health and well-being. High levels of stress can contribute to conditions such as high blood pressure, heart disease and stomach or intestinal ulcers. Chronic stress over long periods of time compromises the immune sys- tem, making people more susceptible to a range of other health condi- tions. Self-rated health and mental health: Homeless people in our survey com- pared with the general population Street Health Survey General population Physical health Very good or excellent* 29% 61% Good 29% 30% Fair or poor* 40% 9% Mental health Very good or excellent* 33% 74% Good* 27% 21% Fair or poor* 38% 5% *statistically significant difference Source of information on general population: Statistics Canada. Canadian Community Health Survey (CCHS) Cycle 3.1 (2005). Public Use Microdata File. Pain Regular experiences of pain were common among homeless people. Forty-one percent of respondents said that they were usually in some pain or discomfort. Fourteen percent of all respondents said that this pain is severe. These high levels of pain and discomfort suggest that many people may have disabilities and medical conditions that are not acknowledged, diagnosed, or treated. Experiences of pain can also lead to low energy levels, which in turn limit people?s ability to care for them- selves and participate economically and socially in the community. Pain also affects one?s ability to sleep. Almost one third (30 percent) of re- spondents said that they found it hard to sleep because of pain or dis- comfort. Physical Health Conditions Of the homeless people we interviewed, 74 percent had at least one seri- ous physical health condition,5 The homeless population carries a disproportionate burden of many serious health conditions compared to the general population. In the ar- eas for which we have comparable data (Statistics Canada, 2005), our results show that homeless people are: and 52 percent had two or more. We also found that for people without any serious health conditions, the average time homeless was 3.7 years, whereas the average length of time home- less for people with at least one serious health condition was 5.1 years. The significant difference between these two averages suggests that be- ing homeless for a longer period increases one?s likelihood of serious illness. ? 20 times as likely to have epilepsy as members of the general popula- tion; ? 5 times as likely to have heart disease; ? 4 times as likely to have cancer; ? 3? times as likely to have asthma; ? 3 times as likely to have arthritis or rheumatism; ? twice as likely to have diabetes. Moreover, we would expect many of these conditions to be more common among older adults, yet the average age of survey respondents was 42, and the oldest person interviewed was 66. 5 A ?serious physical health condition? was defined as any of 22 serious conditions, including: cardiovascular and respiratory diseases, hepatitis and other liver diseases, gastrointestinal ulcers, diabetes, anemia, epilepsy, cancer, and HIV/AIDS. Chronic or ongoing physical health condition Street Health Survey General Population Arthritis or rheumatism* 43% 14%** Allergies other than food allergies* 33% 24%** Migraines* 30% 11%** Liver disease* 26% 10%*** in Canada Hepatitis C* 23% 0.8%? in Canada Problem walking, lost limb, other physical handicap 23% n.a. Asthma* 21% 6%** Heart disease* 20% 4%** High blood pressure* 17% 13%** Chronic obstructive pulmonary disease 17% 1%** Stomach or intestinal ulcers 15% 2%** Skin disease (e.g. eczema, psoriasis) 13% n.a. Angina* 12% 2%?? in Ontario Anemia 11% n.a. Diabetes 9% 4%** Heart attack in lifetime 7% 2%?? in Ontario Inactive or latent tuberculosis 7% n.a. Epilepsy* 6% 0.3%??? in Canada Fetal alcohol spectrum disorder 5% 1% Stroke in lifetime 4% n.a. Hepatitis B* 4% 0.7-0.9%? in Canada Cancer* 4% 1%** Congestive heart failure* 3% 1%?? in Ontario HIV positive* 2% .006%?? AIDS 1.1% n.a. Unless otherwise noted, comparisons are to the general population of Toronto. * Statistically significant. ** Source: Statistics Canada, 2001. *** Source: Canadian Liver Foundation, 2006. ? Source: Public Health Agency of Canada, 2003. ?? Source: Chow et al., 2005. ??? Source: Statistics Canada, 2005. ? Source: Public Health Agency of Canada, 2007. ?? Estimated prevalence of 15,300 in Toronto. Source: Remis et al., 2006. Emergent Health Issues There are several health conditions that have emerged as important health issues among the homeless population in recent years. Tuberculosis (TB) is a contagious disease that had almost disap- peared in Canada, but that has re-emerged in recent years. TB bacteria commonly attack the lungs but can infect other parts of the body. People can have active or inactive TB. Inactive TB means that you have the TB bacteria in your body but it is not making you sick. Inactive TB can be- come active TB if the immune system is somehow damaged. TB is a ma- jor cause of death for people who also have HIV. Seven percent of the people we surveyed said they had inactive TB. It is likely that the re- ported rate of inactive TB among homeless people in our sample is an underestimation, as many respondents may not know they have it. While the majority of active TB cases in Toronto are among people who have travelled or lived in areas where TB is common, one-third of recent cases were among homeless and underhoused people living in shelters or rooming houses (Toronto Public Health, 2007). Crowded con- ditions in these living situations put this population at high risk for in- fection. Although TB is preventable and curable, and despite recent scaled up efforts by Toronto Public Health, it is still not easy for home- less people to get tested for TB or to access treatment. Hepatitis C is a viral infection that attacks the liver and is transmit- ted through blood to blood contact. While 23 percent of our survey sam- ple reported having Hepatitis C, it is likely that this number is even higher. Hepatitis C progresses slowly and most infected people do not experience symptoms for many years. The Ontario Ministry of Health estimates that one-third of people living with Hepatitis C do not realize they are infected. Despite a high need among this group, homeless peo- ple experience major barriers to accessing treatment or even acquiring basic information about the disease. Without education, many people are transmitting the disease unwittingly. Hepatitis C can be effectively treated, but the treatment is difficult and requires stability and support. Treatment requires following a strict schedule of medication and monitoring by a physician for at least six months. The side effects can be debilitating, and include depression, hair loss, flu-like symptoms, and nausea. Many health care providers are un- willing or unable to provide the extensive support that homeless people need to successfully undergo treatment. Without treatment, Hepatitis C can cause liver disease, including cirrhosis and cancer. Without adequate shelter, and nutritious food, homeless people are even more susceptible to some of these negative outcomes. Further, it is estimated that in On- tario, 25 percent of people with HIV also have Hepatitis C (Public Health Agency of Canada, 2001). HIV and Hepatitis C co-infection is problem- atic, because each disease makes the other worse and it is hard to treat both simultaneously. HIV/AIDS has disproportionately affected homeless people relative to the general population. The prevalence of HIV is over 300 times higher among homeless people than in the general population in To- ronto. It is possible that this condition was under-reported by survey respondents, due to the stigma attached to the disease and because some respondents may not know their HIV status. Homeless people with HIV are at extremely high risk for many other medical conditions. In addition to having a compromised immune system due to HIV, homeless people tend to have their immune systems even further weakened by the harsh- ness of their daily lives, which includes fatigue, poor nutrition and high levels of stress. In addition, homeless people are regularly exposed to countless communicable diseases and infections in crowded spaces such as shelters. The Impact of Living Conditions on Homeless People?s Health In addition to poverty, stress and social isolation, key aspects of home- less people?s unique living situation that affect their health are: ? Food: Homeless people lack control over the food they eat, and lack access to healthy food, which may contribute to, or make worse, conditions such as diabetes and stomach ulcers. ? Violence and Injury: Homeless people are more likely to be injured or assaulted, which often includes head injuries. Head injuries can lead to seizures. Violence also has a broad range of negative physical and psychological effects. ? Density and Crowding: Crowded conditions in shelters put homeless people at risk for infectious diseases like the flu and TB, as well as problems like lice, scabies and bed bugs. ? Exposure to the Elements: Homeless people are also far more exposed to the urban environment and the elements than the average person. Many homeless people spend a major part of the day outside, expos- ing them to damp, cold, extreme heat and pollution. This prolonged exposure may put homeless people at higher risk for arthritis, pneumonia, allergies and asthma. Foot problems among homeless people are common because so many homeless people spend a large part of their day walking or standing, and because homeless people often have to spend the day with cold and wet feet. ? Heat: Climate change is starting to have a dramatic impact on home- less and poorly housed people. In 2005, Toronto?s Medical Officer of Health reported that more Torontonians are dying prematurely of heat-related causes in the summer than of cold-related causes in the winter (McKeown, 2006). Homeless and poorly-housed people, who have very few options to escape the heat, are among those at greatest risk for heat-related illness. The number of smog and extreme heat days reached an all-time high in 2005. Rising temperatures due to climate change threaten to make this problem even worse. Mental Health Mental ?illness? does not cause homelessness; poverty does. A 1998 To- ronto study that examined the societal and personal factors that precipi- tate homelessness concluded that mental illness cannot be seen as a pri- mary pathway to homelessness (Tolomiczenko & Goering, 1998). Their report argues that broader systemic factors need to be taken into account and uses an analogy of ?musical chairs.? As chairs (that is, jobs and af- fordable housing) become scarce, it is not surprising to find people with mental and physical health problems among those without a chair. You get a sense of despair; your self worth goes to hell. (Survey Respondent) Suicidal thoughts were significantly more frequent among the res- pondents in our survey than among the general population in Toronto, where 7 percent reported having suicidal thoughts in their lifetime (Sta- tistics Canada, 2004). The high levels of depression, anxiety and suicidal ideation in our sample reflects the harsh reality of homeless people?s dai- ly lives, and the lack of hope that many homeless people feel. Respondents were asked if they had ever been given a diagnosis for a mental health problem by a doctor or psychiatrist. Thirty-five percent of our sample has received such a diagnosis. The table below shows the prevalence of the most common mental health diagnoses in our sample, compared with that of the general population in Canada. Street Health Survey General population** Depression* 17 % 8 % Anxiety* 11 % 1 % Bipolar* 8 % 1 % Schizophrenia* 5 % 1 % Post-traumatic stress disorder 5 % n/a * significant difference ** Source: Health Canada, 2002. Not reflected in these numbers is the reality that many people with mental illness are initially misdiagnosed and that determining a diagno- sis and a treatment plan is often an ongoing process, negotiated between specialists and clients. Many homeless people, because they do not have stable health care, are unable to go through this process and often live with misdiagnoses and inappropriate treatments. Mental health problems affect people of all income levels. It is esti- mated that one in five Canadians will experience mental illness during his or her lifetime (Health Canada, 2002). Mental health problems do not directly cause homelessness. People with mental health issues become homeless when they lack income stability and appropriate supports. Many of the factors that compromise mental health, such as instability, social isolation and violence, are also part of the daily reality of home- lessness. Many people experience mental health problems, or have exist- ing problems become worse, only after they become homeless. While for some people, mental health issues may be one of the fac- tors that contribute to becoming homeless, it is likely just one of many. Although many survey participants experience mental health issues, very few (5 percent) cited mental health issues as the reason they lost housing or the reason they were unable to find or maintain housing. Ad- diction issues came up as a more prominent reason for losing or not be- ing able to get housing (cited by 23 percent for both questions). Our study did not explore the prevalence of concurrent disorders, the term used when people have a combined mental health and sub- stance use problem. However, other studies estimate that 30 percent of people diagnosed with a mental health disorder also have a substance use disorder at some point in their lives (CAMH, 2006). Having a con- current disorder can make it even more difficult to access treatment. Many mental health services refuse treatment to a person with an active drug or alcohol addiction and some addictions services will not treat people for substance use problems until their mental health problem is treated. Learning Disabilities Learning disabilities are disorders that affect the acquisition, organiza- tion, retention, understanding, or use of verbal or nonverbal information. These are lifelong disorders that can affect self-esteem, work, and rela- tionships. Difficulties faced by adults with learning disabilities may in- clude finding or keeping a job, time management, budgeting and manag- ing money. Homeless people report significantly higher rates of learning disa- bility. Sixteen percent (16%) of our sample said they had been dia gnosed with a learning disability, compared to only 2% of the gen eral popula- tion in Toronto (Statistics Canada, 2005). Substance Use Many homeless people smoke cigarettes: 87 percent of respondents said they currently smoke cigarettes, compared to 18 percent of the general population of Toronto (Statistics Canada, 2005). The proportion of homeless people who had consumed alcohol in the last year (77 percent) is almost identical to that of the general popula- tion of Toronto, 70 percent of whom reported using alcohol at least once in the past year (Statistics Canada, 2005). Differences in patterns of alco- hol use between homeless people in our sample and the general popula- tion occur mainly in the percentage of heavy drinkers. Seventy-two per- cent of people in our survey who reported drinking alcohol, reported heavy drinking (five or more drinks on one occasion) at least once in the past year, compared to 44 percent of the general population (Statistics Canada, 2005). Of those who said they had consumed alcohol in the past year, 55 percent reported heavy drinking, more than once a month in the past year. In the general population of Toronto, 22 percent reported the same (Statistics Canada, 2005). Our survey also found that 7 percent had consumed non-beverage alcohol in the past year and four people said that they do this almost daily. Non-beverage alcohol is alcohol in a form that is not meant to be consumed and includes things like mouthwash, hand sanitizer, cooking wine, and rubbing alcohol. Homeless people may drink non-beverage alcohol because it is less expensive and easily available. Some types of non-beverage alcohol (like methanol, found in anti-freeze) are extremely toxic and can cause blindness or death. Dangerous toxic health effects also result from the mix of other chemicals present in these products. I?ve been looking for counselling and I haven?t been able to find any. I lost my kid in the past year. My coping mechanism ? I?m embarrassed to say it ? but I?ve turned to street drugs. ? Marijuana is illegal but it seems to ease my depression, which makes me eat. If it helps, it helps. (Survey Respondent) Of the people we surveyed, 59 percent use at least one illicit drug regularly (three or more times a week), other than marijuana. Twelve percent use marijuana only and 28 percent said they had not used any illicit drugs regularly in the past year. Nearly half of our total sample reported regular crack use. This is very high compared to the crack use rate of 1 percent reported by the general Toronto population (City of Toronto, 2005). Crack use presents many serious health risks, including Hepatitis, HIV, and respiratory problems. There is also intense stigma surrounding crack use and few treatment or support options are available. More than 1 in 10 of our total sample, or 23 percent of those who had used any drugs regularly, re- ported having injected drugs in the past year. Sharing contaminated needles makes injection drug use one of the leading causes of HIV, hepa- titis and other blood-borne infections. Drugs used regularly by respondents in the past year % Crack 49% Marijuana 48% Cocaine 30% Opiates/analgesics (other than Oxycontin) 16% Sedatives, hypnotics or tranquilizers (other than downers) 16% Oxycontin 15% Morphine 10% Heroin 7% Hallucinogens 7% Methamphetamines (crystal meth, uppers, speed) 4% Downers 6% Methadone 5% Amphetamines (Benzedrine, Ritalin) 4% Solvents and other inhalants 2% People of all income levels use drugs for a variety of individual and systemic reasons. Drugs are often used to help people to cope with ill- ness, trauma, stress or pain, and to relieve isolation and boredom. This is probably the case for many of the people we interviewed, 73 percent of whom said that they had used alcohol or drugs in the past year to relieve stress or pain or to feel better about their life. It is likely that many peo- ple in our survey are ?self-medicating? themselves to relieve symptoms of problems for which they cannot get medical treatment, and using ille- gal drugs because they are easier to obtain than prescription medica- tions. Access to Health Care Homeless people often experience difficulties obtaining the health care they need. One in ten reported not using any health care services at all in the year before the survey. Also, 59 percent do not have a family doctor, compared to only 9 percent of Toronto population (Statistics Canada, 2005). Hospital emergency departments were the most frequently used source of health care for homeless people in our survey and many had been hospitalized in the past year. Sources of health care used by respondents in the past year % Aver. # of times Emergency Department 54% 5 Doctor?s office 44% 12 Services at shelters, drop-ins, health bus 42% 15 Community Health Centre 31% 11 Walk-in Clinic 29% 4 Hospitalization (at least one night) 24% 2 Hospital Outpatient Clinic 13% 9 Aboriginal health centre 6% 7 Alternative health centre 1% 10 Community health centres are a model of health care designed to promote access to health for people facing barriers to care, and address a wide range of health-related needs. This makes them well-suited to pro- vide health services to homeless people, but barriers still exist, such as the lack of walk-in services and community health programs that do not focus on the specific needs of homeless people. Only 16 percent of those in our survey cited community health centres as one of their usual sources of care. Outreach-based services are designed to address the barriers of the mainstream health care system. In 1989, there were perhaps four or five street nurses, but today there are more than a hundred street nurses working across Canada (Crowe, 2007). Street nursing services are deliv- ered outside mainstream health care settings, in places where homeless people spend time and where they feel comfortable. Some shelters and meal programs also offer on-site nursing care during set times each week. Some health agencies operate mobile health vans or buses that drive around the city offering care at specific spots and along the way. Outreach workers and nurses take knapsacks and walk around parks, beneath bridges and in ravines, to reach people who might not otherwise be able to access health care on their own. While many homeless people rely on these services for health care, they are not intended to provide comprehensive care or to replace the mainstream health care system. The increase of this type of health ser- vices is a reflection of increasing homelessness and homeless people?s poor access to the mainstream health care system. Almost one-third (29 percent) of homeless people in our survey said that they did not have a usual source of health care. They gave us the following reasons for this situation. Reasons given by those respondents with no usual source of health care % Seldom or never get sick 42% Don?t use doctors or treat self 24% Don?t have a health card 19% Move around a lot within Toronto 15% Negative past experience 12% Recently moved to Toronto 11% Don?t know where to find care 10% Many respondents cited not needing health care as a main reason for not having a stable health care provider. This is surprising and unlikely, considering that three-quarters of respondents have at least one serious physical health condition. This suggests that some homeless people have a lower sense of entitlement and lower expectations about their health and their right to access care. This is also related to homeless people?s difficult living situations, where they often have to prioritize more immediate needs such as shelter, and do not have the luxury of addressing preventive health care. Barriers to Health Care and Social Services Multiple barriers affect homeless people?s access to various types of health care, including hospitals, primary health care, eye doctors and dentists. Many of these barriers relate specifically to homeless people?s poverty and the difficulty of life without a permanent home. Economic barriers include not having money to get to medical appointments or to pay for prescriptions. Other barriers include not having a telephone or stable address and needing to prioritize survival needs such as food and shelter. Health care providers remain a critical access point for a multitude of health and social benefits. Forty-one percent of survey respondents said they had needed a health care provider to fill out a form in order to obtain health or social benefits in the past year. But 59 percent of our sample doesn?t have a regular family doctor to sign their forms. Some doctors also charge a fee for getting forms signed, which presents an ad- ditional barrier. The burdensome and complicated process of having to get medical forms filled out in order to receive social assistance has been cited as a major barrier that prevents homeless people from receiving Ontario Disability Support Program benefits, in a separate study con- ducted by Street Health (Street Health, 2006). Ontario Health Cards Twenty-eight percent of all respondents had been refused health care in the past year because they did not have an Ontario Health Card, and 34 percent did not have such a card. Of those without health cards, only 7 percent (9) said they were not eligible for one. The other reasons for not having a card were: 66 percent had either lost it or had it stolen; 14 per- cent were waiting for a card they had applied for; and 4 percent said their health card had expired. Several respondents said that they had lost their identification as a result of being arrested, going to jail or because the police had taken it from them and had not returned it. Having identification taken by police or losing track of it while in the prison system was also noted in a 2006 Toronto study on homelessness and the criminal justice system (Novac et al., 2006). Beyond health cards, other forms of identification are essential for accessing a wide range of social services and resources. Among our sur- vey respondents, 50 percent did not have a Social Insurance Number6 People in our sample cited many essential services that they were not able to access due to lack of identification documents. card and 29 percent did not have identification that provides proof of citizenship, such as a birth certificate, citizenship card, record of landing and passport. While not having a health card can prevent people from accessing health care, lack of a Social Insurance Number can stop people from accessing income support, training, housing, and from getting a job. Citizenship documents are particularly important, because they en- able people to apply for all other pieces of identification. What respondents could not get due to lack of identification % Ontario Works (welfare) benefits 18% Employment 14% Food bank 12% Housing 11% Training/education 6% Ontario Disability Support Program benefits 4% Discrimination in Health Care Once they see that you?re homeless, their attitude goes from caring to ?get out of here.? (Survey Respondent) Forty percent of those we interviewed said that they had been judged unfairly or treated with disrespect by a doctor or medical staff at least once the past year. Discrimination and poor treatment indicate that, at best, many homeless people are not having their health problems taken seriously or investigated adequately. At worst, it means that they may not be having their health problems treated at all. Discrimination and negative experi- 6 A Social Insurance Number is required to work in Canada and to receive government benefits. ences are real and serious barriers to health care, and prevent many homeless people from getting much-needed care. Reasons respondents felt they experienced discrimination by health care providers % Homelessness 66% Respondent?s use of alcohol or drugs 53% Perception that respondent was drug-seeking 47% Gender 14% Race or ethnic background 13% Ability to speak English 7% Sexual orientation 5% I was helping my friend and he was dirty and did not look good, so [ho s- pital] security gave us a hard time and told us to go away. (Survey Respondent) Negative experiences with hospital security were also commonly reported by people in our survey, with twenty-one percent of respon- dents having had at least one such experience: ? 12 percent had been denied access or told to go away; ? 12 percent had been threatened or verbally assaulted; ? 8 percent had been physically removed; ? 5 percent had been physically assaulted. These hospital security findings are even more startling and signifi- cant when we consider that homeless people use hospitals and emer- gency departments at very high rates. Conclusion Overall, homeless people in Toronto have much poorer health than the general population. Homeless people in our study carry an alarmingly high burden of many serious physical and mental health conditions. The most important factors impacting the health of homeless people are the result of social policy decisions that have been made by our governments in the past 15 years, particularly the cuts to social assistance and the lack of investment in new affordable social housing. Some of the key cuts over the years are outlined below: ? 1993: The federal government cancelled all funding for new afford- able housing. ? 1995: The Ontario government cancelled its funding for new afford- able housing, and 17,000 homes already approved for development. ? 1996: The federal government downloaded responsibility for afford- able housing to the provinces and territories, and began a steady de- cline in federal housing spending. ? 1998: The Ontario government downloaded responsibility for af- fordable housing to municipalities. Starved of funding and programs by the provincial and federal governments, and forced to take on the responsibility for affordable housing, the City of Toronto has a poor record of developing much- needed affordable housing. In the past decade, Toronto has completed only about 1,500 new affordable homes. In 23 of the city?s 44 municipal wards, not a single new affordable home has been completed (Wellesley Institute, 2006). There is an urgent need to take action to: ? Address the poverty and inequality that underlies homelessness; ? Improve access to affordable and appropriate housing; ? Improve immediate living conditions for homeless people; ? Improve access to health care and support for homeless people. Although adequate incomes and housing are the core solutions to improving homeless people?s health and health care access, homeless people need good access to quality health care now. Proper access to good primary and mental health care, dental and vision care, as well as prescription drugs, prevent illnesses from becoming more serious and costly to the health care system. There is an immediate need to address barriers in the health care system for homeless people, and to assist homeless people in navigating the complex systems that deliver health and related services. This chapter is drawn from The Street Health Report 2007, published in To- ronto, September 2007, and prepared by Erika Khandor (Research & Evalu a- tion Coordinator, Street Health) and Kate Mason (Street Health Survey Coo r- dinator, Street Health). The research team also included Laura Cowan (Executive Director, Street Health) and Dr. Stephen Hwang (Research Scien- tist, Centre for Research on Inner City Health, St. Michael?s Hospital). To read the full report, go to www.streethealth.ca. References Ambrosio, E., Baker, D., Crowe, C., & Hardill, K. (1992). The Street Health report. Toronto: Street Health. Canadian Centre for Policy Alternatives (2006). Growing gap, growing concerns: Canadian attitudes towards income inequality. Ottawa: Author. Canadian Liver Foundation. (2006). More Canadians being diagnosed with hepa- titis. Media release, March 23. Toronto: Canadian Liver Foundation. Centre for Addiction and Mental Health. (2006). What are concurrent disorders? Available at: www.camh.net/About_Addiction_Mental_Health/Concurrent _Disorders/Concurrent_Disorders_Information_Guid e/what_are_cd_infog u ide.html Chow, C.M., Donovan, L., Manuel, D., Johansen H., Tu, J.V. (2005). Regional Variation in self-reported heart disease prevalence in Canada. Canadian Journal of Cardiology 21(14), 1265-71. City of Toronto (2003). The Toronto report card on homelessness and housing. To- ronto: Author. City of Toronto (2005). The Toronto Drug Strategy, appendix E: Substance use in To- ronto: Issues, impacts & i nterventions. Toronto: Toronto Drug Strategy Advi- sory Committee. City of Toronto, Shelter, Support & Housing Administration (2006). Quick facts. Toronto. Crowe, C. (2007). Dying for a home. Toronto: Between the Lines. Engeland, J. (2004, December 14). Excluded from acceptable housing: What does it mean to be in core housing need? (PRI Conference Presentation). Ottawa: Ca- nadian Mortgage and Housing Corporation. Falvo, N. (2003). Gimme shelter! : Homelessness and Canada's social housing crisis. Toronto: Centre for Social Justice. Force on Modernizing Income Security for Working-Age Adults (2006). Time for a Fair Deal. Toronto: St. Christopher House, Toronto City Summit Alliance. Health Canada. (2002). A Report on Mental Illness in Canada. Ottawa: Health Can- ada. Housing Connections (2006). Annual report. Toronto: Housing Connections. Hwang, S. W., Svoboda, T. J., De Jong, I. J., Kabasele, K. J., & Gogosis, E. (2005). Bed bug infestations in an urban environment. Emerging Infectious Diseases, 11(4), 533-538. Marmot, M., & Wilkinson, R. (Eds.). (2003). Social determinants of health: The solid facts (2nd ed.). Copenhagen: World Health Organization. McKeown, D. (2006, February 13). Hot weather response plan update (Toronto Staff Report to Board of Health). Toronto: City of Toronto. Novac, S., Hermer, J., & Paradis, E. (2006). Justice and injustice: Homelessness, crime, victimization, and the criminal justice system. Toronto: Centre for Urban and Community Studies, University of Toronto. Public Health Agency of Canada (2001). Final report: Estimating the number of per- sons co-infected with Hepatitis C Virus and Human Immunodeficiency Virus in Canada. Ottawa: Public Health Agency of Canada. Public Health Agency of Canada. (2003). Canadian Data on the trends of HCV. Ottawa: PHAC. Available at: http://www.phacaspc.gc.ca/hcai-iamss/bbp- pts/hepatitis/hep_c_e.html. Public Health Agency of Canada. (2007). About FASD: Frequently Asked Ques- tions. Ottawa: PHAC. Available at: www. phac-aspc.gc.ca/fasd-etcaf/faq_e. htm l#6. Raphael D. (2002). Social justice is good for our hearts: Why societal factors ? not life- styles ? are major causes of heart disease in Canada and elsewhere. Toronto: CSJ Foundation for Research and Education. Remis, R.S., Swantee, C., Schiedel, L., Merid, M.F., Liu, J. (2006). Report on HIV/AIDS in Ontario 200 4. Ontario: Ministry of Health and Long-term Care. Shapcott, M. (2006). TO?s ?sheltered? homeless: Up or down? Toronto: Wellesley Institute. Shapcott, M. (2007). Housing action: Municipal budget. Toronto: Wellesley Institute. Statistics Canada. (2001). Canadian Community Health Survey, 2000/01. Table (105- 0004). Statistics Canada. (2004). Canadian Community Health Survey, Mental Health and Well-Being, 2002 . Available at: http:// www.statcan.ca/english/freepub/82- 617-XIE/htm/5110066.htm Statistics Canada. (2005). Canadian Community Health Survey, Cycle 3.1. Ottawa: Statistics Canada. Street Health (2006). Failing the homeless: Barriers to the Ontario Disability Support Program for homeless people with disabilities. Toronto: Street Health. Tarasuk, V. (2007). Filling Their Bellies: Community Initiatives to Feed the Homeless. Toronto: University of Toronto, Centre for Urban Health Initiatives. Presen- tation, May 2, University College. TD Economics (2007). An update to TD Economics? 200 2 Report on the Greater T o- ronto Area Economy. Toronto: TD Bank Financial Group. Tolomiczenko, G., Goering, P. (1998). Pathways into homelessness: Broadening the perspective. Psychiatry Rounds 2(8). Toronto: Centre for Addiction and Mental Health. Toronto Police Service (2006). 2 0 05 Annua l statistical report. Toronto: Author. Toronto Public Health (2007). Communicable diseases in Toronto 2005 . Toronto: City of Toronto. Wellesley Institute. (2006). The Blueprint to End Homelessness in Toronto. Toronto: Wellesley Institute. Zee, P., & Turek, F. (2006). Sleep and health: Everywhere and in both directions. Archives of Internal Medicine, 166, 1686-1688. Chapter 2.3 Mental Health, Mental Illness, and Homelessness in Canada CANADIAN P OPULATI ON HEALTH INITIATIVE O F THE CANADIAN INSTITUTE F OR HEALTH IN F ORMATI ON Pathways into Homelessness Homelessness or the risk of homelessness is a harsh reality for many Ca- nadians. It is not confined to any one group in society, but may affect youth, men and women, one- or two-parent families, the elderly, new immigrants, Aboriginal Peoples, and others (Shortt et al., 2006). It is not an individual characteristic, but rather a life circumstance that can be temporary, episodic or relatively long lasting (Begin et al., 1999). At present there is no universally agreed-upon definition of homelessness, nor is there a clear picture of the prevalence and composition of Cana- da?s homeless population. Studies show that people who are homeless are more likely to expe- rience compromised mental health and mental illness (Hwang, 2001; Public Health Agency of Canada, 2006). For some, these issues can pre- cede the onset of homelessness (Mental Health Policy Research Group, 1997). For others, they can be worsened with continued homelessness (Frankish et al., 2005). At the same time, it is important to note that not all people with mental illness are homeless, and not all people who are homeless report a mental illness. For example, a 1997 Toronto study of 300 shelter users found that while two-thirds of respondents reported a lifetime diagnosis of mental illness (Goering et al., 2002), mental illness was the least reported reason for becoming homeless (4 percent); loss of job or insufficient income to pay rent was the main reason (34 percent) (Mental Health Policy Research Group, 1997). How Are Mental Health and Homelessness Related? The terms ?mental health? and ?mental illness? are sometimes used in- terchangeably or are seen as two ends of a single continuum. However, many definitions emphasize that mental health is more than the absence of mental illness. For example, the Public Health Agency of Canada (2006) says that, ?mental health is the capacity of each and all of us to feel, think and act in ways that enhance our ability to enjoy life and deal with the challenges we face. It is a positive sense of emotional and spiri- tual well-being that respects the importance of culture, equity, social jus- tice, interconnections and personal dignity? (p. 2). Many studies show that people who are homeless are more likely to experience compromised mental health and mental illness than the gen- eral population (Hwang, 2001; Public Health Agency of Canada, 2006). For some, these issues can precede the onset of homelessness or, through their interaction with other determinants such as income and employ- ment influences, contribute to homelessness (Centre for Addiction and Mental Health, 2003). They may also worsen with continued homeless- ness (Frankish et al., 2005) or contribute to the duration of homelessness (Centre for Addiction and Mental Health, 2003). Patterns of mental health can be influenced by a number of factors, including personal coping skills; perceived self-worth; one?s social envi- ronment; and other physical, cultural and socio-economic characteristics (Public Health Agency of Canada, 2006). Many of these factors may also be related to the risk of becoming or remaining homeless (Canadian Mental Health Association, 1997; Federation of Canadian Municipalities, 2004). Experts indicate that, through mental health promotion, positive mental health can play a role in one?s recovery process. Mental health promotion ?empowers people and communities to interact with their environments in ways that enhance emotional and spiritual strength? through strategies to increase self-esteem, coping skills, social support, and well-being (Public Health Agency of Canada, 2006, p. 21). Increa- singly, Canadian and international studies of the homeless population have examined these aspects of mental health and mental health promo- tion, particularly among homeless youth. Examples of their findings are highlighted below. Stress Studies in Canada and elsewhere suggest that stress levels are higher among the homeless than among the population as a whole. Overall, data from the 2003 Canadian Community Health Survey (CCHS) indi- cate that 24 percent of Canadian adults report having ?quite a lot? of stress (Canadian Institute for Health Information, 2006). Like other simi- lar studies, however, this survey was not administered to homeless pop- ulations. Studies involving the homeless often use other measures and are thus not directly comparable. In some cases, comparisons can be made to published scores compiled from the general population; in other cases, studies include non-homeless comparison groups. Two Canadian examples found that: ? In Kitchener-Waterloo, Ontario, street youth reported more stressors in the past year than non-homeless youth (10.4 on average versus 7.2) (Ayerst, 1999). ? In Ottawa, Ontario, homeless male youth reported an overall stress level that was more than two times higher than that reported by a group of non-homeless male youth (Votta & Manion, 2003). In the U.S., a Los Angeles, California, study involving youth who were homeless or at risk of homelessness found increases in depressive symptoms and substance abuse disorders, as well as poorer self-rated health with increased stress (Unger et al., 1998). Coping Coping skills have been linked to health and well-being (Public Health Agency of Canada, 2004). A number of studies have looked at how homeless individuals cope with stress. Research suggests that homeless youth have a tendency toward using coping styles and strategies that work to distance them from a stressor rather than actively attempting to solve it. For example: ? In Kitchener-Waterloo, Ontario, a study found that street youth were more likely to engage in substance use and self-harm as a means of coping; non-homeless youth were more likely to cope by talking to someone they trusted or through productive problem-solving (Ayerst, 1999). ? In Ottawa, Ontario, homeless male youth were more likely to use strategies such as avoiding the problem, withdrawing from social networks, and avoiding negative thoughts and emotions to cope than were non-homeless youth (Compas, 1997; Votta & Manion, 2003, 2004) Among homeless youth only, this style of coping was re- lated to depressive symptoms and various internalizing behaviour problems?the latter of which was measured by anxiety/depressive symptoms, withdrawal, and somatic complaints (that is, unex- plained physical problems) (Votta & Manion, 2003). ? In Los Angeles, California, homeless male and female youth who reported using such strategies as wishing the problem would disap- pear or using substances tended to have higher levels of stress, social isolation, symptoms of depression, and poor self-rated health; in contrast, homeless youth who tried to solve a problem or change a situation reported good self-rated health (Unger et al., 1998). ? Likewise, among adults, a U.S.-based study found that homeless men with a persistent mental illness reported significantly less use and effectiveness of cognitive (for example, problem-solving me- thods), socio-cultural (for example, seeking social support), and spi- ritual (for example, prayer) coping strategies than did homeless men with an addiction and homeless men dealing with a specific crisis situation (Murray, 1996). Social Support Social support has also been linked to health and well-being, and it can play a role in helping people cope with stress (Johnson et al., 2005a; Pub- lic Health Agency of Canada, 2004; Thoits, 1995). Examples include the number of social relationships, frequency of contact, connections among members of social networks, availability of social support, and the type of support received, for example, emotional support (Thoits, 1995). The evidence about links between social support and mental health are noteworthy, given the reported lack of social support among various segments of the homeless population. For example, one study in Ottawa, Ontario, found that homeless male youth reported less perceived parental support than non-homeless male youth (Votta & Manion, 2003). Another Ottawa study found that 15 percent of adults living on the street reported receiving no social support (Farrell et al., 2001). Various studies report associations between social support and mental health outcomes among people who are homeless. As seeking social support can be a way for people to cope, it is not surprising that these findings are similar to those reported in the coping literature. Ex- amples of findings from existing studies involving youth include the fol- lowing: ? In Toronto, Ontario, compared to street youth with lower levels of social support, street youth with a high level of social support re- ported a significantly lower mean depression score (Smart & Walsh, 1993). ? Among homeless youth in Los Angeles, California, increased availa- bility of social support was associated with reduced depressive symptoms and better self-rated health (Unger et al., 1998). ? In Washington, D.C., the 26 percent of runaway and homeless youth who did not indicate they had a current social network had higher odds of using illicit drugs and engaging in risky sexual behaviours (Ennett et al., 1999). Among adults, Nyamathi et al. (2000a) found that 51 percent of homeless women in Los Angeles reported no current substantial source of social support. Compared to these women, homeless women report- ing support from individuals who were not substance users reported higher self-esteem, more active coping, greater life satisfaction, and low- er levels of both anxiety and depression. Self-Esteem Self-esteem is another factor often discussed in relation to mental health and well-being. A Toronto study found that street youth with high self- esteem reported being less depressed than those with lower reported self-esteem (Smart & Walsh, 1993). A study of youth in substance abuse treatment programs in Ontario found that compared to 66 percent of non-homeless youth, 50 percent of homeless youth reported feeling good about themselves (Smart & Ogborne, 1994). Similar findings have also been noted in international studies. For instance: ? Relative to different groups of non-homeless youth, a study in Syd- ney, Australia, found that homeless youth scored significantly lower in four areas of self-concept: impulse control (control of aggression, anxiety, resentment, fear), emotional tone (feelings of tension, sad- ness, loneliness, inferiority), family relations, and level of psychopa- thology. Among homeless youth, hopelessness was associated with lower overall self-esteem (Miner, 1991). ? Low self-esteem, along with low support from positive sources, higher support from deviant sources (drug-using family/friends or drinking partners), and avoidant coping (for example, withdrawing from others), was significantly related to high mental distress scores among homeless women in Los Angeles, California (Nyamathi et al., 2000b). ? Another Los Angeles study found that 16 percent of street youth re- ported low self-esteem, which was itself associated with increased risk of both alcohol and drug use and suicidal thoughts/attempts (Unger et al., 1997). Suicidal Behaviours Although much remains unknown about the causal pathways between mental health and suicide, suicidal behaviours have been linked to as- pects of mental health among homeless individuals. Qualitative studies have found that feelings of hopelessness, loneliness, worthlessness, and being trapped were themes underlying homeless youths? experiences with suicide (Kidd, 2004). Existing research shows an association be- tween suicidal behaviours and coping. Among homeless male youth, suicidal behaviours were associated with having a coping style that does not involve actively trying to solve a problem or cope with a stressor (Votta & Manion, 2004). A number of Canadian studies report higher rates of suicidal thoughts and suicide attempts among homeless youth than among youth who are not homeless. According to the Public Health Agency of Canada (2006), 12 percent of males and 19 percent of females aged 15 to 24 report having had suicidal thoughts at some point in their lifetime. Fewer (2 percent of males and 6 percent of females aged 15 to 24) report having attempted suicide. Findings from studies involving homeless youth in- clude the following: ? A 2006 survey of youth across British Columbia indicated that com- pared to 4 percent of males and 10 percent of females in schools, 15 percent of males and 30 percent of females who were street-involved and marginalized reported having attempted suicide at least once in the previous 12 months (McCreary Centre Society, 2007). ? In Ottawa, Ontario, compared with 4 percent of non-homeless male youth, 21 percent of homeless male youth reported at least one past suicide attempt. Compared with 34 percent of non-homeless youth, 43 percent of homeless youth reported suicidal thoughts (Votta & Manion, 2004). ? Of homeless youth sampled in Toronto, Ontario, and Vancouver, British Columbia, 46 percent reported a past suicide attempt (Kidd, 2004). ? In Richmond Hill, Ontario, 20 percent of homeless youth reported at least one suicide attempt in their lifetime; 25 percent reported suicid- al thoughts (Cameron et al., 2004). How Are Mental Illness and Homelessness Related? The Public Health Agency of Canada (2006) defines mental illness as ??alterations in thinking, mood or behaviour ?or some combination thereof?associated with significant distress and impaired functioning.? (p. 2) Compared with the general population, research shows a greater incidence and prevalence of persons with serious mental illnesses be- coming or remaining homeless (Levine, 1984). Other research has docu- mented a higher prevalence of mental disorders among the homeless than among the general population (D?Amore et al., 2001). In Toronto, Ontario, 67 percent of shelter users in the Pathways into Homelessness Project reported a lifetime diagnosis of mental illness (Goering et al., 2002). Schizophrenia and Personality Disorder In Statistics Canada?s 2002 Mental Health and Well-being Survey, less than 1 percent of adults in the general population reported having been professionally diagnosed with schizophrenia (Public Health Agency of Canada, 2006). Canadian and U.S. studies, including the following, re- port higher rates of schizophrenia among the homeless: ? In Toronto, Ontario, 6 percent of 300 shelter users reported a psy- chotic disorder, primarily schizophrenia (Mental Health Policy Re- search Group, 1997). ? A Vancouver, British Columbia, study reported that 24 of 124 shelter users had a mental health problem; of these, 7 identified their mental health problem as schizophrenia (Acorn, 1993). Toronto?s Pathways into Homelessness Project also found that 29 percent of shelter users met criteria for anti-social personality disorder, often in addition to another diagnosis such as depression, post-traumatic stress disorder (PTSD) or psychotic disorder (Mental Health Policy Re- search Group, 1997). PTSD is a disorder associated with a traumatic event and characte- rized by various symptoms, including persistent and recurring thoughts or images and avoidance behaviours (Rothschild, 2000). Research indi- cates that physical and sexual abuse occurring while people are homeless is a risk factor for the onset of PTSD (Mueser et al., 2004). In a study in- volving homeless youth, 24 percent met criteria for PTSD; 40 percent who met the criteria for substance abuse disorder also met the criteria for PTSD (Johnson et al., 2005b). Substance Abuse and Concurrent Disorders Among the general population, data from Statistics Canada?s 2002 Men- tal Health and Well-being Survey indicate that among females, 4 percent of young women (15 to 24 years of age) and 1 percent of adult women (25 to 44 years of age) report alcohol dependence in the previous 12 months. Fewer (2 percent of young women and less than 1 percent of adult women) report illicit drug dependence in the previous 12 months. Rates are higher among males: 10 percent of young men and 4 percent of adult men report alcohol dependence, while 4 percent of young men and 1 percent of adult men report illicit drug dependence in the previous 12 months (Public Health Agency of Canada, 2006). Canadian studies indicate that rates of substance abuse are higher among homeless individuals than among the general population. For example, in Toronto, Ontario, 68 percent of shelter users reported a life- time diagnosis of substance abuse or dependence (Goering et al., 2002). Other Canadian studies have found that: ? In Vancouver, British Columbia, 44 percent of homeless adults re- ported use of non-prescription drugs such as marijuana and cocaine within the past month (Acorn, 1993). ? In Edmonton, Alberta, 40 percent and 55 percent of homeless youth reported drinking alcohol and using marijuana, respectively, at least two to three times a week (Baron, 1999). ? Various Canadian studies also report high levels of opioid and non- opioid drug use among the homeless. For example, in Edmonton, Alberta, 55 percent of street youth reported using at least one of four drugs (cocaine, heroin, amphetamines or tranquilizers) in the past year (Baron, 1999). A Montr?al , Quebec, study of street youth over a five-year period noted an incidence rate of drug injection use of 8.2 per 100 person-years among a cohort of 415 street youth (Roy et al., 2003)?at study entry, these youth had never used injection drugs. Some individuals have both substance abuse disorders and mental illness diagnoses, known as ?concurrent disorders? (Shortt et al., 2006). Other terms used include ?dual diagnosis,? ?dual disorder,? ?comorbidi- ty? or ?co-occurring substance abuse disorders and mental disord- ers?(Centre for Addiction and Mental Health, 2006). Published literature reviews suggest that homeless individuals with concurrent disorders are likely to remain homeless longer than other homeless people (Drake et al., 1991). In Toronto, Ontario, almost all of the shelter users in the Path- ways into Homelessness Project who reported a lifetime diagnosis of mental illness also had a substance abuse disorder (Mental Health Policy Research Group, 1997). Depressive Symptoms and Major Depressive Disorder (MDD ) Research also suggests that depression is more common among home- less Canadians than among others. Among the general population, 14 percent of 15- to 24-year-old females and 17 percent of 25- to 44-year-old females report having been diagnosed with depression at some point in their life. Reported rates are lower among male youth and adults?7 per- cent and 10 percent, respectively (Public Health Agency of Canada, 2006). Methods used in research among the homeless are not directly comparable, but studies have found that: ? Homeless male youth in Ottawa, Ontario, were more likely than non-homeless male youth to report scores for depressive symptoms (39 percent versus 20 percent) and internalizing behaviour problems (44 percent versus 24 percent) that were within a clinical range. As noted previously, the latter were measured based on the frequency of withdrawal behaviours, symptoms of anxiety/depression, and un- explained physical problems (Votta & Manion, 2003). ? One-third (33 percent) of a sample of Ottawa, Ontario?s adult street population self-reported mental health difficulties; of these, 20 per- cent reported depression (Farrell et al., 2001). ? In Kitchener-Waterloo, Ontario, street youth had a significantly higher mean level of depression than non-runaway youth. About half of the street youth in this study (48 percent) reported a decrease in their depression level since leaving home, while 28 percent re- ported an increase (Ayerst, 1999). Research involving the homeless in the U.S. reports a range of find- ings. For example, one study conducted in a large northwestern U.S. city found that 12 percent of 523 homeless youth reported a diagnosis of de- pression. Rates of depression were higher among females than males (20 percent versus 7 percent). About three-quarters of those surveyed (73 percent) reported experiencing their first depressive episode before leav- ing home (Rohde et al., 2001). This variation may reflect a number of is- sues including the use of different measures for assessing prevalence rates (Boivin et al., 2005) or the use of different terminology to reflect symptoms or diagnoses (Susser et al., 1989). Determining the Status of Mental Health and Mental Illness Among the Homeless Accurately measuring mental health status and mental illness among Canada?s homeless population, as well as their use of appropriate mental health services, is complicated. Methodological issues include: ? the different means by which mental illness among the homeless is defined (Susser et al., 1989), which limits the comparisons that can be made between cities, over time, or with the general population; ? variations in the nature of information reported in terms of specific diagnoses; ? a lack of representative information across the provinces and territo- ries; ? the use of terms such as ?mental illness,? ?mental health problems,? ?mental health concerns,? and ?mental health difficulties??to name a few?interchangeably. Use of Mental Health Services Dozens of different mental health services exist, although the types of service available?and the populations to whom they are available? vary across the country. Not everyone with mental health problems uses these services. This is true for both the homeless population and others, although the circumstances may be somewhat different. For example, while two-thirds of homeless respondents in a Toronto, Ontario, study reported having been diagnosed with a mental illness at some time dur- ing their life, only 25 percent reported receiving psychiatric outpatient services in the previous year (Mental Health Policy Research Group, 1997). Likewise, homeless men with schizophrenia in New York City, New York, were less likely to report having received assistance with dis- charge planning for living arrangements, aftercare and finances upon release from hospital than non-homeless men with schizophrenia (Caton, 1995). Recent research has also explored the barriers that homeless people report encountering in attempting to get help. A Los Angeles, California, study reported that 218 of 688 homeless youth perceived a need for help with mental health problems; 95 had received help and 123 had not. Those who identified a need for help, but who did not get help, cited various reasons, such as not knowing where or what services to use (53 percent), feeling embarrassed (47 percent), not having money to get to the service (36 percent), fears the service provider would contact family (36 percent) or police/social worker (36 percent), thinking the service would not help (33 percent), and the cost of the service (14 percent) (So- lorio et al., 2006). When the homeless do use services, studies indicate that there may be a tendency to use clinics and emergency departments (EDs). A study of over 2,900 homeless patients in the U.S. found that 63 percent received medical care at locations such as outpatient clinics and shelters in the previous year (Kushel et al., 2001). Published reports put the proportion of the homeless population who have received medical care in the ED in the past year at 32 percent (Kushel et al., 2001) to 40 percent (Kushel et al., 2002). Factors associated with ED use included symptoms of ill health, injuries, substance dependence, and depressive symptoms (among homeless men) (Padgett et al., 1995), as well as being a victim of crime, unstable housing, and medical comorbidity (Kushel et al., 2002). Data from the Canadian Institute for Health Information (CIHI) in- dicate that mental health and behavioural disorders account for a larger share of ED visits and hospital stays among the homeless than among the population as a whole (Canadian Institute for Health Information, 2007). (The data track ED use in Ontario and a handful of other centres, as well as hospital use outside of Quebec.) Most of the inpatient hospita- lizations tracked for the homeless took place in Vancouver, British Co- lumbia; Calgary, Alberta; and Toronto, Ontario. Toronto accounted for 78 percent of all ED visits by the homeless in Ontario. Mental health and behavioural disorders were the most common reason for ED visits by the homeless, but were not in the top five reasons for visits by other patients. These conditions accounted for more than one-third (35 percent) of visits by the homeless. Within this category, the most common type of mental disorder was psychoactive substance use (54 percent) followed by ?schizophrenia, schizotypal and delusional dis- orders? (20 percent). Reasons for visits for mental health and behavioural disorders varied for homeless men and women. Psychoactive substance use predominated for men (accounting for 62 percent of visits in this cat- egory), but it represented only 30 percent of visits for women. In both cases, ?schizophrenia, schizotypal and delusional disorders? was the next most common reason for visits for mental health and behavioural disorders (28 percent for homeless women and 18 percent for men). Mental diseases and disorders were also the most common reason for acute care hospitalization among the homeless, but were not as common among the rest of the population. In 2005? 2006, 52 percent of inpatient hospitalizations among the homeless (outside Quebec) were primarily for these conditions (Canadian Institute for Health Informa- tion, 2007). Mental Health Policy in Canada Starting in the 1960s, many psychiatric inpatients were discharged to the community when psychiatric hospitals or wards were closed and/or the number of beds in psychiatric facilities reduced (Herman & Smith, 1989; Nelson, 2006; Sealy & Whitehead, 2004). While there is no consensus on the impact of deinstitutionalization on the prevalence of homelessness, some researchers have suggested that deinstitutionalization was asso- ciated with the growth of new forms of residential or institutional care (Herman & Smith, 1989) as well as increased rates of homelessness (Commission on the Future of Health Care in Canada, 2002; Nelson, 2006; Public Health Agency of Canada, 2006). It has also been noted that community mental health services did not increase at the same rate as patients were deinstitutionalized (Sealy & Whitehead, 2004). Traditionally, community mental health programs had a communi- ty treatment and rehabilitation focus. Approaches reflected such values as reducing symptoms, preventing hospitalization, professionally pre- scribed treatment, community-based support, vocational training, and housing with an element of support (for example, group homes and halfway houses). The 1990s saw a shift toward recovery and empower- ment that reflected values consistent with the principles outlined in the 1988 federal discussion paper, Striking a Balance: Mental Health for Cana- dians: emphasis on recovery, recognizing strengths, consumer choice and control, community integration, informal supports, supported employ- ment, and independent housing with flexible support (National Health and Welfare, 1988; Nelson, 2006). In May 2006, the Senate?s report, Out of the Shadows at Last: Trans- forming Mental Health, Mental Illness and Addiction Services in Canada (Standing Senate Committee on Social Affairs Science and Technology, 2006) recommended the establishment of a Canadian Mental Health Commission and a national mental health strategy; funding was an- nounced by the federal government in March 2007 (Government of Can- ada, 2007). The report also noted that affordable housing is a key issue for people living with a mental illness: ?the percentage of Canadians who are living with mental illness who need access to [adequate, appr o- priate, and affordable] housing is almost double the percentage of people in the general population whose housing needs are not being met? (Standing Senate Committee on Social Affairs Science and Technology, 2006, p. 462). Mental Health Policy at the Provincial Level Several provinces have developed specific initiatives, plans, or frame- works to guide their policies and services. In many cases, they specifical- ly address issues related to homelessness (for example, the provision of supportive housing). Examples include: ? British Columbia?s Mental Health and Addictions Reform Initiative. The British Columbia Ministry of Health formed best practices working groups to identify various services and strategies that produce posi- tive health outcomes for individuals. The working group?s findings are reflected in BC?s Mental Health Reform Best Practices, which in- cludes reports specific to housing and Assertive Community Treat- ment (ACT) (British Columbia Ministry of Health, 2002). ? Alberta?s Mental Health Plan: Advancing the Mental Health Agenda. This plan highlights strategies targeted to specific population groups, in- cluding the homeless. Programs that provide the homeless with access to mental health programs and referral services on-site at shel- ters or drop-in centres are recommended. The plan also outlines var- ious priority strategies and actions such as the provision of safe and supportive housing for individuals with severe and persistent men- tal health problems (Alberta Mental Health Board, 2004). ? Manitoba?s Mental Health System. Various housing and community living programs are made available to individuals with mental health problems who may be experiencing difficulties living inde- pendently. These programs provide participants with several hous- ing services, including residential care facilities and supportive hous- ing (Manitoba Health, 2007). ? Ontario?s Mental Health Homelessness Initiative. Announced in 1999, this initiative aimed to ?address the housing needs of people with mental illness who were either homeless or at risk of becoming homeless.? (pg. 91) A comprehensive evaluation of the first phase of this initiative determined that housing choice/control and housing quality were related to subjective quality of life (Nelson et al., 2007). ? Quebec?s Mental Health Action Plan, 200 5 ?2010 . The goal of this plan is to improve access to quality mental health services for those with mental health disorders and/or those who have a high risk of suicide through action, rehabilitation, accessibility, continuity of services, partnerships, and efficiency. Quebec?s Minist?re de la Sant ? et des Services Sociaux is committed to prioritizing access to front-line mental health services and to reducing the stigma often associated with having a mental disorder, so that individuals feel comfortable seeking help. The plan will support the provision of quality mental health services to the entire population (for example, youth, adults, communities and Aboriginal Peoples) (Sant? et Services S ociaux Qu?bec, 2005). ? Newfoundland and Labrador?s Framework to Support the Development of a Provincial Mental Health Policy. As part of this framework, the com- munity resource-based model identifies housing as a key element for supporting the well-being of persons with mental health needs. The framework also aims to incorporate best-practice knowledge in housing and case management services (Government of Newfound- land and Labrador, 2001). Mental Health Promotion Among the Homeless: Housing Programs CIHI?s Improving the Health of Canadians: An Introduction to Health in Ur- ban Places (Canadian Institute for Health Information, 2006b) highlighted the roles that housing, both as a physical structure and the meaning it holds for individuals, can play in physical and mental health outcomes. It presented evidence of a relationship between the lack of affordable housing and both psychological distress (Cairney, 2005) and increased risk of homelessness (Bunting et al., 2004). Research also shows that se- curing physical housing resources can be associated with reduced psy- chological distress among the homeless (Wong & Piliavin, 2001) and play a role in supporting individuals recovering from severe mental illness (Borg et al., 2005). Different types of housing are available to individuals who are homeless and have mental health issues, such as supportive and sup- ported housing. Housing of this nature tends to be small-scale and fo- cused on rehabilitation and community integration (Centre for Addiction and Mental Health, 2005). Research also indicates that the costs asso- ciated with supportive housing are lower than the costs associated with emergency shelters (British Columbia Ministry of Social Development and Economic Security, 2001; Pomeroy, 2005). Supportive housing includes on-site staff support that varies de- pending on residents? needs (for example, group homes). Supported housing does not include on-site support staff but does include elements of recovery and empowerment (Centre for Addiction and Mental Health, 2005). Continuum of Care (Treatment First) and Housing First are two models designed to provide housing to the homeless while addressing existing mental health issues. Continuum of Care Models (Treatment First) The Continuum of Care model consists of several components. In the first phase, outreach, clients are encouraged to accept a referral to a second-step program such as a shelter or drop-in centre. In the next phase, clients are provided with, and required to take part in, any neces- sary psychiatric or substance abuse treatment. Permanent housing is made available to participants in the final stage, after treatment is com- pleted (Tsemberis & Eisenberg, 2000; Tsemberis et al., 2004). Housing First Models In 1992, Pathways to Housing (PTH) Inc., a non-profit New York City agency, developed the Housing First model (Padgett et al., 2006; Tsem- beris & Eisenberg, 2000; Tsemberis et al., 2004). Housing First programs offer those who are homeless and mentally ill immediate access to hous- ing that is not contingent on sobriety or treatment. These programs tend to promote harm reduction (that is, diminish the harm caused by drink- ing or drug use) instead of requiring abstinence (Padgett et al., 2006). They also offer clients a variety of services through interdisciplinary as- sertive community treatment (ACT) teams, thereby helping to engage those not reached by more traditional approaches. Effectiveness of Treatment First and Housing First Programs A number of studies have documented the effectiveness of the Housing First approach in housing retention among individuals who were home- less and mentally ill. Evaluations have not typically included evaluation of long-term health outcomes. ? A New York City study found that after a five-year period (1993 to 1997), 88 percent of participants in the Pathways to Housing pro- gram remained housed, compared with 47 percent of participants in Treatment First programs. This study also found that while dual di- agnosis reduced housing tenure among participants in both pro- grams, dually diagnosed participants in the PTH program had a higher housing rate than those in the Treatment First program (Tsemberis & Eisenberg, 2000). ? A recent randomized experiment involving homeless individuals who had a diagnosis of severe and persistent mental disorder found the Housing First approach to be more effective than the Treatment First approach in reducing homelessness (Tsemberis et al., 2003). Another study found that homeless participants with a major mental illness such as schizophrenia or bipolar disorder who were enrolled in PTH spent more time in stable housing and less time in hospitals than those in Treatment First programs (Gulcur et al., 2003). Mental Health Promotion Among the Homeless: Community Mental Health Programs As the pathways out of homelessness and into secure housing are not always easily found or immediate, there is value in understanding what strategies are effective at promoting mental health and addressing men- tal illness among individuals experiencing homelessness. Individuals who are homeless and have a mental illness are often reluctant to engage in some of the more traditional, office-based approaches to providing services (Johnsen et al., 1999). As a result, a number of community men- tal health programs have been developed (Dickey, 2000). Some provide outreach services, while others provide longer-term services in the form of assertive community treatment (ACT), intensive case management (ICM), or service integration. Outreach Services Outreach programs serve as a point of first contact for persons not linked to other models of service. They provide assessment and linking to other longer-term services. For example: ? The Psychiatric Outreach Team of the Royal Ottawa Hospital is a multidisciplinary team comprising an addiction worker, an occupa- tional therapist, a psychiatric nurse practitioner, a psychiatrist, a psychologist, a recreational therapist, and social workers. The team provides psychiatric services to the individual who is homeless or at risk of homelessness, as well as to the partner agency serving the particular individual (Farrell et al., 2005). ? The Street Outreach and Stabilization (SOS) program of the Cana- dian Mental Health Association, Calgary Region, provides individu- als who are homeless with help in obtaining mental/health services, financial resources, housing, legal assistance, daily life skills training, transportation training, opportunities for leisure and recreation ac- tivities, and information about and access to community resources (Canadian Mental Health Association, 2007). Most evaluations of outreach programs are of a formative and process nature (number of people served, number of referrals, links to other longer-term services, etc.) and do not look at long-term health out- comes. Assertive Community Treatment (ACT) ACT teams, typically comprising psychiatrists, psychologists, social workers, addiction specialists, and other professionals, offer intensive case management and support services for individuals with severe and persistent mental health problems. Services are provided on a long-term basis and often right within the client?s home community (Centre for Addiction and Mental Health, 2007). Evaluation studies indicate that, compared to those receiving tradi- tional health services, homeless individuals living with a severe and per- sistent mental illness who were ACT program participants had im- proved housing and clinical outcomes, as well as greater satisfaction with their general well-being, their neighbourhoods, and their health. ACT participants also had fewer psychiatric inpatient hospital days (35 versus 67) and emergency department visits (1 versus 2), and more out- patient mental health visits (103 versus 40)?these findings suggested a shift from crisis-oriented services to ongoing outpatient care (Lehman et al., 1997). Intensive Case Management (ICM) ICM is a client-directed form of mental health case management and, like ACT, provides program participants with intensive services and long- term support. Unlike the ACT approach, ICM?s services are provided through individual case managers as opposed to a multidisciplinary team (Community Mental Health Evaluation Initiative, 2003; Ontario Community Mental Health Initiative, 2007). The Community Mental Health Evaluation Initiative (CMHEI) is a six-year multi-site assessment of community mental health programs in Ontario (Ontario Community Mental Health Initiative, 2007). As part of this assessment, a clinical trial in Ottawa compared the service use and outcomes of homeless and mentally ill clients receiving ICM to those of clients receiving standard care. Many participants also had other chal- lenges such as concurrent substance abuse (Community Mental Health Evaluation Initiative, 2003; Aubry et al., 2006). Results showed im- provements among clients receiving both ICM and standard care in housing stability and community functioning, as well as decreases in hospitalizations and substance abuse. At the 24-month follow-up, ICM clients showed significantly lower levels of housing instability (10 per- cent versus 27 percent) and fewer hospitalizations (13 percent versus 32 percent) than those receiving standard care (Aubry et. al., 2006; Commu- nity Mental Health Evaluation Initiative, 2003; Ontario Community Men- tal Health Evaluation Initiative, 2007). Service Integration Another focus area specific to mental health and homelessness is the in- tegration of various services. In 1993, the U.S. Department of Health and Human Services initiated the 18-site Access to Community Care and Ef- fective Services and Supports (ACCESS) demonstration program as part of a nation-wide agenda to address homelessness among the seriously mentally ill. The goals of the program were twofold: ??to identify promising approaches to systems integration and to evaluate their effec- tiveness in providing services to this population.? (Randolph et al., 1997, pp. 369? 370). Findings from the ACCESS demonstration program are presented in many published reports. One study found no differences in mental health status and achievement of independent housing between experimental and control-group clients. However, it did find a positive association among participants enrolled in systems that became more integrated with better housing outcomes (Rosenheck et al., 2002). Mental Health Promotion Among the Homeless: A Population Health Approach Mental health promotion strategies, combined with specific treatment for a mental illness, can empower people to achieve well-being, develop healthy relationships, and maintain a form of housing and employment (Public Health Agency of Canada, 2007). The links between the determinants of mental health and the de- terminants of homelessness are interrelated and numerous and indicate a role for continued discussion and action to promote mental health among this population. Strategies to achieve this are related to the popu- lation health approach: ? ?focusing on the needs of the population as a whole as well as sub- populations with particular needs; ? addressing the determinants of mental health and their interactions; ? basing decisions on evidence of need and the effectiveness of inter- ventions; ? increasing investments on the social and economic determinants of health; ? applying multiple strategies in multiple settings and sectors; ? collaborating across sectors and levels of government; ? employing mechanisms for meaningful public involvement; and ? demonstrating accountability for health outcomes.? (Public Health Agency of Canada, 2006, p. 21). Conclusions The pathways linking mental health and homelessness are numerous and interrelated. For some individuals, the pathways into homelessness may be upstream, reflecting issues such as housing, income level, or em- ployment status. For others, the pathways may be more personal or in- dividual, reflecting issues such as compromised mental health and well- being, mental illness, and substance abuse. Many of these personal and upstream issues are linked to each other. Some studies suggest that the homeless are at higher risk for com- promised mental health and mental illness. Other research has found that those with compromised mental health or mental illness are more likely to become homeless. As most studies involving the homeless tend to be cross-sectional, it is difficult to identify causal pathways between the onset of mental illness and homelessness. Improving the Health of Canadians: Mental Health and Homelessness was the first report in a series of three produced by CPHI on the theme of mental health and resilience. Due to scoping restrictions, there were a number of areas the report did not address, including the role of positive traits as protective factors against negative mental health outcomes; age and sex differences in the onset of mental illness among the homeless; the impact of the length of time homeless; co-addictions; the impact of development disabilities as a concurrent diagnosis with mental illness; and the prevalence of Fetal Alcohol Spectrum Disorder (FASD). Lastly, given the availability of evidence, the report primarily focused on home- less youth and single adult males. It is important to identify the preva- lence of homelessness among other subgroups of the population, along with their mental health issues and needs. Understanding the link between mental health and homelessness requires consideration of both individual-level factors and the broader social determinants of health. Findings indicate that there may be value in clinical, outreach, and research programs that target specific issues such as coping skills, self-worth, and social support along with interven- tions and policies that target mental illness, addictions, and the other determinants of homelessness, such as housing, income, and employ- ment. With this understanding, there is a greater opportunity for inter- ventions and policies to address homelessness and the mental health and mental illness issues affecting the homeless. This chapter contains extracts from the CIHI report titled Improving the Health of Canadians: Mental Health and Homelessness, released August 2007. This cha pter is printed with the permission of CIHI. The full report was prepared by CPHI/CIHI staff Elizabeth Votta, Nadine Valk, Keith Denny, Ste- phanie Paolin and Anne Markhauser, as part of CPHI?s Improving the Health of Canadians three-report series on mental health and resilience. It can be found at http://www.cihi.ca/cphi. ? 200 7 Canadian Institute for Health Info rmation. References Acorn, S. (1993). Mental and physical health of homeless persons who use emer- gency shelters in Vancouver. Hospital & Community Psychiatry , 44(9), 854- 857. Alberta Mental Health Board (2004). Advancing the mental health agenda: A provin- cial mental health plan for Alberta. Edmonton: Alberta Mental Health Board. Aubry, T., Cousins, B., & Flynn, R. (2006). The community mental health evaluation initiative in Ottawa (pp. 1-2). Ottawa: University of Ottawa Centre for Re- search on Community Services. Ayerst, S. L. (1999). Depression and stress in street youth. Adolescence, 34(135), 567? 575. Baron, S. W. (1999). Street youths and substance use: The role of background, street lifestyle, and economic factors. Youth & Society , 31(1), 3? 26. Begin, P., Casavant, L., Chenier, N. M., & Dupuis, J. (1999) Homelessness. Ottawa: Parliamentary Research Branch. Retrieved June 26, 2006, from http://www.parl.gc.ca/information/library/PRBpubs/prb991-e.pdf. Boivin, J. F., Roy, E., Haley, N., & du Fort, G. G. (2005). The health of street youth: A Canadian perspective. Canadian Journal of Public Health, 96(6), 432? 437. Borg, M., Sells, D., Topor, A., Mezzina, R., Marin, I., & Davidson, L. (2005). What makes a house a home: The role of material resources in recovery from se- vere mental illness. American Journal of Psychiatric Rehabilitation, 8(3), 243? 256. British Columbia Ministry of Health and Ministry Responsible for Seniors (2002). B.C.?S mental health reform: Housing?best p ractices. Vancouver, B.C. British Columbia Ministry of Social Development and Economic Security (2001). Homelessness?c auses & effects: The cost of homelessness in British Columbia. Vancouver, B.C. Bunting, T., Walks, R. A., & Filion, P. (2004). The uneven geography of housing affordability stress in Canadian metropolitan areas. Housing Studies, 19(3), 361? 393. Cairney, J. (2005). Housing tenure and psychological well-being during adoles- cence. Environment and Behavior, 37(4), 552? 564. Cameron, K. N., Racine, Y., Offord, D. R., & Cairney, J. (2004), Youth at risk of homelessness in an affluent Toronto suburb. Canadian Journal of Public Health, 95(5), 352? 356. Canadian Institute for Health Information (2007). Improving the health of Cana- dians: Mental health and homelessness. Ottawa. Canadian Institute for Health Information (2006a). Health indicators. Ottawa. Re- trieved April 30, 2007, from http://secure.cihi.ca/indicators/june_2006/en/hlthind06_e.shtml. Canadian Institute for Health Information (2006b). Improving the health of Cana- dians: An introduction to health in urban places. Ottawa. Canadian Mental Health Association (1997). Mental health and homelessness: A CMHA Ontario position paper (pp. 1? 4). Ottawa. Canadian Mental Health Association (2007). Our programs: Independent living support, Calgary region. Ottawa. Retrieved April 11, 2007, from http://www.cmha.calgary.ab.ca/programs/independent_living_support.asp x Caton, C. L. (1995). Mental health service use among homeless and never- homeless men with schizophrenia. Psychiatric Services, 46(11), 1139? 1143. Centre for Addiction and Mental Health (2003). Current issues in addictions and mental health. Toronto, Ontario. Retrieved April 23, 2007, from http://www.camh.net/Public_policy/Public_policy_papers /current_issues_p aper03.pdf.pdf. Centre for Addiction and Mental Health (2006). Information about concurrent disorders. Retrieved April 23, 2007, from http://www.camh.net/About_Addiction_Mental_Health/ Concurrent_ Disor ders/index. Centre for Addiction and Mental Health (2007). More intensive and specialized support. Retrieved April 27, 2007, from http://www.camh.net/Care_Treatment/Resources_clients_families_ friends/Challenges_and_Choices/challenges_choices_intspecsuppor.html . Centre for Addiction and Mental Health (2005). Types of housing: Suppor- tive/supported housing. Retrieved April 30, 2007, from http://www.camh.net/Care_Treatment/Community_and_social_supports/H ousing/Types_of_Housing/csru_supportive_ed_housing.html . Commission on the Future of Health Care in Canada (2002). Building on values: The future of health care in Canada. Ottawa: Health Canada. Community Mental Health Evaluation Initiative (2003, Summer). Community Mental Health Evaluation Initiative. Helping homeless people with mental ill- ness, 7, 1? 3. Ontario Community Mental Health Evaluation Initiative (2007). Making a differ- ence: Ontario?s Community Mental Health Evaluation Initiative. Toronto. Retrieved May 24, 2007, from www.ontario.cmha.ca/cmhei/images/report/Making_a_Difference.pdf. Compas, B. (1997). Effortful and involuntary responses to stress: Implications for coping with chronic stress. In B. H. Gottlieb (Ed.) Coping with chronic stress (pp. 105? 130). New York: Plenum Press. D?Amore, J., Hung, O., Chiang, W., & Goldfrank, L. (2001). The epidemiology of the homeless population and its impact on an urban emergency depart- ment. Academic Emergency Medicine 8(11), 1051? 1055. Dickey, B. (2000). Review of programs for persons who are homeless and mental- ly ill. Harvard Review of Psychiatry, 8(5), 242? 250. Drake, R. E., Osher, F. C., & Wallach, M. A. (1991). Homelessness and dual diag- nosis. American Psychologist, 46(11), 1149? 1158 (Reprinted from Alcohol and drug abuse and mental health problems among homeless persons: A re- view of the literature, 1980? 1990, by P. J. Fisher, 1990, Rockville, MD: Na- tional Institute on Alcohol Abuse and Alcoholism and National Institute of Mental Health). Ennett, S. T., Bailey, S. L., & Federman, E. B. (1999). Social network characteristics associated with risky behaviors among runaway and homeless youth. Jour- nal of Health and Social Behavior, 40(1), 63? 78. Farrell, S. J., Huff, J., MacDonald, S.-A., Middlebro, A., & Walsh, S. (2005). Taking it to the street: A psychiatric outreach service in Canada. Community Mental Health Journal, 41(6), 737? 746. Farrell, S. J., Aubry T., & Reissing, E. (2001). Street needs assessment: An investiga- tion of the characteristics and service needs of persons who are homeless and not currently using emergency shelters in Ottawa. Ottawa: University of Ottawa. Federation of Canadian Municipalities (2004). Incomes, shelter and necessities (Quality of life in Canadian communities, theme report #1) . Ottawa. Frankish, C. J., Hwang, S. W., & Quantz, D. (2005). Homelessness and health in Canada: Research lessons and priorities. Canadian Journal of Public Health, 96(Suppl. 2), 23? 29. Goering, P., Tomiczenko, G., Sheldon, T., Boydell, K., & Wasylenki, D. (2002). Characteristics of persons who are homeless for the first time. Psychiatric Services, 53(11), 1472? 1474. Government of Canada (2007). Budget 2007: Preserving our environment and mod er- nizing our health care system. Ottawa: Public Works and Government Servic- es Canada. Government of Newfoundland and Labrador (2001). Valuing mental health: A framework to support the development of a provincial mental health policy for Newfoundland and Labrador. St. John?s, NL: Department of Health and Community Services. Gulcur, L., Stefancic, A., Shinn, M., Tsemberis, S., & Fischer, S. N. (2003). Hous- ing, hospitalization and cost outcomes for homeless individuals with psy- chiatric disabilities participating in Continuum of Care and Housing First programmes. Homelessness: Integrating international perspectives [Special issue] . Journal of Community & Applied Social Ps ychology, 13(2), 171? 186. Herman, N. J., & Smith, C. M. (1989). Mental hospital depopulation in Canada: Patient perspectives. Canadian Journal of Psychiatry, 34(5), 386? 391. Hwang, S. W. (2001). Homelessness and health. Canadian Medical Association Journal, 164(2), 229? 233. Johnsen, M., Samberg, L., Calsyn, R., Blasinsky, M., Landow, W., & Goldman, H. (1999). Case management models for persons who are homeless and men- tally ill: The ACCESS Demonstration Project. Community Mental Health Journal, 35(4), 325? 346. Johnson, K. D., Whitbeck, L. B., & Hoyt, D. R. (2005a). Predictors of social net- work composition among homeless and runaway adolescents. Journal of Adolescence, 28(2), 231? 248. Johnson, K. D., Whitbeck, L. B., & Hoyt, D. R. (2005b). Substance abuse disorders among homeless and runaway adolescents. The Journal of Drug Issues, 35(4), 799? 816. Kidd, S. A. (2004). The walls were closing in, and we were trapped: A qualitative analysis of street youth suicide. Youth & Society , 36(1), 30? 55. Kushel, M. B., Perry, S., Bangsberg, D., Clark, R., & Moss, A. R. (2002). Emergen- cy department use among the homeless and marginally housed: Results from a community-based study. American Journal of Public Health, 92(5), 778? 784. Kushel, M. B., Vittinghoff, E., & Haas, J. S. (2001). Factors associated with the health care utilization of homeless persons. Journal of the American Medical Association, 285(2), 200? 206. Lehman, A. F., Dixon, L. B., Kernan, E., DeForge, B. R., & Postrado, L. T. (1997). A randomized trial of assertive community treatment for homeless persons with severe mental illness. Archives of General Psychiatry, 54(11), 1038? 1043. Levine, I. S. (1984). Homelessness: Its implications for mental health policy and practice. Psychosocial Rehabilitation Journal, 8(1), 6? 16. Manitoba Health, Mental Health and Addictions Branch (n.d.). Guide to the Men- tal Health System in Manitoba. Retrieved April 25, 2007, from http://www.gov.mb.ca/health/mh/system.html. McCreary Centre Society (2007). Against the odds: A profile of marginalized and street-involved youth in BC. Vancouver, B.C. Mental Health Policy Research Group (1997, November 3). Mental illness and pathways into homelessness: Findings and implications. Paper presented at the Mental Illness and Pathways into Homelessness Conference, Toronto. Miner, M .H. (1991). The self-concept of homeless adolescents. Journal of Youth and Adolescence, 20(5), 545? 560. Mueser, K. T. Salyers, M. P., Rosenberg, S. D., Goodman, L. A., Essock, S. M., Osher, F. C., Swartz, M. S., & Butterfield, M. I. (2004). Interpersonal trauma and posttraumatic stress disorder in patients with severe mental illness: Demographic, clinical, and health correlates. Schizophrenic Bulletin, 30(1), 45? 57. Murray, R. B. (1996). Stressors and coping strategies of homeless men. Journal of Psychosocial Nursing and Mental Health Services, 34(8), 16? 22. Nelson, G. (2006). Mental health policy in Canada. In A. Westhues (Ed.) Canadian social policy: Issues and perspectives (pp. 245? 260). Waterloo, Ontario: Wilfrid Laurier University Press. Nelson, G., Sylvestre, J., Aubry, T., George, L. & Trainor, J. (2007). Housing choice and control, housing quality, and control over professional support as contributors to the subjective quality of life and community adaptation of people with severe mental illness. Administration and Policy in Mental Health & Mental Health Services Research , 34(2), 89? 100. Nyamathi, A. M., Leake, B., Keenan, C., & Gelberg, L. (2000a). Type of social support among homeless women: Its impact on psychosocial resources, health and health behaviors, and use of health services. Nursing Research, 49(6), 318? 326. Nyamathi, A. M., Stein, J. A., & Bayley, L. J. (2000b). Predictors of mental distress and poor physical health among homeless women. Psychology & Health , 15(4), 483? 500. Padgett, D. K., Struening, E. L., Andrews, H., & Pittman, J., (1995). Predictors of emergency room use by homeless adults in New York City: The influence of predisposing, enabling and need factors. Social Science & Medicine, 41(4), 547? 556. Padgett, D. K., Gulcur, L., & Tsemberis, S. (2006). Housing First services for people who are homeless with co-occurring serious mental illness and sub- stance abuse. Research on Social Work Practice, 16(1), 74? 83. Pomeroy, S. (2005). The cost of homelessness: Analysis of alternate responses in four Canadian cities. Ottawa: Focus Consulting. Public Health Agency of Canada (2007). Canadian Health Network. Retrieved May 15, 2007, from http://www.canadian-health-network.ca. Public Health Agency of Canada (2006). The human face of mental health and mental illness in Canada 2006 . Ottawa: Minister of Public Works and Government Services Canada. Public Health Agency of Canada (2004). What Determines Health? . Retrieved April, 9, 2007, from http://www.phac-aspc.gc.ca/ph-sp/determinants/index- eng.php. Randolph, F., Blasinsky, M., Leginski, W., Parker, L. B., & Goldman, H. H. (1997). Creating integrated service systems for homeless persons with mental ill- ness: The ACCESS Program. Psychiatric Services, 48(3), 369? 373. Rohde, P., Noell, J., Ochs, L., & Seeley, J. R. (2001). Depression, suicidal ideation and STD-related risk in homeless older adolescents. Journal of Adolescence, 24(4), 447? 460. Rosenheck, R. A., Lam. J., Morrissey, J. P., Calloway, M. O., Stolar, M., & Ran- dolph, F. (2002). Service systems integration and outcomes for mentally ill homeless persons in the ACCESS Program. Psychiatric Services, 53(8), 958? 966. Rothschild, B. (2000). Overview of Posttraumatic stress disorder (PTSD): The impact of trauma on body and mind. In The body remembers: The psychology of trauma and trauma treatment (pp. 3? 14). New York: W. W. Norton & Com- pany. Roy, ?., Haley, N., Leclerc, P., C?dras, L., Blais, L., & Boivin, J.-F. (2003). Drug injection among street youths in Montreal: Predictors of initiation. Journal of Urban Health, 80(1), 92? 105. Sant? et S ervices Sociaux Qu?bec. (2005). Plan d?action en sante mentale: La force des liens 200 5 ?2010 . Quebec. Sealy, P., & Whitehead, P. C. (2004). Forty years of deinstitutionalization of psy- chiatric services in Canada: An empirical assessment. Canadian Journal of Psychiatry, 49(4), 249? 257. Shortt, S., Hwang, S., & Stuart, H. (2006). Homelessness and health: A policy synthe- sis on approaches to delivering primary care for homeless persons. Kingston, ON: Centre for Health Services and Policy Research, Queen?s University. Smart, R. G., & Walsh, G. W. (1993). Predictors of depression in street youth. Ado- lescence, 28(109), 41? 53. Smart, R. G., & Ogborne, A. C. (1994). Street youth in substance abuse treatment: Characteristics and treatment compliance. Adolescence, 29(115), 733? 745. Solorio, M. R., Milburn, N. G., Andersen, R. M., Trifskin, S., & Rodriguez, M. A. (2006). Emotional distress and mental health service use among urban homeless adolescents. Journal of Behavioral Health Services and Research, 33(4), 381? 393. Standing Senate Committee on Social Affairs Science and Technology (2006). Out of the shadows at last: Transforming mental health, mental illness and addiction services in Canada. Ottawa: The Senate. Susser, E., Conover, S., & Struening, E. L. (1989). Problems of epidemiologic me- thod in assessing the type and extent of mental illness among homeless adults. Hospital & Community Psychiatry , 40(3), 261? 265. Thoits, P. A. (1995). Stress, coping, and social support processes: Where are we? What next?. Forty years of medical sociology: The state of the art and direc- tions for the future [Extra issue]. Journal of He alth and Social Behaviour, 35, 53? 79. Tsemberis, S., & Eisenberg, R. F. (2000). Pathways to housing: Supported housing for street-dwelling homeless individuals with psychiatric disabilities. Psy- chiatric Services, 51(4), 487? 493. Tsemberis, S. J., Moran, L., Shinn, M., Asmussen, S. M., & Shern, D. L. (2003). Consumer preference programs for individuals who are homeless and have psychiatric disabilities: A drop-in center and a supported housing program. American Journal of Community Psychology, 32(3? 4), 305? 317. Tsemberis, S., Gulcur, L., & Nakae, M. (2004). Housing first, consumer choice, and harm reduction for homeless individuals with a dual diagnosis. Ameri- can Journal of Public Health, 94(4), 651? 656. Unger, J. B., Kipke, M. D., Simon, T. R., Montgomery, S. B., & Johnson, C. J. (1997). Homeless youths and young adults in Los Angeles: Prevalence of mental health problems and the relationship between mental health and substance abuse disorders. American Journal of Community Psychology, 25(3), 371? 394. Unger, J. B., Kipke, M. D., Simon, T. R., Johnson, C. J., Montgomery, S. B., & Iver- son, E. (1998). Stress, coping, and social support among homeless youth. Journal of Adolescent Research, 13(2), 134? 157. Votta, E., & Manion, I. G. (2003). Factors in the psychological adjustment of homeless adolescent males: The role of coping style. Journal of the American Academy of Child and Adolescent Psychiatry, 42(7), 778? 785. Votta, E., & Manion, I. (2004). Suicide, high-risk behaviours, and coping style in homeless adolescent males? adjustment. The Journal of Adolescent Health, 34(3), 237? 243. Wong, Y.-L. I., & Piliavin, I. (2001). Stressors, resources, and distress among homeless persons: A longitudinal analysis. Social Science & Medicine , 52(7), 1029? 1042. Chapter 2.4 The Health of Street Youth in Canada: A Review of the Literature JEAN - F RAN ? O IS B O IVIN , ? L ISE R OY , NANCY HALEY, AND GUILLAUME GALBAUD DU F O RT Street youth are exposed to a number of factors that may detrimentally affect their health, including unsafe sexual practices, drug use, poor diet, inadequate shelter, exposure to violence, low levels of social support, and limited access to medical care (Noell et al., 2001a; Rohde et al., 2001). In recent literature, the term street youth has been used to describe youth living or working on the streets of major urban centres, and it is usually associated with varying degrees of homelessness. In 1998, the Canadian Paediatric Society indicated that estimates of the number of runaways in Canada ranged from 45,000 to 150,000 (Canadian Paediatric Society, 1998), for a population of approximately four million subjects in the age group of 10 to 19 years. There are, however, considerable difficulties in arriving at such estimates (Ringwalt et al., 1998), and these figures represent only expert opinion. Epidemiologic studies of the health status of street youth are rela- tively recent. In 1989, the Council on Scientific Affairs of the American Medical Association published a report on the health care needs of homeless and runaway youth (Council on Scientific Affairs, 1989); only one peer-reviewed epidemiologic study was cited by the Council at that time (McCormack et al., 1986). Since then, numerous studies have been conducted. The objective of the current paper is to review the existing scientific knowledge on the health status of street youth, with a specific focus on Canadian data. Methods We identified the epidemiologic studies for our review from searches of the MEDLINE database and the bibliographies of published papers. The keywords used in MEDLINE searches were: ?homeless youth,? ?street youth,? and ?runaways.? We excluded studies of homeless youth when these focused on young people living with their homeless family. We did not include technical reports and other documents not subjected to peer review by scientific journals. The main health outcomes assessed were blood-borne and sexually transmitted infections, mental health problems, pregnancy, violence and mortality. We concentrated on research that included teenagers. We allowed, however, broader age definitions, from the pre-teens to 30 years, as long as adolescents were also included. We focused on studies on Canada and other countries with some- what similar cultural and social contexts, namely the United States, the United Kingdom and Australia. We restricted our search to the peer- reviewed literature published between 1980 and 2003. Throughout our review, we paid particular attention to the compar- ison of street youth data to reference data for non-street youth. In the case of infectious diseases, for which the reviewed papers generally did not include any non-street comparison group, we sought reference fig- ures from the published literature. For the other health outcomes, we relied on data (if any) provided by the authors of the reviewed papers. Table 1: Prevalence and Inc idence of Infectious Disease Markers Among Canadian Street Youth and Comparison Populations Infectious diseases We identified 16 reports providing prevalence (Alderman et al., 1998; DeMatteo et al., 1999; Haley et al., 2002; Noell et al., 2001a; Ochnio et al., 2001; Pfeifer & Oliver, 1997; Rouget et al., 1994; Roy et al., 1999, 2000, 2001, 2002a; Sherman, 1992; Stricof et al., 1991; Sweeney et al., 1995; Wang et al., 1991) or incidence (Noell et al., 2001a; Roy et al., 2003) esti- mates for markers of past or present infectious diseases in street youth, all based on laboratory tests. Table 1 presents results from Canadian studies. We also present comparison figures, based on data cited by the authors of the reviewed papers or from papers identified through other sources such as the Health Canada Population and Public Health Branch website (Glasgow et al., 1997; Levy et al., 2001; Miller et al., 2004; Renzullo et al., 2001; Ro- thon et al., 1997; Sweeney et al., 1995; Zou et al., 2000). Some of these comparison figures are drawn from American studies, because appro- priate Canadian figures could not always be identified. Results are pre- sented by age subgroups where available. These data indicate that prevalence of hepatitis B and hepatitis C are significantly higher among street youth than among non-street per- sons of similar age; there is also an indication of an increased prevalence of Chlamydia trachomatis genital infection among younger subjects. On the other hand, the prevalence of hepatitis A is not increased. Table 1 also gives estimates of the prevalence and of the incidence of HIV infec- tion. These data suggest that HIV infection is also higher among street youth. It was particularly difficult, however, to identify comparison fig- ures for HIV infection. For prevalence, Table 1 gives two comparison figures, one for British Columbia young offenders, and one for U.S. sen- tinel adolescent clinics; in both cases, however, these comparison esti- mates were restricted to youth below 20 years of age. Fragmentary evi- dence based on AIDS cases reported to the Centre for Infectious Disease Prevention and Control (Health Canada, 2004) suggests that the HIV in- fection prevalence observed in older street youth (20 to 24 years old) is also in excess of expectation, but no data confirming this impression were found. For the incidence of HIV infection, we compared street youth data to incidence estimates for U.S. army personnel (Renzullo et al., 2001), and rates were higher for street youth in each age category. These results must be interpreted with caution, since the studies of street youth and those of non-street youth used different recruitment and diagnostic methods, and since different geographic locations are being compared. Table 2: Risk Factors Associated with Infectious Disease Markers in Montreal Street Youth Multivariable analyses of risk factors for infections have been re- ported for street youth from Vancouver (hepatitis A), Toronto (hepatitis B), and Montreal (hepatitis A, B, and C, and HIV infection). The Vancou- ver study included street youth, injection drug users, and men who have sex with men (Ochnio et al., 2001), and the prevalence of hepatitis A was higher in subjects born in countries with high rates of hepatitis A. The Toronto study included street youth as well as adolescents who lived with their family (Wang et al., 1991); the number of lifetime sexual part- ners and the practice of anal intercourse were associated with the pres- ence of hepatitis B markers. Table 3: Prevalence of Menta l Health Disorders in Street Youth Compared to Non -street Youth In the Montreal study, analyses were restricted to street youth (Roy et al., 1999, 2000, 2001, 2002a, 2003). The prevalence of hepatitis B, hepati- tis C, and HlV infection markers increased with age. Drug injection was associated with hepatitis B, hepatitis C, and HIV infection. Crack cocaine use was associated with hepatitis C, and prostitution with HIV infection. More detailed results are provided in Table 2. Mental health and addiction We identified 25 surveys of mental health problems among street youth (Adlaf et al., 1996; Booth & Zhang, 1996; Cauce et al., 2000; Dadds et al., 1993; Feitel et al., 1992; Greenblatt & Robertson, 1993; Greene & Ring- walt, 1996; Hier et al., 1990; Kipke et al., 1997; McCarthy & Hagan, 1992; McCaskill et al., 1998; McCormack et al., 1986; Molnar et al., 1989; Mun- dy et al., 1990; Rohde et al., 2001; Rotheram-Borus, 1993; Shade et al., 1998; Smart & Adlaf, 1991; Smart et al., 1994; Smart & Walsh, 1993; Stiff- man, 1989; Warheit & Biafora, 1991; Whitbeck et al., 2000; Unger et al., 1997; Yoder, 1999). Some investigators have used standardized survey instruments to assess prevalence of mental health problems, while others have modified existing instruments or developed their own. Some instruments, such as the Diagnostic Interview Schedule (Robins et al., 1981), the Schedule for Affective Disorders and Schizophrenia for School-Age Children, Present Episode (Chambers et al., 1985), and the Diagnostic Interview Schedule for Children Version 2.3 (Shaffer et al., 1996) are compatible with diag- nostic criteria of the Diagnostic and Statistical Manual of Mental Disord- ers (DSM) of the American Psychiatric Association (1987), while other instruments were not designed with this purpose in mind. Fifteen of the 25 reviewed studies included comparisons of data be- tween street and non-street youth, the latter group either from within the same study or drawn from the literature (Cauce et al., 2000; Dadds et al., 1993; Greene & Ringwalt, 1996; Greenblatt & Robertson, 1993; Hier et al., 1990; Kipke et al., 1997a; McCaskill et al., 1998; Robertson et al., 1989; Rohde et al., 2001; Smart & Adlaf, 1991; Smart et al., 1994; Stiffman, 1989; Warheit & Biafora, 1991; Whitbeck et al., 2000; Yoder, 1999). Fifteen studies assessed correlates of mental health problems in street youth using multivariable statistical models (Adlaf et al., 1996; Booth & Zhang, 1996; Cauce et al., 2000; Greene & Ringwalt, 1996; Hier et al., 1990; Kipke et al., 1997a; McCarthy & Hagan, 1992; Molnar et al., 1998; Mundy et al., 1990; Smart & Adlaf, 1991; Smart & Walsh, 1993; Stiffman, 1989; Unger et al., 1997; Whitbeck et al., 2000; Yoder, 1999). Table 3 summarizes the results of the only three surveys of street youth providing DSM-compatible diagnoses and presenting compari- sons of prevalence estimates between street and non-street youth (Le- winsohn et al., 1993; McCaskill et al., 1998; Regier et al., 1988; Rohde et al., 2001; Warheit & Biafora, 1991). These three American studies are pre- sented here because no equivalent study was identified for Canadian youth. In these studies, the prevalence estimates for the mental health dis- orders were always higher (to some extent) in street youth than in com- parison populations. Some of the results shown in Table 3 suggest a so- cial class effect, however. In the Fort Lauderdale and Oregon studies, prevalence figures among street youth were compared to those in gener- al populations of subjects of similar ages, and differences were marked. By contrast, in the Detroit study, McCaskill et al. (1998) matched home- less and housed adolescents for neighbourhood, and prevalence of alco- hol abuse and dependence, and of depression/dysthymia were some- what closer in value. Table 4: Victimization of Runaway Youth in Toronto (n=18 7) (Janus et al., 1995 ) The general pattern of increased prevalence of mental health prob- lems described above is also reflected in other studies conducted in Can- ada, the United States, and Australia, using scales not designed to yield DSM-compatible diagnoses (Dadds et al., 1993; McCaskill et al., 1998; Rotheram-Borus, 1993; Smart et al., 1994; Stiffman, 1989). Canadian re- sults are summarized here. In Toronto, Smart et al. (1994) compared 217 street youth to 199 stu- dents with respect to depression and alcohol problems, using the CAGE questionnaire (Mayfield et al., 1974) and items from the Centre for Epi- demiologic Studies Depression Scale (Radloff, 1977). Greater percentages of street youth reported alcohol problems and feelings of depression. Smart et al. (1993) also reported that low self-esteem and the number of months having lived in a hostel were associated with higher depression scores. In other analyses of Toronto subjects, the number of previous street experiences and length of time on the street were associated with suicide attempt (McCarthy & Hagan, 1992). Pregnancy Greene and Ringwalt (1998) compared pregnancy histories of three groups of female youth aged 14 to 17 years in the United States: a repre- sentative sample of 169 runaway and homeless youth residing in 23 funded shelters in metropolitan areas, a convenience sample of 85 street youth living in 10 American cities, and a nationally representative sam- ple of 1,609 household youth included in the 1992 National Health Inter- view Survey. Youth living on the street had the highest lifetime occur- rence of pregnancy (48.2 percent), followed by youth residing in shelters (33.2 percent), and household youth (7.2 percent). Twenty percent of the street youth, 12.6 percent of the shelter youth, and 1.5 percent of the household youth reported two or more pregnancies. No equivalent study, comparing street and household youth in Canada, was identified. Victimization while on the street Street youth experience high levels of violence and victimization of vari- ous kinds, both before leaving home and while on the street (Janus et al., 1995; Kipke et al., 1997; Kufeldt & Nimmo, 1987; Noell et al., 2001b; Rohde et al., 2001; Whitbeck et al., 1997). Results presented in Table 4 confirm the importance of this phenomenon in Toronto: a very large proportion of runaway youth reported being physically abused or as- saulted, threatened, or subjected to other similar abuse during street liv- ing. In Calgary, more than 50 percent of a sample of 489 runaway and homeless youth indicated having been approached to participate in il- legal activities (Kufeldt & Nimmo, 1987). Mortality Street youth experience high mortality rates (Hwang, 2000; Hwang et al., 1997; Roy et al., 1998; Roy et al., 2002b; Shaw & Dorling, 1998). In Mon- treal, the mortality rate among 1,013 street youth over a two-year follow- up period was 0.89 deaths per 100 person-years, which corresponded to 11 times the rate expected for subjects of corresponding age and sex in the province of Quebec (Roy et al., 1998, 2002b). Twenty-six deaths were observed, including 13 suicides, 8 associated with overdose, and 2 trau- matic deaths. In Toronto, the age-adjusted mortality rate ratio was 8.3, comparing men 18 to 24 years old using homeless shelters to men in the general population; the leading causes of death were unintentional poi- sonings, other accidents, and suicide (Hwang, 2000). Discussion Our review indicates that street youth are affected by several problems, including infections, mental health disorders, and high mortality. Epi- demiologic studies quantifying specific disease risks in street youth, however, remain limited; only a single estimate, for example, is currently available on the incidence of HIV infection (Roy et al., 2003). Studies of mental health problems present several important limitations. Only three of the 25 studies we reviewed on this topic included a comparison group of non-street youth (Dadds et al., 1993; McCaskill et al., 1998; Smart et al., 1994). Of the remaining 22 studies, only 12 provided a comparison of their results for street youth with literature results for non-street young people (Greenblatt & Robertson, 1993; Greene & Ringwalt, 1996; Hier et al., 1990; Kipke et al., 1997a; Robertson et al., 1989; Rotheram-Borus, 1993; Smart & Adlaf, 1991; Stiffman, 1989; Warheit & Biafora, 1991; Whitbeck et al., 2000; Yoder, 1999). No longitudinal studies providing incidence data for mental health problems appear to exist. Similarly, the important question of victimization of street youth remains poorly inves- tigated: research instruments require further development, standardiza- tion, and validation and studies comparing the experience of street and non-street youth are needed. No or very limited data are available on various other outcomes such as dental health, reproductive history, and various infections. The need for Canadian data is particularly acute in specific areas. Only 6 of the 25 reviewed studies on mental health problems were con- ducted in Canada (Adlaf et al., 1996; McCarthy & Hagan, 1992; McCor- mack et al., 1986; Smart & Adlaf, 1991; Smart & Walsh, 1993; Smart et al., 1994) and none assessed DSM-compatible psychiatric diagnoses. No study of youth pregnancy, comparing street and non-street young people, has been reported in Canada. As well, no data are available on important sexually transmitted infections such as herpes virus infection and syphilis. Our review presents several limitations. The street youth popula- tions under study were very heterogeneous. The general epidemiologic profile of the different urban populations among which street youth live also differs, thereby affecting risks for various diseases and the interpre- tability of some results. Comparison populations of non-street youth were rarely included in the reviewed studies, and comparative figures obtained from other sources are affected by various limitations such as differences in geographic areas covered and age groups included. In summary, current research results are useful to orient public health interventions for street youth, but further epidemiologic research is required to better define the needs of this vulnerable population. Jean- Fran?ois Boivin is a Professor in the Department of Epidemiology & Biostatistics, at McGill University. Elise Roy is a physician specializing in public health and adjunct professor at McGill University. Nancy H aley is a physician with the Montreal Public Health Department. G ui llaume Galbaud du Fort is a psychiatric epidemiologist based at the Clinical Epidemiology Unit of the Sir Mortimer B. Davis-Jewish General Hospital and the Culture and Mental Health Research Unit. References Adlaf, E. M., Zdanowicz, Y., & Smart, R. G. (1996). Alcohol and other drug use among street-involved youth in Toronto. Addiction Research, 4(1), 11-24. Alderman, E. M., Shapiro, A., Spigland, I., Coupey, S. M., Bashir, M., & Fox, A. S. (1998). Are there risk factors for hepatitis B infection in inner-city adoles- cents that justify prevaccination screening? Journal of Adolescent Health, 22(5), 389-393. American Psychiatric Association. (1987). Diagnostic and statistical manual of men- tal disorders (3rd ed.). Washington, D.C.: Author. Booth, R. E., & Zhang, Y. (1996). Severe aggression and related conduct problems among runaway and homeless adolescents. Psychiatric Services, 47(1), 75-80. Canadian Paediatric Society. (1998). Bringing street youth out of the shadow. CPS News, May-June, 5-6. Cauce, A. M., Paradise, M., Ginzler, J. A., Embry, L., Morgan, C. J., Lohr, Y., & Theofelis, J. (2000). The characteristics and mental health of homeless ado- lescents: Age and gender differences. Journal of Emotional and Behavioral Dis- orders, 8(4), 230-239. Chambers, W. J., Puig-Amich, J., Hirsch, M., Paez, P., Ambrosini, P. J., Tabriz, M. A., & Davies, M. (1985). The assessment of affective disorders in children and adolescents by semistructured interview: Test-retest reliability of the schedule for affective disorders and schizophrenia for school-age children, present episode version. Archives of General Psychiatry, 42(7), 696-702. Council on Scientific Affairs. (1989). Health care needs of homeless and runaway youths. The Journal of the American Medical Association, 262(10), 1358-1361. Dadds, M. R., Braddock, D., Cuers, S., Elliott, A., & Kelly, A. (1993). Personal and family distress in homeless adolescents. Community Mental Health Journal, 29(5), 4l3-422. DeMatteo, D., Major, C., Block, B., Coates, R., Fearon, M., Goldberg, E., King, S. M., Millson, M., O?Saughnessy, M., & Read, S. E. (1999). Toronto street youth and HIV/AIDS: Prevalence, demographics, and risks. Journal of Ado- lescent Health, 25(5), 358-366. Feitel, B., Margetson, N., Chamas, J., & Lipman, C. (1992). Psychosocial back- ground and behavioral and emotional disorders of homeless and runaway youth. Hospital & Community Psychiatry, 43(2), 155-159. Glasgow, K. W., Schabas, R., Williams, D. C., Wallace, E., & Nalezyty, L. A. (1997). A population-based hepatitis B seroprevalence and risk factor study in a northern Ontario town. Canadian Journal of Public Health, 88(2), 87-90. Greenblatt, M., & Robertson M. J. (1993). Life-styles, adaptive strategies, and sex- ual behaviors of homeless adolescents. Hospital & Community Psychiatry , 44(12), 1177-1180. Greene, J. M., & Ringwalt, C. L. (1996). Youth and familial substance use's asso ci- ation with suicide attempts among runaway and homeless youth. Substance Use & Misuse, 3l(8), 1041-1058. Greene, J. M., & Ringwalt, C. L. (1998). Pregnancy among three national samples of runaway and homeless youth. Journal of Adolescent Health, 23(6), 370-377. Haley, N., Roy, E., Leclerc, P., Lambert, G., Boivin, J. F., Cedras, L., & Vincelette, J. (2002). Risk behaviours and prevalence of Chlamydia trachomatis and Neisseria gonorrhoeae genital infections among Montreal street youth. In- ternational Journal of STD & AIDS, 13(4), 238-245. Health Canada. (2004). Surveillance and Risk Assessment Division, Centre for Infectious Disease Prevention and Control. HlV/AlDS EPl Updates, May 2004. Ottawa, ON. Hier, S. J., Korboot, P. J., & Schweitzer, R. D. (1990). Social adjustment and symp- tomatology in two types of homeless adolescents: Runaways and throwa- ways. Adolescence, 25(100), 761-771. Hwang, S. W. (2000). Mortality among men using homeless shelters in Toronto, Ontario. The Journal of the American Medical Association, 283(16), 2152-2157. Hwang, S. W., Orav, E. J., O'Connell , J. J., Lebow, J. M., & Brennan, T. A. (1997). Causes of death in homeless adults in Boston. Annals of Internal Medicine, 126(8), 625-628. Janus, M. D., Archambault, F. X. , Brown, S. W., & Welsh, L. A. (1995). Physical abuse in Canadian runaway adolescents. Child Abuse & Neglect, 19(4), 433- 447. Kipke, M. D., Montgomery, S. B., Simon, T.R., & Iverson, E. F. (1997a). ?Sub- stance abuse? disorders among runaway and homeless youth. Substance Use & Misuse, 32(7-8), 969-986. Kipke, M. D., Simon, T. R., Montgomery, S. B., Unger, J. B., & Iversen, E. F. (1997b). Homeless youth and their exposure to and involvement in violence while living on the streets. Journal of Adolescent Health, 20(5), 360-367. Kufeldt, K., & Nimmo, M. (1987). Youth on the street: Abuse and neglect in the eighties. Child Abuse & Neglect, 11(4), 531-543. Levy, I., Chen, D., Sherman, M., Smith, D., & Krajden, M. (2001). Hepatitis A vi- rus seroprevalence in 1,000 university students in Toronto. Canada Commu- nicable Disease Report, 27(11), 93-96. Lewinsohn, P. M., Hops, H., Roberts, R. E., Seeley, J. R., & Andrews, J.A. (1993). Adolescent psychopathology: I. Prevalence and incidence of depression and other DSM-III-R disorders in high school students. Journal of Abnormal Psy- chology, 102(4), 133-144. Mayfield, D., McLeod, G., & Hall, P. (1974). The CAGE questionnaire: Validation of a new alcoholism screening instrument. American Journal of Psychiatry, 131(10), 1121-1123. McCarthy, B., & Hagan, J. (1992). Surviving on the street: The experience of homeless youth. Journal of Adolescent Research, 7(4), 412-430. McCaskill, P. A., Toro, P. A., & Wolfe, S. M. (1998). Homeless and matched housed adolescents: A comparative study of psychopathology. Journal of Clinical Child Psychology, 27(3), 306-319. McCormack, A., Janus, M. D., & Burgess, A. W. (1986). Runaway youths and sexual victimization: Gender differences in an adolescent runaway popula- tion. Child Abuse and Neglect, 10(3), 387-395. Miller, W. C., Ford, C. A., Morris, M., Handcock, M. S., Schmitz, J. L., Hobbs, M. M., Cohen, M. S., Harris, K. M., & Udry, J. R. (2004). Prevalence of chla- mydial and gonococcal infections among young adults in the United States. The Journal of the American Medical Association, 291(18), 2229-2236. Molnar, B. E., Shade, S. B., Krai, A. H., Booth, R. E., & Waiters, J. K. (1998). Sui- cidal behavior and sexual/physical abuse among street youth. Child Abuse & Neglect, 22(3), 213-222. Mundy, P., Robertson, M., Robertson, J., & Greenblatt, M. (1990). The prevalence of psychotic symptoms in homeless adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 29(5), 724-731. Noell, J., Rohde, P., Ochs, L., Yovanoff, P., Alter, M. J., Schmid, S., Bullard, J., & Black, C. (2001a). Incidence and prevalence of chlamydia, herpes, and viral hepatitis in a homeless adolescent population. Sexually Transmitted Diseases, 28(1), 4-10. Noell, J., Rohde, P., Seeley, J., & Ochs, L. (2001b). Childhood sexual abuse, ado- lescent sexual coercion and sexually transmitted infection acquisition among homeless female adolescents. Child Abuse and Neglect, 25(1), 137-148. Ochnio, J. J., Patrick, D., Ho, M., Tailing, D. N., & Dobson, S. R. (2001). Past infec- tion with hepatitis A virus among Vancouver street youth, injection drug users and men who have sex with men: Implications for vaccination pro- grams. Canadian Medical Association Journal, 165(3), 293-297. Pfeifer, R. W., & Oliver, J. (1997). A study of HIV Seroprevalence in a group of homeless youth in Hollywood, California. Journal of Adolescent Health, 20(5), 339-342. Radloff, L. (1977). The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1(3), 385-401. Regier, D. A., Boyd, J. H., Burke, J. D., Jr., Rae, D. S., Myers, J. K., Kramer, M., Robins, L. N., George, L. K., Karno, M., Locke, B.Z. (1988). One-month pre- valence of mental disorders in the United States ? Based on five epidemi- ologic catchment area sites. Archives of General Psychiatry, 45(11), 977-986. Renzullo, P. O., Sateren, W. B., Garner, R. P., Milazzo, M. J., Birx, D. L., & McNeil, J. G. (2001). HIV-I seroconversion in United States Army active duty per- sonnel, 1985-1999. AIDS, 15(12), 1569-1574. Ringwalt, C. L., Greene, J. M., Robertson, M., & McPheeters, M. (1998). The pre- valence of homelessness among adolescents in the United States. American Journal of Public Health, 88(9), 1325-1329. Robertson, M., Koegel, P., & Ferguson, L. (1989). Alcohol use and abuse among homeless adolescents in Hollywood. Contemporary Drug Problems, 16(3), 415-452. Robins, L. N., Helzer, J. E., Croughan, J., & Ratcliff, K. S. (1981). National Institute of Mental Health diagnostic interview schedule: Its history, characteristics, and validity. Archives of General Psychiatry, 38(4), 381-389. Rohde, P., Noell, J., Ochs, L., & Seeley, J. R. (2001). Depression, suicidal ideation and STD-related risk in homeless older adolescents. Journal of Adolescence, 24(4), 447-460. Rotheram-Borus, M. J. (1993). Suicidal behavior and risk factors among runaway youths. American Journal of Psychiatry, 150(1), 103-107. Rothon, D. A., Strathdee, S. A., Cook, D., & Cornelisse, P. G. (1997). Determinants of HIV-related high risk behaviours among young offenders: A window of opportunity. Canadian Journal of Public Health, 88(1), 14-17. Rouget, A., Mah, J., Lang, R., & Joffres, M. (1994). Prevalence of sexually trans- mitted diseases in juvenile prostitutes and street youth. Canadian Journal of Infectious Diseases, 5(1), 21-27. Roy, E., Boivin, J. F., Haley, N., & Lemire, N. (1998). Mortality among street youth. Lancet, 352(9121), 32. Roy, E., Haley, N., Leclerc, P., Boivin, J. F., Cedras, L., & Vincelette, J. (2001). Risk factors for hepatitis C virus infection among street youths. Canadian Medical Association Journal, 165(5), 557-560. Roy, E., Haley, N., Lemire, N., Boivin, J. F., Leclerc, P., & Vincelette J. (1999). He- patitis B virus infection among street youths in Montreal. Canadian Medical Association Journal, 161(6), 689-693. Roy, E., Haley, N., Leclerc, P., Cedras, L., Bedard, L., & Allard, R. (2002a). Sero- prevalence and risk factors for hepatitis A among Montreal street youth. Canadian Journal of Public Health, 93(1), 52-53. Roy, E., Haley, N., Leclerc, P., Cedras, L., Weber, A., Claessens, C., & Boivin, J. F. (2003). HIV incidence among street youth in Montreal, Canada. AIDS, 17(7), 1071-1075. Roy, E., Haley, N., Leclerc, P., Lemire, N., Boivin, J. F., Frappier, J. Y., & Claes- sens, C. (2000). Prevalence of HIV infection and risk behaviours among Montreal street youth. International Journal of STD & AIDS , 11(4), 241-247. Roy, E., Haley, N., Leclerc, P., Sochanski, B., & Boivin, J. F. (2002b). Explaining mortality among street youth. In Full program and abstracts. 1st International Conference on Inner City Health, 3-6 October 2002. Toronto, Canada. Abstract Session 6, Special Populations. Shaffer, D., Fisher, P., Dulcan, M. K., Davies, M., Piacentini, J., Schwab-Stone, M. E., Lahey, B. B., Bourdon, K., Jensen, P. S., Bird, H. R., Canino, G., & Regier, D. A. (1996). The NIMH Diagnostic Interview Schedule for Children Ver- sion 2.3 (DISC-2.3): Description, acceptability, prevalence rates, and per- formance in the MECA Study. Journal of the American Academy of Child and Adolescent Psychiatry, 35(7), 865-877. Shaw, M., & Dorling, D. (1998). Mortality among street youth in the UK. Lancet, 352(9129), 743. Sherman, D. J. (1992). The neglected health care needs of street youth. Public Health Reports, 107(4), 433-440. Smart, R. G., & Adlaf, E. M. (1991). Substance use and problems among Toronto street youth. British Journal of Addiction, 86(8), 999-1010. Smart, R. G., Adlaf, E. M., Walsh, G. W., & Zdanowicz, Y. (1994). Similarities in drug use and depression among runaway students and street youth. Cana- dian Journal of Public Health, 85(1), 17-18. Smart, R. G., & Walsh, G. W. (1993). Predictors of depression in street youth. Ado- lescence, 28(109), 41-53. Stiffman, A. R. (1989). Suicide attempts in runaway youths. Suicide and Life Threatening Behavior, 19(2), 147-159. Stricof, R. L., Kennedy, J. T., Nattell, T. C., Weisfuse, I. B., & Novick, L. F. (1991). HIV Seroprevalence in a facility for runaway and homeless adolescents. American Journal of Public Health, 81(Suppl. 5), 50-53. Sweeney, P., Lindegren, M. L., Buehler, J. W., Onorato, I. M., & Janssen, R. S. (1995). Teenagers at risk of human immunodeficiency virus type 1 infec- tion: Results from seroprevalence surveys in the United States. Archives of Pediatrics & Adolescent Medicine, 149(5), 521-528. Unger, J. B., Kipke, M. D., Simon, T. R., Montgomery, S. B., & Johnson, C. J. (1997). Homeless youths and young adults in Los Angeles: Prevalence of mental health problems and the relationship between mental health and substance abuse disorders. American Journal of Community Psychology, 25(3), 371-394. Wang, E. E., King, S., Goldberg, E., Bock, B., Milner, R., & Read, S. (1991). Hepati- tis B and human immunodeficiency virus infection in street youths in To- ronto, Canada. Pediatric Infectious Disease Journal, 10(2), 130-133. Warheit, G., & Biafora, F. (1991) Mental health and substance abuse patterns among a sample of homeless post-adolescents. International Journal of Ado- lescence and Youth , 3(1-2), 9-27. Whitbeck, L. B., Hoyt, D., & Ackley, K. (1997). Abusive family backgrounds and later victimization among runaway and homeless adolescents. Journal of Re- search on Adolescence, 7(4), 375-392. Whitbeck, L. B., Hoyt, D. R., & Bao, W. N. (2000). Depressive symptoms and co- occurring depressive symptoms, substance abuse, and conduct problems among runaway and homeless adolescents. Child Development, 71(3), 721- 732. Yoder, K. A. (1999). Comparing suicide attempters, suicide ideators, and nonsui- cidal homeless and runaway adolescents. Suicide and Life-Threatening Beha- vior, 29(1), 25-36. Zou, S., Tepper, M., & Giulivi, A. (2000). Current status of hepatitis C in Canada. Canadian Journal of Public Health, 91(Suppl.), S10-S15. Chapter 2.5 Understanding the Health, Housing, and Social Inclusion of Formerly Homeless Older Adults LYNN MCDONALD , PETER D ONAHUE , JULIA JANES, AND LAURA CLEGH ORN Despite considerable research on the homeless population in Canada, relatively little is known about the characteristics, circumstances, health, housing and service needs of older homeless adults, especially after they have been housed. As the number of homeless older adults is expected to increase with the aging of the baby boomers, improving service delivery to reach this population is important. While the experience of homeless- ness impacts the health and well-being of older adults, aging adds a new dimension creating unique challenges for programming, policy and ser- vice provision. We conducted a study in two cities, Toronto and Calgary, using a mixed-method approach drawing on data from 237 survey interviews with older adults in supportive and supported housing, 53 qualitative interviews with formerly homeless people, six focus groups with former- ly homeless people and service providers, and Personal Health Informa- tion (that is, OHIP data released by the Ontario Ministry of Health and Long-term Care) from 136 consenting participants to investigate the health and housing outcomes of formerly homeless older adults. We also helped develop a working group of research participants who carried out dissemination of the results through a postcard campaign and a speakers? bureau. Our findings suggest that housing is a critical determinant of health and that health care utilization after people have been housed is asso- ciated with improved health outcomes and more effective and cost- efficient use of health care services. We also found that this population can be appropriately and stably housed in a number of different forms of housing. Clearly, investment in age-appropriate, affordable housing and supports can help formerly homeless older adults find their way ?in from the streets.? The survey group Of the 237 participants who responded to the survey in Toronto and Calgary, the majority were male, which accurately reflects the proportion of men to women in the homeless population. The average age was 57 in both samples. Most of the participants were born in Canada and identi- fied as ?white,? although Toronto had a larger percentage of immigrants and Calgary had a higher percentage of Aboriginal peoples. Most were single or divorced, over half in both samples had attended or completed high school and close to one third had attended college or university. Over 60 percent of the participants in Toronto and 56 percent of the formerly homeless participants in Calgary had been homeless more than once, with men reporting significantly more homeless episodes than women in both cities. In Toronto, 71 percent lived in supportive housing compared to 42 percent in Calgary and the remainder lived in supported housing with the help of community supports. In Toronto, about 50 percent had been housed for over five years compared to only eight percent for Calgary. The last episode of homelessness in Calgary was much shorter than for Toronto, suggesting a quicker turn-around in interventions that pro- vided support and housing. Characteristics of the Formerly Homeless The literature suggests that 50 is an appropriate demarcation of ?old? in the homeless population. ?Accelerated aging? was linked to ?homeless effects,? which emerged as a central theme in both the qualitative and focus group analyses. Participants spoke of the stressful conditions of living without housing as having not only immediate adverse impacts on their health and well-being, but lingering effects that persisted once they were housed. Poor nutrition, trauma, and lack of access to health care while homeless left participants feeling considerably older than their chronological age. Most of the participants were single, divorced, or widowed. Both male and female participants expressed considerable loneliness and dis- connection and the trauma of not being able to trust and build relation- ships, as a result of the experience of homelessness. Another barrier to participants? capacity to ?get connected? was the internalized stigma and shame many participants felt due to their homeless experience, their re- ceipt of income assistance, the depth of their poverty, and their residence in social housing. In both the focus groups and the qualitative interviews, service pro- viders and formerly homeless participants expressed frustration about the inappropriateness of the employment supports and the ageism that limits labour market participation. Recovery and employment programs were characterized as paternalistic and insensitive to individual needs and capacities. Ageism, coupled with episodic unemployment while homeless, constrained the employment options for many participants. Also, other ?homeless effects,? such as poor health and mental health and ongoing challenges in adapting to ?normal? schedules after years of chaotic living without permanent housing, made it extremely difficult for some participants to secure employment. These limitations were par- ticularly salient, given that qualitative data revealed that the majority of participants did not see themselves as ?retired? and were either actively looking for employment or were intending to do so in the near future. Despite the desire for employment, only one quarter of the partici- pants reported any income from employment in the previous six months and, of this group, the majority reported part-time or casual work. A larger percentage of the Calgary sample was employed full-time, reflect- ing the robust nature of the Alberta economy that may override the bar- riers to employment as a result of the stigmatization of this population. In both the qualitative and focus group analyses, participants hig- hlighted the struggles they experienced ?making ends meet.? Frequent descriptions of choosing between paying the rent or eating and the pro- hibitive luxury of new clothing or a fast food burger were a clear indica- tion that participants had moved along?but not off?the poverty conti- nuum. The high proportion of participants who relied on food banks and meal programs in both cities was a testimony to the challenges of secur- ing enough to eat. Although most participants were getting enough to eat through the use of food banks, meal programs and the groceries they were able to purchase, the nutritional value of the food was poor. The majority of participants in Toronto reported a yearly income for 2004 well below the current Low Income Cut-Off for a single individual in an urban centre. The average yearly incomes in Calgary were slightly higher, but still below the LICO for a city of this size. Not only were the levels of income available through assistance programs seen as inade- quate in meeting basic needs, program policies were characterized as ?welfare or poverty walls? that were difficult or impossible to transcend. Disincentives to work such as the clawback of earned income, the possi- bility of losing disability status, or the loss of health benefits were all de- scribed as formidable barriers to employment. Health and Well-being of Formerly Homeless Older Adults Overall the health and well-being of formerly homeless older adults were improving in comparison to the poorer scores reported for older homeless adults interviewed for the 2004 study. However, their health was lower than similar indicators reported in general population sur- veys. The double jeopardy of ?homeless? and ?accelerated aging? effects were limiting participants? abilities to move toward better health and well-being. Nevertheless, formerly homeless participants, once stably housed, reported greater access to health care. Results from the survey data indicated that the Calgary respondents scored considerably lower on the physical health scale, yet were much less likely to have visited a physician?s office in the previous six months than their Toronto counterparts. This may be attributable to a number of factors, including the fact that fewer had health cards and the shortage of physicians in Alberta as a result of rapid population growth there. Con- ditions that may have existed while they were homeless, but which re- mained undiagnosed, negatively influence measurements of health. In short, the identification and treatment of undiagnosed or latent condi- tions that occurred during homelessness affected health and well-being outcomes long after they had moved into housing. In both the qualitative interviews and the focus group analyses, participants stressed that re- covery from the experience of homelessness was ongoing, and that one year of homeless experience required several years of stable housed ex- perience to heal. The data collected in the questionnaire on the mental health status of participants indicated poorer mental health than evidenced in similar- ly aged adults in the general population. However, some improvement was indicated by the higher mean score than that of the homeless older adults interviewed for the 2004 study. Analysis of the qualitative data revealed that for many participants, poor mental health was an ongoing struggle, but that they felt ?less despairing? than they had when they were homeless. An important paradox raised by several participants was the flawed assumption that proximity and access to support would sig- nificantly improve their ability to seek help. A mental health crisis was described as ?not rational,? a process and state where the crisis itself prevented participants from seeking help. However, participants did express greater confidence that being housed facilitated earlier identifica- tion of imminent crises that would allow them to seek support to fores- tall a health crisis. Several participants described ?homeless effects? as lingering trau- ma adversely affecting their mental health, using terms similar to those associated with Post Traumatic Stress Disorder. The magnitude of the trauma experienced during the homeless period also emerged as a key issue in the focus groups with service providers who spoke of isolation and exile as maladaptive responses to ?homeless effects.? Service pro- viders and formerly homeless participants spoke of the critical need for supports and services to be sensitive and responsive to the residual ef- fects of the traumatic events experienced while homeless. Analyses from the service provider focus groups and from the sur- vey interviews also revealed that Alzheimer?s Disease and other forms of dementia were much less common than mood and schizophrenic disord- ers. This may be a sampling effect due to the relatively ?young? average age and better health of those older adults who were willing and able to participate in the interviews. The focus group participants did, however, indicate that it was difficult to separate the effects of overlapping health issues such as cognitive impairment and alcoholism, as alcohol misuse remains a problem for some formerly homeless older adults. Well-Being Barely more than half of the formerly homeless participants in Toronto and slightly less than half in Calgary reported satisfaction with their lives. Perhaps more significant is the number of remaining participants who rated life satisfaction as ?neutral? or ?dissatisfied.? A key theme emerging from the qualitative and focus group analyses is that housing ends ?houselessness,? but much more is needed to bring people into wellness, inclusion, and other positive dimensions associated with quali- ty of life. Key areas participants identified as limiting quality of life were factors like isolation, loneliness, discrimination, internalized stigma, and lack of opportunities for meaningful participation (within and outside of the labour market). Formerly homeless participants were at considerable risk of social isolation and continued to rely heavily on service providers for support. A significant difference was found between housing types and risk of social isolation in Toronto. Interestingly, those living in supportive hous- ing were at significantly greater risk of social isolation than those in sup- ported housing with the use of community supports. This finding is con- trary to much of the literature (Lum et al., 2005; Pynoos et al., 2004; Cannuscio, 2003), which suggests that the presence of on-site staff exerts a positive effect on social connection and interaction. However, these studies sample from the general population of older adults living in supportive and supported housing. Consequently, as Ridgeway et al. (1994) suggest, formerly homeless persons may have a greater need for privacy and self-determination and find staff intervention intrusive, which may undermine social connections. Another factor influencing this unexpected outcome is the selection bias that may result in formerly homeless older adults with greater needs and challenges being placed in supportive rather than more independent housing settings. Although qualitative analyses from the focus groups and qualitative interviews revealed that many formerly homeless persons were connect- ing and reconnecting with family and friends, a significant proportion remained disconnected from their housing and neighbourhood com- munities. Many factors limiting social support were cited, including dis- crimination (e.g., for having been homeless, for residing in social hous- ing, for receiving income assistance, for being labelled ?hard-to-house?), shame, distrust, lack of age-appropriate venues for social interaction, crime-ridden housing and neighbourhood environments, limited mobili- ty, poor mental health, and prohibitive transportation costs. Many partic- ipants expressed frustration with funding and programming that under- valued social capital, commenting that the focus was on the measurable outcomes of employment supports and that supports to social inclusion and quality of life were neglected. Although feelings of insecurity and threat were frequently mentioned by participants, overall, the formerly homeless reported fewer violations of personal safety than the homeless older adults interviewed in 2004. Use of Health Care Services Analyses of the survey on the use of acute care (hospital emergency de- partment visits) reported for the previous six months was similar for both the formerly homeless interviewed during this study and the homeless adults interviewed in 2004. However, analyses of the second- ary OHIP data on the use of health care services by formerly homeless participants in Toronto before and after housing, indicated that the ac- tual mean number of visits dropped significantly after being housed as did in-patient and day patient care. These changes suggest that housing may contribute to more stable health care for the homeless once they are housed. The changes also imp- ly reductions in the cost of care for this group as a result of being housed, since ambulatory care and inpatient care are expensive health services. Further, findings from the OHIP data analyses are consistent with the survey findings indicating that the overall health of the newly housed has improved compared to the health indicators for the earlier 2004 study of homeless older adults but below that for the general popula- tion. Service, Support and Housing Needs of Formerly Homeless lder Adults Just over half the participants reported finding out about and having received assistance in applying for their current housing from a social service worker. However, a significant minority located and secured their housing by themselves or with the assistance of informal supports such as family and friends. The analyses of the qualitative and focus group data found that some participants stressed that having profes- sional ?allies? or ?advocates? was essential to navigating the social ser- vice and housing systems, while others stressed the power and value of informal networks and resources. Many participants suggested that re- sources should be directly accessible to users and that those resources should be informed by peer knowledge. Peer-based resources that incor- porate the ?lived experience? of the homeless and formerly homeless persons were seen as more responsive and more accurate. The primary finding is that there is an acute shortage of affordable, age-appropriate housing and support options. This is an issue of supply, but also an issue of lack of variability in housing and support packages. Because of the very low vacancy rates in Calgary, respondents in sup- ported housing were probably forced to live in very poor circumstances, which they indicated in the survey and confirmed in the focus groups and individual interviews. Participants stressed that a variety of housing and support options was critical to achieving a ?good fit ? between indi- vidual needs and preferences and living arrangements. Participants indi- cated that the degree to which a ?good fit? was achieved influenced housing stability and health and well-being. No single model could ade- quately address the diversity of needs and preferences. Another critical aspect of achieving a ?good fit,? identified by both formerly homeless participants and service providers, was that the process must be client-directed. Self-determination and autonomy were highly valued by participants and were related to feelings of loss of trust and loss of control experienced while homeless. These ?homeless effects? were best addressed by models of service that were client-centred and stressed relationship building. Although the survey responses to questions about current housing and supports was, for the most part positive, a few areas emerged where needs were clearly not being met. When asked whether their housing was equipped to assist people with impaired mobility, the majority of participants in Toronto reported living in housing without accessibility accommodations. This finding has significant implications to formerly homeless older adults? abilities to age in place. The Calgary data painted a very different picture, with the majority reporting that their housing was equipped to deal with the challenges faced by those with mobility issues. This is a reflection of the much newer housing stock available in Calgary. Of those participants who indicated linked supports and services, the three most significant areas of unmet need identified were transpor- tation supports, special services for older people, and skills develop- ment. Transportation issues were identified as particularly relevant in Calgary, a city that is more geographically dispersed and with a far less developed public transportation infrastructure than Toronto. In the qua- litative interviews and focus groups, participants frequently reported that they could not afford transportation to health care facilities or meal programs and that many services were insensitive to the needs of older people (e.g., to slower mobility and diminished memory). For a great many participants identifying as ?too old to be young and too young to be old? (the demographic ?gap?), age-appropriate services were even more difficult to obtain. Another area of concern was shared living arrangements. Almost two-thirds of participants shared accommodation, but the vast majority expressed a clear preference for self-contained units. The conflicts arising in shared living arrangements became especially troubling in housing sites where tenants were clustered according to similar health and men- tal health challenges. Although some participants felt that residing with tenants who shared similar challenges promoted greater understanding and acceptance, most felt that diversity of age, gender, ability, health and mental health status and of tenure (i.e., mixed subsidized and market rentals) prevented ?ghettoization.? Participants spoke of cluster housing settings as creating dangerously vulnerable and disadvantaged housing communities. A number of participants reported feeling unsafe in their housing, and identified criminal activity and inadequate security along with a fear of fellow tenants as reasons for feeling unsafe. Building safety and per- sonal safety emerged as major themes in the qualitative data; partici- pants said they wanted to have unregulated guest visits, but feared that not screening guests was dangerous. Both service providers and partici- pants felt that more than any single policy or intervention, security and safety were best supported by ?community building,? which empha- sized participation, inclusion, and self-regulating tenant communities. Recommendations for Effective Models that Support Health, Housing, and Inclusion No single housing model was identified as most effective in supporting the health and well-being of formerly homeless adults. Although several models are identified below, the most significant theme was that a broad menu of housing, health and support options must be available to meet a diversity of needs and preferences of older homeless people. Client-Centred Models A primary theme emerging from the qualitative analyses was that the processes of finding and maintaining housing and supports should fol- low a client-centred model of delivery. Participants spoke of the necessi- ty for relationship-building and establishing trust with housing and support workers. Sound client-worker relationships were described as critical to early intervention to prevent returns to homelessness. Client determination of housing/support packages was viewed by participants as central to securing a ?good fit? without which housing instability might ensue. Continuity Models The theme of continuity of support was linked to relationship building. In some cases this meant continuity of support from shelter to housing, and in other cases, the focus was on continuity across moves to different types of housing. The former was contentious. Some participants de- scribed the link from the shelter to housing as effective, while others felt that it was undesirable, even traumatic, to maintain links to homeless services. However, almost all service providers and participants stressed that continuity across housing settings was critical to maintaining hous- ing stability and health. Several mechanisms for continuity were suggested, such as portable supports?for example, case management?that were de-linked from any single housing site or, alternatively, developing off-site partnerships with community-based agencies that would stay with a person and act as an adjunct to linked housing supports. Integrated models Integrated team models were championed as a means of providing lay- ers of support in a coordinated seamless delivery. In this model, coordi- nated interdisciplinary teams provide a combination of care across a number of housing settings, which may or may not have on-site staff. Service providers emphasized the challenges to staff in supporting a di- versity of needs in a single service setting, because of the scarcity of staff trained to support the mental health and personal care needs of aging formerly homeless tenants. Formerly homeless participants emphasized the challenges of negotiating fragmented, inaccessible service systems, where staff were either overwhelmed or inaccessible, a process exacer- bated by the lack of support from a professional advocate. ?Housing First? Models ?Housing first? models, though typically associated with independent ?low-demand? housing with client-determined, community-based sup- port, do not necessarily imply the absence of on-site staff. The distinction made by service providers and formerly homeless participants was that the housing was not contingent on the tenant using any particular sup- port or meeting any standard other than those demanded of all tenants (e.g., prohibition on criminal activities and on behaviours that interfere with reasonable enjoyment of other tenants). Both service providers and participants strongly endorsed a framework of universal rights and re- sponsibilities as an appropriate tool for accessing housing and mediating conflict. Harm Reduction Models Harm reduction was seen by service providers and participants as an integral component of a ?low-demand? ?housing-first? model, which would ensure that active users, often the most vulnerable of homeless persons, were not excluded from housing. However, service providers expressed concern that housing sites formally adopting a harm reduction model might be subject to unfair scrutiny and stigma, despite substantial evidence-based research attesting to the effectiveness of harm reduction approaches (Hunt, 2004; Marlatt and Witkiewitz, 2002; Riley and O?Hare, 2001; MacPherson, 1999). As an alternative, service providers felt that a rights and responsi- bilities framework subjects tenants to the same prohibitions on substance use enforced in the general population without the problems associated with formal sanction of harm reduction. However, such an ?informal? model of harm reduction may mean that the supports associated with formal harm reduction are not available, such as service and supplies to support safer consumption. Community Development Models that Stress Participation and Engagement Formerly homeless participants spoke of the need to build healthy hous- ing and neighbourhood communities. Community building was accom- plished by programs that stressed participation in decision making. For example, participants and service providers spoke of tenant councils that addressed everything from social recreational programs to providing the first intervention in the event of risk of eviction. Self-regulating housing communities were valued for fostering social connections; enhancing feelings of security, safety and autonomy; and providing a mechanism for skill-building transferable to other settings. An extension of skill building was developing micro-enterprises within the housing commu- nity to support transitions to paid work and combat the ageism and oth- er forms of discrimination confronting employment seeking formerly homeless older adults. An integral component of community building models was that they engage and incorporate peer knowledge. Participants highly valued ?lived experience? and spoke of ?word on the street? (and in the drop- ins) as a vital and responsive resource. A central theme emerging in both the qualitative and focus group analyses was that formerly homeless older adults had a tremendous amount of knowledge and resources that could be integrated into programming, materials, and policies affecting the homeless community. Models that Emphasize Diversity and are Integrated into the ?Mainstream? Although some participants expressed a preference for ?clustered? set- tings (that is, living in facilities with people with similar mental health challenges), most endorsed diversity as desirable across age, rental status (subsidized and market rents) and health status. Clustering was per- ceived as dangerous and described as ?ghettoization? that induced con- flict and vulnerability to victimization. Service providers were less clear on the subject of diversity versus clustering. Many providers felt that diversity was an valuable principle, but difficult to implement?that is, selective placement may not always be possible and staffing to accom- modate a diversity of needs was challenging. A variant of the theme of diversity was that of ?mainstreaming.? Many participants described the stigma and shame associated with using food banks and meal programs and residing in social housing clearly demarcated from the rest of the housing in a neighbourhood. Integrating service, supports, and housing into the mainstream was identified by many participants as a way to reduce the stigma. Participants suggested a number of examples, such as some sort of invisible proxy that could be used to buy food and meals in mainstream venues or community kitch- ens open to all members of the public with nominal or subsidized fees. Models that Support Transitions Formerly homeless participants were adamant that models of services, supports, and housing must support transition and be flexible to shifts in need and preferences. Participants wished to move to different housing sites, toward better health, well-being and inclusion, and toward greater economic security. Many participants expressed frustration with models that assumed the status quo was sufficient and that ?maintenance was progress.? However, participants were sensitive to the risk that models emphasizing transition may marginalize or adversely impact those per- sons who cannot or will not make those transitions, again suggesting that client-centred, flexible models would be able to accommodate both options. Key Challenges to Effective Delivery of these Models The focus group and qualitative interview analyses revealed limitations to the delivery of the above models to formerly homeless older adults: notably ?homeless effects?; accelerated ?aging effects?; ageism, especial- ly that confronting those 50 to 65 years of age; classism; ?poverty or wel- fare walls?; and a lack of affordable age-appropriate housing and sup- ports. ?Homeless effects? and ?accelerated aging effects? are clearly in- fluencing the ability of formerly homeless older adults to recover and to improve health and well-being. Consequently, supports must be sensi- tive to these effects. For example, health interventions should stress the recovery of nutritional deficits incurred over the homeless period or ac- commodate, without pathologizing the lingering effects of trauma expe- rienced while homeless. The varied and pervasive forms of discrimination experienced by the older adults limited their ability to secure employment, and housing, and to realize meaningful social integration. Classism and all its variants, identified in the analyses by such phrases as ?hard-to-house,? ?welfare bum ? and ?living in the projects? (social housing), are critical barriers that housing and support models must overcome. One way that housing and support models can address these stigmatizing labels is to avoid ?clustering? and ?naming? disadvantage whether through ensuring di- versity or ensuring that any disadvantage associated with a program is as invisible as possible. Ageism, as is evident in the general population, seriously eroded the ability of formerly homeless older adults to secure employment. Age- ism in employment-seeking was further exacerbated for this group by the ?homeless effect,? which created significant breaks in their employ- ment history or made skill sets obsolete. These limitations were particu- larly significant for those participants who saw themselves as members of the ?demographic gap? between 50 to 65 years of age who were ac- tively seeking employment and not ready to retire. Participants reported feeling caught between the conflicting assumptions that they were too old to find employment in a competitive and ageist labour market, yet were receiving income assistance?related employment support programs premised on the expectation of future employment and the cessation of income assistance. Skill development, training, and employment support programs for formerly homeless older adults should be based on realistic assumptions of labour market participation and options to exercise skills in volunteer settings. The issue of the invisibility of the demographic ?gap ? extended to other areas of programming and was seen by participants as a serious limitation to appropriate service delivery. Service models should adapt and accommodate what participants refer to as a group that is ?too old to be young and too young to be old.? ?Poverty or welfare walls? were a serious impediment to formerly homeless older adults achieving greater economic security. Participants, despite receiving income assistance and housing subsidies, were still liv- ing considerably below established Low-income Cut-offs (NCW, 2006). The reliance on food banks and meal programs reported by participants indicates the depth of poverty that many formerly homeless older adults experienced. As reported in the discussion of the socio-economic status of formerly homeless older adults, income assistance was not only in- adequate, given the cost of living, but also presented formidable barriers to getting out of poverty (e.g., asset ceilings) and into employment (e.g., loss of health benefits). For formerly homeless older adults subject to discrimination and persistent ?homeless effects? and ?accelerating aging effects,? income support programs designed to be temporary and resi- dual were inappropriate to their needs and challenges. A final and significant limitation is housing and support models that assume a static level of support with no effective means for transi- tion to other housing settings. Formerly homeless participants spoke of the desire to move to other housing settings; many were looking for set- tings with more independence and less support while some required higher levels of support and more accessible accommodations. Some formerly homeless participants and service providers spoke of the need to accommodate higher levels of support in the earlier stages of housing, which may no longer be necessary as greater health and housing stability is achieved. Although the most formidable barrier to housing transition is the scarcity of affordable, age-appropriate housing and support options, any available transfers were reported to be problematic and inadequately supported. For example, both service providers and participants noted the vulnerability introduced in moving to new locations and establishing new supports. Portable or community-based supports were mentioned as mediating the risks to the social connections and housing stability as- sociated with relocation. Other suggestions made by service providers were that transitions should be ?trialed? and barriers removed so people could return to their original housing situation. For formerly homeless older adults, the risks to stability of health, well-being, and of housing associated with adapting to a new setting must be mediated by models that offer ongoing links to supports established prior to the move. Conclusion The most significant implications of these findings for practice, program development, and policy-making are fourfold. First, the findings em- phasize that it is critical that health, support, and housing programs are sensitive to ?homeless effects? and accelerated ?aging effects.? Recogniz- ing and supporting recovery from the persistent trauma induced by these effects is essential to preventing formerly homeless older adults from cycling back to homelessness. Rapid intervention is critical and must support people as they make transitions and during the first years of housing. Second, developing and evaluating age-appropriate affordable housing and supports are of primary importance. However, the findings highlight that policy, programming, and research must be premised on social inclusion so that issues such as community integration, belonging, participation, overcoming discrimination and stigma, and other meas- ures of quality of life can be addressed. Third, assumptions around income support and employment sup- port for this group need to be revisited. There is a significant disconnect between expectations embedded in these programs and the significant barriers experienced by formerly homeless older adults. Finally, the findings suggest that homelessness and former home- lessness must be situated as points on the poverty continuum so that pol- icy and programming do not address them as discrete or disconnect them from other socio-economically marginalized groups and from the general population of older adults. Lynn MacDonald References is a professor in the Faculty of Social Work and Director of the Institute for Life Course and Aging at the University of Toronto. Peter D o- nahue is an Assistant Professor at the Faculty of Social Work, University of Calgary. Julia Janes is a researcher with the Institute for Life Course and Ag- ing at the University of Toronto. Laura Cleghorn of Cardinal Consultants has conducted research for ten years in the field of social and health services. Cannuscio, C., Black, J., & Kawachi, I. (2003). Social capital and successful aging: The role of senior housing. Annals of Internal Medicine, 139(5, Part 2): 395-39. Hunt, N. (2004). A review of the evidence-base for harm reduction approaches to drug use. Retrieved January 28, 2005 from www.forward-thinking-on- drugs.org/review2-print.html Lum, J., Ruff, S. & Williams, P. (2005). When Home is Community: Community Sup- port Services and the Well-Being of Seniors in Supportive and Social Housing. To- ronto: United Way of Greater Toronto. MacPherson, D. (2001). A Framework for Action: A Four Pillar Approach to Drug Problems in Vancouver. Vancouver: City of Vancouver. Marlatt, G. & Witkiewitz, K. (2002). Harm Reduction approaches to alcohol use: health promotion, prevention and treatment. Addictive Behaviours, 27, 867- 886. National Council of Welfare (NCW) (2006). National Council of Welfare: Welfare Incomes 200 5 . Retrieved August 28, 2006, from http://www.ncwcnbes.net/htmdocument/reportWelfareIncomes2005/WI200 5ENG.pdf Pynoos, J., Liebig, P., Alley, D., & Nishita, C. M. (2004). Homes of choice: To- wards more effective linkages between housing and services. Journal of Housing for the Elderly, 18(3/4): 5-49. Ridgeway, P., Simpson, Wittman, F. & Wheeler, G. (1994). Home Making and Community Building: Notes on Empowerment and Place. Journal of Mental Health Administration, 21(4), 407-418. Riley, D. & O?Hare, P. (1999). Harm Reduction: History, Definition and Practice. In J. Inciardi & L. Harrison. eds., Harm Reduction: National & International Perspectives. Thousand Oaks, California: Sage. Chapter 2.6 Traumatic Brain Injury in the Homeless Pop ulation: A Toronto Study STEPHEN W. HWANG, ANGELA C OLANT ONI O , SHIRLEY CHIU, GE ORGE T OL OMIC ZENK O , ALE X KISS, LAURA C OWAN , D ONALD A. REDELMEIER, AND WENDY LEVINS O N Traumatic brain injury is caused by ?a blow or jolt to the head or a pene- trating head injury that disrupts the normal function of the brain? and most commonly results from falls, motor vehicle traffic crashes and as- saults (National Center for Injury Prevention and Control, 2008). Trau- matic brain injury is a leading cause of permanent disability in North America. Traumatic brain injury may be common in the homeless popu- lation (Waldmann, 2004). Exposure to physical abuse during childhood, which could result in traumatic brain injury, is a known risk factor for homelessness as an adult (Herman et al., 1997). Substance abuse in- creases the risk of homelessness (Susser et al., 1993) and the risk of trau- matic brain injury (Corrigan, 1995). Homeless people experience high rates of injury of all types and are frequently victims of assault (Kushel et al., 2003; Zakrison et al., 2004). Finally, traumatic brain injury could be a factor contributing to the 3 to 8 percent prevalence of cognitive dysfunc- tion among homeless adults (Kass & Silver, 1990; Spence et al., 2004). Providing health care for homeless patients can be challenging for various reasons, including difficult behavioural patterns. These behaviours may be related in part to the unrecognized consequences of traumatic brain injury and may include cognitive impairment, attention deficits, disinhibi- tion, impulsivity and emotional lability.' Two previous studies have reported the prevalence of traumatic brain injury among homeless people in London, England, and Milwaukee, Wis- consin. These studies were limited by small sample sizes, recruitment at a single shelter, and a lack of data from women (Bremner et al., 1996; Solli- day-McRoy et al., 2004). We conducted this study to determine the lifetime prevalence of traumatic brain injury in a representative sample of homeless men and women across an entire city, and to identify the temporal relation between traumatic brain injury and the onset of homelessness. We also sought to characterize the association between a history of traumatic brain injury and current health problems in this population. Our primary hy- pothesis was that a history of traumatic brain injury would be associated with poor current health. Appropriate support services may be able to minimize the adverse impact of these behaviours. Survey and analysis Study design and population We used a cross-sectional survey design. We recruited a representative sample of homeless people in Toronto, Ontario, where about 5,000 people are homeless each night and about 29,000 people use shelters each year (Brown, 2006; City of Toronto, 2003). We defined homelessness as living within the last seven days at a shelter, public place, vehicle, abandoned building or someone else?s home, and not having a home of one?s own. Based on a pilot study, we determined that about 90 percent of homeless people in Toronto slept at shelters, and that 10 percent did not use shelters but did use meal programs (Hwang et al., 2005). We therefore recruited 90 percent of our study participants at shelters and 10 percent at meal pro- grams. We contacted every homeless shelter in Toronto and obtained permis- sion to enrol participants at 50 (89 percent) out of 56 shelters (20 shelters for men, 12 for women, 6 for men and women, and 12 for youths aged 16? 25 years). The number of beds at each shelter ranged between 20 and 406. Re- cruitment at meal programs took place at 18 sites selected at random from 62 meal programs in Toronto that served homeless people. Because the goal of recruiting at meal programs was to enrol homeless people who did not use shelters, we excluded people at meal programs who had used a shelter within the last seven days. We recruited participants over 12 consecutive months in 2004? 2005. We stratified enrolment to achieve a 2:1 ratio of men to women. The num- ber of participants recruited at each site was proportionate to the number of homeless people served monthly. We selected participants at random from bed lists or meal lines using a random number generator and assessed their eligibility. We excluded people who did not meet our definition of home- lessness, who were unable to communicate in English, and who were un- able to give informed consent. We also excluded homeless shelter users encountered at meal programs and those who did not have a valid Ontario health insurance number, which was required to track health care use after the recruitment interview. Previous studies have shown that homeless parents with dependent children differ substantially from homeless people without children. Homeless parents have lower rates of mental illness and substance abuse and are more likely than those without children to have become homeless for purely economic reasons (Robertson & Winkleby, 1996; Shinn et al., 1998). Because of these differences, this report does not include homeless parents with dependent children who were enrolled in the study. Survey instrument Each participant provided written informed consent and received $15 for completing the survey. This study was approved by the research ethics board at St. Michael?s Hospital. Research team members administered the survey to each participant by a face-to-face interview conducted immediately after recruitment at shelters and meal programs. We obtained information on demographic characteris- tics and health conditions. We collected data on ethnic background because previous studies have reported racial disparities in rates of traumatic brain injury. Participants self-identified their ethnic background from categories adapted from the Statistics Canada Ethnic Diversity Survey (Statistics Canada, 2002). The most commonly selected categories were white, black, and First Nations. All other categories were classified as ?other.? Mental health problems, alcohol problems and drug problems in the last 30 days were assessed using the Addiction Severity Index (McGahan et al., 1986; McLellan et al., 1992). The Addiction Severity Index has been validated with homeless people and used in numerous studies, including a nationwide survey of homeless people in the United States (Burt et al., 2001; Drake et al., 1995; Joyner et al., 1996; Zanis et al., 1994). We classified participants as having mental health problems if their mental health score on the Addiction Severity Index was ? 0.25. They were classified as having alcohol problems if their alcohol score was ? 0.17 and were classified as having drug problems if their drug score was ? 0.10 (Burt et al., 1999). We used the SF-12 health survey, a health status instrument that has been validated in homeless populations (Larson, 2002), to generate scores for the physical and mental component subscales (Ware et al., 1995). These scores range continuously from 0 to 100 (best), standardized to a mean of 50 and standard deviation of 10 in the general population in the United States (Ware et al, 1995). Problems were dichotomized as present or absent by use of cut-off scores established for homeless populations (Burt et al., 1999). We determined a history of traumatic brain injury using questions from a study of prison inmates (Slaughter et al., 2003). Lifetime prevalence of traumatic brain injury was determined using the question, ?Have you ever had an injury to the head which knocked you out or at least left you dazed, confused or disoriented?? Participants were asked how many such injuries they had over their lifetime. For the first injury and up to two subsequent injuries, we obtained the date or age at injury, whether the injury resulted in unconsciousness, and the duration of unconsciousness. We used the age at which the participant first experienced homelessness to determine the temporal relation between the first traumatic brain injury and the onset of homelessness. A mild traumatic brain injury was defined as a head injury that left the person dazed, confused, or disoriented, but resulted in no unconsciousness or un- consciousness for less than 30 minutes. A moderate or severe traumatic brain injury was defined as a head injury that resulted in unconsciousness for more than 30 minutes. These definitions are consistent with standard- ized consensus criteria (Kay et al., 1993) Statistical analyses . We compared the characteristics of people with and without a history of traumatic brain injury using chi-square and T tests. We developed re- gression models to determine if a history of traumatic brain injury was associated with health conditions and health status indicators, after ad- justment for sex, age, ethnic background, place of birth, education and lifetime years of homelessness. We used generalized estimating equa- tions to account for possible clustering of the sample within shelters or meal programs. History of traumatic brain injury was entered into models as a cate- gorical variable representing the severity of the worst traumatic brain injury ever experienced (none, mild or unknown, or moderate or severe). In our secondary analyses, both severity of the worst traumatic brain in- jury and the lifetime number of traumatic brain injuries were entered into models. We assessed independent variables for multicolinearity be- fore the analyses, and no problems were detected. Analyses were con- ducted with unweighted data. Of 1,679 people screened at homeless shelters and meal programs, we included 904 people in our study. In total, 489 (29 percent) were ineligible for inclusion: 222 (13 percent) did not meet our definition of homelessness, 61 (4 percent) were unable to communicate in English, 54 (3 percent) were homeless shelter users encountered at meal programs, and 51 (3 percent) were unable to give informed consent. Because this study was part of a lar- ger study of the use of health care services by homeless people, we ex- cluded 101 people (6 percent) because they did not have an Ontario health insurance number. Most of these 101 people were refugees, refugee claim- ants or had recently migrated to Ontario. Of 1,190 eligible people, 283 declined to participate. We enrolled 907 (76 percent of eligible people) in the study. We obtained information about traumatic brain injury for 904 participants. The characteristics of the study participants are shown in Table 1. Results The lifetime prevalence of traumatic brain injury was 53 percent. The prevalence was significantly higher among men (58 percent) than among women (42 percent, p < 0.001). Those with a history of trau matic brain injury were more likely to be male, white and born in Canada; to have become homeless for the first time at a younger age; and to have experi- enced more years of homelessness over their lifetime. Compared to those without a history of traumatic brain injury, participants with a history of traumatic brain injury had a significantly higher lifetime prevalence of seizures (8 percent v. 22 percent, p <0.001); higher prevalence of mental health problems (33 percent v. 43 percent, p = 0.001), alcohol problems (28 percent v. 42 percent, p < 0.001) and dr ug problems (40 percent v. 57 percent, p < 0.001). They also had poorer mental health (mean score 43.8 v. 39.0, p < 0.001) and phys ical health (mean score 48.1 v. 43.9, p < 0.001) as measured by the SF-12 health survey (Table 1). The mean age at first traumatic brain injury was 17.8 years. Although 40 percent of participants with traumatic brain injuries reported only 1 such injury, 21 percent reported 2 injuries, 12 percent reported 3 injuries, 7 per- cent reported 4 injuries, and 20 percent reported 5 or more injuries. The severity of the worst traumatic brain injury was mild for 66 percent of par- ticipants, moderate or severe for 23 percent and unknown for 11 percent. In all analyses involving traumatic brain injury severity, we grouped injuries of unknown severity with mild injuries. Analyses in which injuries of un- known severity were considered to be a separate category gave essentially identical results. The temporal relation between the first traumatic brain injury and the first episode of homelessness is shown in Figure 1. For 70 percent of par- ticipants, the first traumatic brain injury occurred before the onset of home- lessness. The injury occurred in the same year as the onset of homelessness for 7 percent of participants, and after the onset of homelessness for 22 per- cent. We could not determine the relation between the first traumatic brain injury and the first episode of homelessness for 2 percent of participants. When we considered the influence of sex, age, ethnic background, place of birth, education and lifetime years of homelessness, a history of traumatic brain injury was significantly associated with seizures, mental health and drug problems, and poorer physical and mental health status (Table 2). In additional models that included both the severity of the worst traumatic brain injury and the total lifetime number of traumatic brain inju- ries as covariables, a higher number of traumatic brain injuries was associ- ated with significantly increased odds of seizures and mental health, alco- hol and drug problems. Figure 1: Homeless participants ( n = 461 ) who experienced a traumatic brain injury before or after becoming homeless. Interpretation We found a high prevalence of traumatic brain injury in a representative sample of homeless people. A history of traumatic brain injury was more common among homeless men (58 percent) than among homeless women (42 percent). These rates are five or more times greater than the 8.5 percent lifetime prevalence rate of traumatic brain injury in the gen- eral population in the United States (Silver et al., 2001) and are within the range reported in studies of traumatic brain injury among prison inmates (Morrell et al., 1998; Schofield et al., 2006; Slaughter et al., 2003) Only two previous studies have reported the prevalence of traumatic brain injury among homeless people. In a study of 80 consecutive entrants to a men?s shelter in London, England, 46 percent of entrants had a lifetime history of head injury severe enough to cause unconsciousness (Bremner et al., 1996). A study of 90 homeless men at a shelter in Milwaukee, Wiscon- sin, found that 80 percent of participants had possible cognitive impairment and 48 percent had a history of traumatic brain injury involving loss of con- sciousness (Solliday-McRoy et al., 2004). In both studies, the sample size was small, and participants were recruited at a single shelter rather than at a broad range of shelters across an entire city. In addition, homeless women and homeless people who did not use shelters were excluded. . Data from the United States have demonstrated higher rates of trau- matic brain injury among African-Americans (National Center for Injury Prevention and Control, 2008). In contrast, our study found a significantly lower prevalence of traumatic brain injury among homeless people who were black (30 percent) compared with those who were white (59 percent). This difference is possibly explained by the fact that traumatic brain injury was much less common among immigrants than among people born in Canada. In our study, 69 percent of participants who were black were im- migrants to Canada. Among homeless people, the first experience of traumatic brain injury often occurred at a young age and usually occurred before the person?s first episode of homelessness. This finding suggests that, in some cases, trau- matic brain injury may be a causal factor that contributes to the onset of homelessness, possibly though cognitive or behavioural consequences of traumatic brain injury. Future research could explore this hypothesis. A history of traumatic brain injury was strongly associated with many adverse health outcomes among homeless people, including seizures, men- tal health problems, drug problems, and poorer physical and mental health status. A history of moderate or severe traumatic brain injury had particu- larly strong associations with both the presence of mental health problems within the past 30 days (OR 2.5, 95 percent; CI 1.5? 4.1) and poorer mental health status (? 8.3 points on the SF-12 mental component sub-scale1). Our cross-sectional study was unable to ascertain the causal pathways respon- sible for these associations. Although the cognitive effects of traumatic brain injury may increase the risk of subsequent mental health and drug problems, it is equally plausible that pre-existing mental health, alcohol and drug problems increase the risk of experiencing traumatic brain injury (Parry-Jones et al., 2006) Strengths and limitations . Likewise, homelessness could be both a contrib- uting cause and a consequence of traumatic brain injury. Clarification of these issues would require data from a prospective longitudinal study of people with traumatic brain injury. 1 10 points equals 1 standard deviation in the general population. Our study has a number of important strengths. We enrolled a large repre- sentative sample of both homeless men and women in a major North American city, including both those who used and those who did not use shelters. We used rigorous methods to select participants randomly at each site. We achieved a high response rate, and successfully recruited 76 per- cent of eligible people. History of traumatic brain injury was assessed using a series of questions from a previously validated survey of prison inmates (Slaughter et al., 2003) Certain limitations of this study should be noted. We did not enrol a control group of non-homeless people. Our findings may not reflect rates of traumatic brain injury among homeless parents with dependent children or homeless persons who do not use shelters or meal programs. The require- ment that study participants have an Ontario health insurance number re- sulted primarily in the exclusion of refugees and refugee claimants, whose history of traumatic brain injury may be different from that of other home- less people. We did not collect information about the mechanism or cir- cumstances of traumatic brain injury. Prevalence and severity of traumatic brain injury as well as age at the time of traumatic brain injury were self- reported by participants and are subject to recall errors. Confirmation of these self-reports through the review of health records was beyond the scope of our study. . Recently, the Traumatic Brain Injury Questionnaire has been described as a promising interview-based instrument to assess the history of trau- matic brain injury in incarcerated adults (Diamond et al., 2007). Future studies including homeless people should consider using this instrument. Finally, participants did not undergo formal testing for neuropsychological dysfunction that may have resulted from brain injuries. Conclusion Our study?s findings underscore the need for clinicians to routinely ask patients who are homeless about a history of traumatic brain injury. Given the apparent dose? response relation between injury severity and current health, clinicians should assess injury severity based on information such as self-reported duration of unconsciousness, admission to hospital after the injury, collateral history and medical records. For people with a history of traumatic brain injury, brief neuropsychological screening can provide valuable information on cognitive function. People with moderate or severe cognitive impairment may be eligible for disability benefits. Re- ferral to rehabilitation and other appropriate community services should be considered, as recent studies have shown that rehabilitation interven- tions improve community integration and other outcomes among people with traumatic brain injury (Gordon et al., 2006). Moreover, appropriate living environments are fundamental to community integration and are particularly important for people with more severe injuries (Kelly & Winkler, 2007). Treatment of concurrent alcohol or substance abuse should also be considered. Future research should expand these findings by using medical re- cords to confirm self-reported traumatic brain injury among homeless peo- ple and by correlating a history of traumatic brain injury with objectively assessed cognitive function. Cohort studies would be helpful to clarify the causal pathways that account for the high prevalence of traumatic brain injury among homeless people. Finally, research should examine the possi- ble benefits of appropriate supportive living environments for homeless people with moderate cognitive dysfunction due to traumatic brain injury. Stephen W. Hwang and Shirley Ch iu are with the Centre for Research on Inner City Health, Keenan Research Centre in the Li Ka Shing Knowledge Insti- tute of St. Michael?s Hospital, Toronto. Donald A. Redelmeier is with the In- stitute for Clinical Evaluative Sciences at Sunnybrook Health Sciences Centre in Toronto. Stephen W. Hwang and Donald A. Redelmeier also work in the Di- vision of General Internal Medicine, Department of Medicine at the University of Toronto. George Tolomiczenko is with the Department of Psychiatry, Uni- versity of Toronto and the Crohn?s and Colitis Foundation of Canada. Wendy Levinson is in the Department of Medicine, University of Toronto. A ngela Colantonio is with the Toronto Rehabilitation Institute, Department of Occu- pational Science and Occupational Therapy. Alex Kiss is with the Department of Research Design and Biostatistics. Laura Cowan works for the Street Health Community Nursing Foundation. References Bremner, A. J., Duke, P. J., Nelson, H. E., & Pantelis, C. (1996). Cognitive function and duration of rooflessness in entrants to a hostel for homeless men. Brit- ish Journal of Psychiatry, 169(4), 434-439. Brown, P. (2006). Street needs assessment: Results and findings. Toronto: Shelter, Support and Housing Administration, The City of Toronto. Retrieved Au- gust 27, 2008, from www.toronto.ca/housing/pdf/streetneedsassessment .pdf. Burt, M. R., Aron, L. Y., & Douglas, T. (1999). Homelessness: Programs and the peo- ple they serve. Findings of the National Survey of Homeless Assistance Providers and Clients. Washington: United States Interagency Council on the Home- lessness. Burt, M. R., Aron, L. Y., & Lee, E. (2001). Helping America?s homeless. Washington: The Urban Institute Press. City of Toronto (2003). The Toronto report card on housing and homelessness. To- ronto: Author. Retrieved August 27, 2008, from www.toronto.ca/homelessness /pdf/reportcard2003.pdf. Corrigan, J. D. (1995). Substance abuse as a mediating factor in outcome from traumatic brain injury. Archives of Physical Medicine and Rehabilitation, 76(4), 302-309. Diamond, P. M., Harzke, A. J., Magaletta, P. R., Cummins, A. G., & Frankowski, R. (2007). Screening for traumatic brain injury in an offender sample: A first look at the reliability and validity of the Traumatic Brain Injury Question- naire. Journal of Head Trauma Rehabilitation, 22(6), 330-338. Drake, R. E., McHugo, G. J., & Biesanz, J. C. (1995). The test-retest reliability of standardized instruments among homeless persons with substance use dis- orders. Journal of Studies on Alcohol, 56(2), 161-167. Gordon, W. A., Zafonte, R., Cicerone, K., Cantor, J., Brown, M., Lombard, L., Goldsmith, R., & Chandna, T. (2006). Traumatic brain injury rehabilitation: State of the science. American Journal of Physical Medicine and Rehabilitation, 85(4), 343-382. Herman, D. B., Susser, E. S., Struening, E. L., & Link, L. B. (1997). Adverse child- hood experiences: Are they risk factors for adult homelessness?. American Journal of Public Health, 87(2), 249-255. Hwang, S. W., Chiu, S., Kiss, A., Katic, M., Tolomiczenko, G., Levinson, W., & Redelmeier, D. (2005). Use of meal programs and shelters by homeless peo- ple in Toronto. Journal of Urban Health, 82(Suppl. 2), 46. Joyner, L. M., Wright, J. D., & Devine, J. A. (1996). Reliability and validity of the Addiction Severity Index among homeless substance misusers. Substance Use and Misuse, 31(6), 729-751. Kass, F., & Silver, J. M. (1990). Neuropsychiatry and the homeless. Journal of Neu- ropsychiatry and Clinical Neuroscience, 2 (1), 15-19. Kay, T., Harrington, D. E., Adams, R., Anderson, T., Berrol, S., Cicerone, K., Dahlberg, C., Gerber, D., Goka, R., Harley, P., Hilt, J., Horn, L., Lehmkuhl, D., & Malec, J. (1993). Definition of mild traumatic brain injury. Journal of Head Trauma Rehabilitation, 8(3), 86-87. Kelly, G., & Winkler, D. (2007). Long-term accommodation and support for peo- ple with higher levels of challenging behaviour. Brain Impairment, 8(3), 262- 275. Kushel, M. B., Evans, J. L., Perry, S., Robertson, M. J., & Moss, A. R. (2003). No door to lock: Victimization among homeless and marginally housed per- sons. Archives of Internal Medicine, 163(20), 2492-2499. Larson, C. O. (2002). Use of the SF-12 instrument for measuring the health of homeless persons. Health Services Research, 37(3), 733-749. McGahan, P. L., Griffith, J. A., Parente, R., & McLellan, A. T. (1986). Addiction Severity Index: Composite scores manual. Philadelphia: The University of Pennsylvania, Veterans Administration Centre for Studies of Addiction. Re- trieved August 27, 2008, from www.tresearch.org/resources/compscores/ CompositeManual.pdf. McLellan, A. T., Kushner, H., Metzger, D., Peters, R., Smith, I., Grissom, G., Pet- tinati, H., & Argeriou, M. (1992). The fifth edition of the Addiction Severity Index. Journal of Substance Abuse Treatment, 9(3), 199-213. Morrell, R. F., Merbitz, C. T., Jain, S., & Jain, S. (1998). Traumatic brain injury in prisoners. Journal of Offender Rehabilitation, 27(3-4), 1-8. National Center for Injury Prevention and Control. (2008). What is traumatic brain injury?. Atlanta: The Centre for Disease Control. Retrieved August 27, 2008, from www.cdc.gov/ncipc/tbi/TBI.htm. Parry-Jones, B. L., Vaugh, F. L., & Cox, V. M. (2006). Traumatic brain injury and substance misuse: A systematic review of prevalence and outcomes re- search (1994-2004). Neuropsychological Rehabilitation, 16(5), 537-560. Robertson, M. J., & Winkleby, M. A. (1996). Mental health problems of homeless women and differences across subgroups. Annual Review of Public Health, 17, 311-336. Schofield, P. W., Butler, T. B., Hollis, S. J., Smith, N. E., Lee, S. J., & Kelso, W. M. (2006). Traumatic brain injury among Australian prisoners: Rates, recur- rence and sequelae. Brain Injury, 20(5), 499-506. Shinn, M., Weitzman, B. C., Stojanovic, D., Knickman, J. R., Jim?nez, L., Duchon, L., James, S., & Krantz, D. H. (1998). Predictors of homelessness among families in New York City: From shelter request to housing stability. Ameri- can Journal of Public Health, 88(11), 1651-1657. Silver, J. M., Kramer, R., Greenwald, S., & Weissman, M. A. (2001). The associa- tion between head injuries and psychiatric disorders: Findings from the New Haven NIMH Epidemiologic Catchment Area Study. Brain Injury, 15(11), 935-945. Slaughter, B., Fann, J. R., & Ehde, D. (2003). Traumatic brain injury in a county jail population: Prevalence, neuropsychological functioning and psychiatric disorders. Brain Injury, 17(9), 731-741. Solliday-McRoy, C., Campbell, T. C., Melchert, T. P., Young, T. J., & Cisler, R. A. (2004). Neuropsychological functioning of homeless men. Journal of Nervous and Mental Disease, 192(7), 471-478. Spence, S., Stevens, R., & Parks R. (2004). Cognitive dysfunction in homeless adults: A systematic review. Journal of the Royal Society of Medicine, 97(8), 375-379. Statistics Canada, Housing, Family and Social Statistics Division (2002). Ethnic diversity survey questionnaire. Ottawa: Author. Retrieved August 27, 2008, from www.statcan.ca/english/sdds/instrument/4508_Q1_V1_E.pdf . Susser, E., Moore, R., & Link, B. (1993). Risk factors for homelessness. Epidemi- ologic Reviews, 15(2), 546-556. Waldmann, C. A. (2004). Traumatic brain injury (TBI). In J. J. O?Connell (Ed.), The health of homeless persons: A manual of communicable disease and common prob- lems in shelters and on the streets (p p. 237-241). Boston: Boston Health Care for the Homeless Program and the National Health Care for the Homeless Council. Ware, J. E., Kosinski, M., & Keller, S. D. (1995). SF-12: How to score the SF -12 phys i- cal and mental health summary scales (2nd ed.). Boston: The Health Institute, New England Medical Center. Zakrison, T. L., Hamel, P. A., & Hwang, S. W. (2004). Homeless people?s trust and interactions with police and paramedics. Journal of Urban Health, 81(4), 596-605. Zanis, D. A., McLellan, A. T., Canaan, R. A., & Randall, M. (1994). Reliability and validity of the Addiction Severity Index with a homeless sample. Journal of Substance Abuse Treatment, 11(6), 541-548. Chapter 2.7 Primary Health Care for Homeless Persons: Evaluating the Options Using a Policy Analysis Ap proach S.E.D. SH ORTT , STEPHEN W. HWANG, HEATHER STUART, MELANIE BED ORE , NADIA Z URBA, MARGARET DARLING THE 1999 Canadian National Homelessness Initiative (now the Home- less Partnering Strategy [ Human Resources and Social Development Canada (HRSDC), 2008] ) defined as homeless ?any person, family or household that has no fixed address or security of tenure.? How many people fall within this definition is unknown, particularly since ?rough sleepers? (persons on the streets) and ?couch surfers? (individuals chronically staying with others) are almost impossible to enumerate. However, the 2001 Census found that 14,145 persons were using shelters at any given time in Canada; by the 2006 Census, that number had risen to 19,630 (Statistics Canada, 2002, 2008). Males, aged 35 to 64 years, were the most common subgroup within this population, followed by males, aged 15 to 34 years (Statistics Cana- da, 2002). Data from Toronto and Ottawa revealed that families consti- tute a significant portion of shelter users, occupying 42 percent and 35 percent of shelter beds in each city, respectively (Hwang, 2001). Abori- ginal people are overrepresented in the homeless population; in Toronto, they accounted for 2 percent of the total population in 1999 but 25 per- cent of the homeless population (Begin, Casavant, Chenier, & Dupuis, 1999). It is difficult to describe with precision the health problems of homeless persons, in part because of the heterogeneity of this population across geographical regions (Lindsey, 1995). A number of studies have attempted to document the health conditions encountered by homeless populations in specific facilities or regions (Blewett, Barnett, & Chueh, 1999; Nuttbrock, McQuistion, Rosenblum, & Magura, 2003; Plescia, Watts, Neibacher, & Strelnick, 1997; Spanowicz, Millsap, McNamee, & Bartek, 1998). It is apparent that certain conditions, such as trauma, res- piratory infections, dermatological conditions, mental illness and sub- stance abuse, are strongly associated with homelessness. Almost all other forms of chronic illness ? such as diabetes, osteoarthritis and high blood pressure ? that are common in both housed and homeless populations are made worse by homelessness because of the inability of homeless people to receive regular care or to self-manage the condition appro- priately. Moreover, diseases such as HIV/AIDS or tuberculosis, which require aggressive treatment, undoubtedly carry a much less favourable prognosis for homeless persons than for the general population. One indicator of the severity of these morbidities is the much higher rate of premature death among homeless persons compared to the housed pop- ulation (Hwang, 2000; Roy, Boivin, Haley, & Lemke, 1998). Despite this substantial burden of illness, homeless persons face a variety of barriers to receiving appropriate health care. A significant ob- stacle to accessing care in Canada is the absence of a valid entitlement document, i.e., a provincial health card (Hwang, Windrim, Svoboda, & Sullivan, 2000). Homeless people may be unable to afford supplies or medications that are not covered under provincial health care plans (On- tario Medical Review [ OMR], 1996). Physicians? offices are seldom lo- cated in areas where homeless people tend to congregate and are usually open only during regular office hours, posing transportation and sche- duling challenges (Gelberg et al., 2002; Kurtz, Surratt, Kiley, & Inciardi, 2005). Homeless people may encounter psychological barriers, such as fear that they will be refused care (Bunce, 2000) or feelings of stigmatiza- tion by health care providers (Gelber, Browner, Lejano, & Arangua, 2004). Finally, homeless individuals may delay seeking medical care be- cause other needs, such as securing food and shelter, are more critical to their daily survival. This chapter asks the question: What is the most effective way to de- liver point-of-first-contact or primary health care to homeless persons? A search of the literature revealed insufficient empirical sources to answer the question using standard systematic review methodology. Instead, we used a policy analysis approach. Data Retrieval A structured literature search was conducted for English-language pub- lications from 1990 to 2006 in the following databases: Medline, Embase, Cinahl and the Cochrane Library, Social Services Abstracts, Social Sciences Citation Index, Social Sciences Index, Sociological Abstracts, CBCA, Canadian Newsstand, JStor, Readers? Guide and PAIS Interna- tional. Throughout the study period, a ?My NCBI Alert? was used to deliver new search results from Medline (PubMed) on a weekly basis, and periodic update searches were conducted in the other databases. Search strategies for each database were developed using natural- language keywords and controlled vocabulary terms specific to each da- tabase. Three related searches covered the following topics: primary health care services for homeless persons; impact of primary health care services for homeless persons; and health problems of homeless persons. Additional sources were identified through a manual search of bib- liographies and references, and the World Wide Web was searched using Google (advanced search mode) to identify grey literature, organizations involved in providing services to the homeless, and examples of pro- grams providing primary care services to homeless persons. All refer- ences were recorded in a database created using Reference Manager 11. Analysis The search revealed that the literature, though extensive, was largely descriptive. There was insufficient empirical data to conduct a systematic review (Bravata, McDonald, Shojania, Sundaram, & Owens, 2005) of primary care delivery methods. There were also too few robust evalua- tions of primary care programs for homeless persons to permit a narra- tive synthesis (Dixon-Woods, Agarwal, Young, Jones, & Sutton, 2004). However, the existing literature did lend itself to a policy analysis ap- proach. Such an approach involves examining the relevance of specific research findings to a policy issue, weighing the evidence, and construct- ing a logical case about the utility of specific policy options for address- ing the issue in light of predetermined policy objectives (Aday & Begley, 1993). The key steps in policy analysis are articulating a broad policy goal; dividing that goal into measurable objectives; selecting evaluation crite- ria by which the attainment of objectives will be assessed; and judging how various policy options are most likely to perform when measured by these predetermined evaluation criteria. In the absence of definitive empirical evidence about the various policy options, this process neces- sarily represents the informed opinion of the policy analysis team. Results The policy goal is to ensure use of the most effective way to provide point-of-first-contact health care to homeless persons. Measurable objec- tives that support this goal may be taken from the seven defining attributes of appropriate primary health care recently identified by the Canadian Institute for Health Information through a comprehensive consultation process (Canadian Institute for Health Information [ CIHI], 2006). These attributes correspond closely to the seven desirable system- level service delivery attributes identified by the Working Group on Homeless Health Outcomes for the United States Department of Health and Human Services (United States Bureau of Primary Health Care, 1996). The objectives are: 1. Ensuring access to primary health care through a regular primary health care provider. 2. Enhancing the population orientation of primary health care ? for example, health promotion strategies that engage and mobilize the community. 3. Providing comprehensive whole-person care that addresses physi- cal, social and psychological dimensions. 4. Enhancing an integrated approach to 24/7 access. 5. Strengthening the quality of primary health care. 6. Building patient-centred care, that is, taking into account the pa- tient?s desire for information and decision-making in an empathetic and open manner. 7. Promoting continuity through integration and coordination. To ensure that the evaluation criteria for each of these objectives are specific to the needs of homeless persons, it is necessary to consult the literature describing the barriers that this disadvantaged population fac- es in obtaining primary care. That is, evaluation criteria are the adapta- tions to the delivery and structure of care necessary to counter the bar- riers. Such adaptations were summarized at the 1998 National Symposium on Homeless Research in the United States (McMurray- Avila, Gelberg, Breakey, & the National Symposium on Homelessness Research, 1998) and may be inferred from the many discussions of bar- riers to care faced by homeless persons (Bunce, 2000; Gelberg et al., 2004; McMurray-Avila et al., 1998; Ontario Women?s Health Council, 2002). The criteria deemed most relevant are listed in Table 1. What are the options for delivering primary care to homeless per- sons? The literature suggests four broad options, distinguished largely by the location at which care is delivered, but also by associated organi- zational features: the status quo based on independent family doctors? offices and three models directed specifically at homeless clients ? stan- dard facility/clinic site, fixed outreach site, and mobile outreach service. Although the literature on homelessness and health includes many brief descriptions of local interventions, no single paper provides a suffi- ciently generic experience upon which broad generalizations can be based. However, from papers on each specific model of care, it is possi- ble to extract common characteristics, which can then be reassembled into an archetypal description of that model. The idealized composite picture that emerges may serve as a paradigm of that model of care when assessing its potential effectiveness. Table 1. Evaluation criteria for homeless primary care Goal Objectives Evaluation criteria To enhance the health of homeless persons through the provision of optimal prima- ry care Ensuring access to prima- ry health care through a regular primary health care provider ? Entitlement documents not required for care or for ancillary services ? Service available at venues likely to suit homeless persons Enhancing the population orientation of primary health care ? Collaboration with public health author- ities on harm reduction strategies Providing comprehensive whole-person care ? Multidisciplinary team care ? Established referral routes for specialty services ? Social work assistance available for benefit entitlement, housing Enhancing an integrated approach to 24/7 access ? Service available at times likely to suit homeless persons ? Evidence of reduced emergency room use Strengthening the quality of primary health care ? Special expertise in areas germane to the clinical conditions of homeless per- sons, e.g., substance abuse, sexually transmitted diseases Building patient-centred care ? User involvement in service planning and operation Promoting continuity through integration and coordination ? Appropriate access to electronic medi- cal records by multiple providers ? Mechanisms to contact patients ? Hospital liaison for planning discharge Primary care status quo Many types of practices can be found in Canada, but physician-centred solo and small group practices are the norm. In the 2001 National Family Physician Workforce Survey, 73 percent of family doctors reported that private offices were their main practice setting. Solo practice is more common in inner cities, with 46 percent of family doctors in these areas reporting solo practice, compared to 19 percent in isolated or remote areas. Between 1989 and 2000, the number of physicians reporting that they operate ?office-only? practices ? meaning they did not make house calls, provide hospital or nursing home care, work in emergency de- partments, or provide obstetrical services ? rose from 14 percent to 24 percent. Most family doctors in Canada are paid on a fee-for-service basis. Physicians submit bills to provincial or territorial health insurance plans for each service provided. Alternative payment structures accounted for 11 percent of total clinical payments in 2000-2001 but are increasing (CIHI, 2003). In 2001, 94 percent of Canadians aged 15 and over received care from a family physician, commonly during regular office hours. However, almost one in five of those who sought ?first-contact? services in 2001 had difficulty accessing care at some point in that year (CIHI, 2003). The 2004 National Physician Survey found that only 20 percent of practices were open to new patients, and a Decima poll reported that five million Canadians over 18 years of age were unable to find a family doc- tor in the 12 months preceding the survey (College of Family Physicians of Canada [ CFPC], 2004). A recently described typology of Canadian primary care models summarized the status quo under the term ?professional contact model.? This model facilitates a care-seeking person?s ability to make first contact with the health care system. Individuals usually travel to the physician?s office, a single location where the physician may practise alone or in a group. Such physicians are rarely associated with other health profes- sionals and are commonly paid on a fee-for-service basis. With the pro- fessional contact model, there is no tool beyond patient loyalty to ensure long-term continuity of care, and there is no formal mechanism to ensure integration with other health services. The model facilitates accessibility and responsiveness to patients, but performs poorly in terms of effec- tiveness, productivity, equity, and quality (Canadian Health Services Research Foundation [ CHSRF] et al., 2003). Standard facility or clinic site Descriptions are available in the literature of standard facilities or clinic sites exclusively dedicated to serving homeless persons in Miami (Four- nier, Perez-Stable, & Greer, 1993), New York (Morrow, Halbach, Hopkins, Wang, & Shortridge, 1992) and Los Angeles (Gelber, Doblin, Leake, 1996); some additional details on the operation of such initiatives were drawn from other published sources. Such clinics may originate as charitable and volunteer initiatives, but generally are affiliated with an institution such as a hospital or community health centre. Academic links providing training for nursing and medical students are common. Care is delivered by multidisciplinary teams, with non-clinical services available from social workers or legal staff. Close connections are main- tained with social service agencies and public health units to which clients can be referred. A hospital affiliation facilitates referrals to spe- cialists, but some specialty care may be available on site. Clinics are often found near shelters, and in some cases outreach visits to these sites may take place. Typically, clinics have both daytime and evening hours of operation. The emphasis is on immediate care for acute illnesses, with the hope that persons requiring more complex care can be successfully integrated into the general health system. Screening and health education are common elements of care (Edwards, Kaplan, Barnett, & Logan, 1998; Macnee, Hemphill, & Letran, 1996). Care is pro- vided without charge, as are a limited range of medications and labora- tory tests. More sophisticated testing may be available from affiliated organizations. A significant number of patient encounters are repeat vis- its. A broad array of clinical services available in a timely manner may reduce emergency room use. Fixed outreach model A composite picture of fixed outreach programs can be constructed from descriptions of initiatives in New York (Plescia et al., 1997), Boston (Kline & Saperstein, 1992), New Orleans (Steele & O?Keefe, 2001), California (Fiore, 1995) and Ohio (DiMarco, 2000), with additional details extracted from other sources. ?Outreach? in this model refers to care that is pro- vided in non-traditional settings frequented by, or convenient to, home- less persons, in the absence of which such individuals would be unlikely to access services (Morse et al., 1996). The care may be delivered at schools (Berti, Zylbert, & Rolnitzky, 2001; Nabors et al., 2004), in community drop-in centres (Cunnane, Wy- man, Rotermund, & Murray, 1995; Reuler, 1991) or in transitional hous- ing settings (Rog, Holupka, & Combs-Thornton, 1995), but the most com- mon location is at shelters for the homeless. Regularly scheduled sessions are held at these venues and are staffed predominantly by nurses but with physicians, social workers, and counsellors on the team as well. Care is delivered without charge, and some medications may be available free of charge to patients. Mechanisms may be in place to ex- pedite registration for benefit programs for those patients who are eligi- ble. Services include acute care for minor and chronic conditions, pre- ventive care and education, and referral to other providers or agencies. Outreach clinics usually have good linkages with many other health and social agencies, including public health units to which patients can be referred; referrals to community clinics and specialty care at nearby hos- pitals are common. There may be formal administrative and funding ties between the outreach clinic and established health care facilities in the region. Brief clinical records are commonly kept, providing the basis for activity reports that focus on types and volume of services but only rare- ly on outcomes (Bradford, Gaynes, Kim, Kaufman, & Weinberger, 2005; Cunningham et al., 2005; Tischler, Vostanis, Bellerby, & Cumella, 2002). Increasingly, these records are kept in electronic format (Blewett et al., 1999). In a large number of cases patients are seen on only one occasion, but a small number of patients become regular users of these sites. By becoming frequent users with attendant documentation, such individu- als assist the clinics accomplish what is often their main goal in addition to the provision of immediate care: helping individuals reintegrate into mainstream care programs by eventually transferring care to more tradi- tional care venues. Mobile outreach service model Descriptions from New York (Redlener & Redlener, 1994) and Georgia (Testani-Dufour, Green, Green, & Carter, 1996; Tollett & Thomas, 1995), supplemented with details from other programs, provide sufficient in- formation to construct a composite picture of the mobile outreach service model. Mobile services operate from vehicles of various descriptions at sites convenient to homeless persons, such as at shelters or on the streets. Often the units visit their sites on a regular schedule so that clients can anticipate their arrival. The target population may be specialized, such as youth (Auerswald, Sugano, Ellen, & Klausner, 2006) or persons with mental illness (Farrell, Huff, MacDonald, Middlebro, & Walsh, 2005; Morris & Warnock, 2001), or it may focus on anyone without a home. Visits may be scheduled or offered on a walk-in basis, and there is no cost to the user. The services provided may be determined by a pre- liminary needs assessment and modified on the basis of subsequent client input. Space may limit the range and volume of services available, but common services include diagnosis, including the performance of basic laboratory tests; the treatment of acute and chronic conditions, for which a limited range of medications may be dispensed; screening and prevention activities; educational interventions; and referrals to other community agencies or specialized care. These services are provided by a team weighted towards nurses, but including a variable physician presence and other providers, such as social workers. Point-of-contact electronic records may be linked to a central database, and handheld devices may be used to enter new en- counter data (Buck, Rochon, & Turley, 2005; Bunschoten, 1994). Success may be measured by such programs on the basis of tabulations of the numbers of client encounters, repeat visits or referrals, or by surveying clients and providers. Sponsors may include independent charitable or- ganizations or health care institutions such as hospitals; extensive colla- boration with other agencies is common. Costs relative to other delivery methods are seldom reported because they are challenging to assess and may depend on location or funding source (Wray et al., 1999). Based on the data presented above, it is now possible, as shown in Table 2, to apply the evaluation criteria to the four options. Table 2. Evaluation of four models Evaluation criteria Status quo model Standard facility/ clinic site Fixed outreach site Mobile outreach service Entitlement documents not required for health care or for ancillary services poor excellent excellent excellent Service available at venues likely to suit homeless persons poor well excellent excellent Collaboration with public health authorities on harm reduction strategies poor well adequate adequate Multidisciplinary team care poor excellent excellent excellent Established referral routes for specialty services excellent excellent excellent adequate Social work assistance available for benefit entitlement, housing poor excellent excellent well Service available at times likely to suit homeless persons poor well adequate excellent Evidence of reduced emergency room use poor adequate unknown unknown Special expertise in areas germane to the clinical conditions of homeless persons, e.g., substance abuse, sexually transmitted diseases poor excellent excellent well User involvement in service planning and operation poor poor poor adequate Appropriate access to electronic medical records by multiple providers poor well adequate well Mechanisms to contact patients poor well Fair fair Hospital liaison for planning discharge poor unknown poor poor The status quo performs poorly by all but one of the 13 evaluation criteria. While there is variable performance on individual measures, the remaining three models all perform well. This finding implies that some factor other than performance on the specified measures should be used to choose a specific model. Such factors might include comparative costs, feasibility for staffing, geographic distribution of the population served or local preferences Conclusion Primary care in Canada has witnessed the appearance of a number of new models of payment and organization over the last two decades. Some of these may be better suited to meeting the needs of homeless persons than others, but the literature as yet contains no evidence to support this assertion. Indeed, the lack of published research on Cana- dian programs for the care of homeless persons was a striking finding in this project. To better understand this deficit, 42 primary care programs target- ing homeless individuals across Canada were approached to take part in key-informant interviews; 18 agreed. Not one was able to provide pub- lished or unpublished program descriptions or evaluations. There was a consensus among informants that the programs lacked the evaluation skills to create such documents and that any costs associated with creat- ing documents would reduce already inadequate clinical care budgets. It is easy to assume that a health system such as Canada?s, which provides universal first-dollar coverage, meets the health needs of home- less persons. But the concept of ?horizontal equity? that underlies the system ? equal needs receive equal resources ? fails to appreciate the different and far greater needs present in vulnerable groups. These populations require a system that incorporates ?vertical equi- ty,? that is, the capacity to meet unequal needs with unequal resources. The disproportionate burden of illness borne by the homeless population constitutes a dramatic inequality of health need, yet in comparison to specialized services designed to meet these needs, the current model of primary care in Canada is inadequate. To ignore this inadequacy by fail- ing to provide specialized care is to permit the operation of what has been termed the ?inverse care law,? which states that ?the availability of good medical care tends to vary inversely with the need for it in the population served? (Hart, 1971, p. 405). If, as has been proposed, a measure of any health system?s merit is the way in which it treats its most vulnerable citizens (Brownell, Roos, & Roos, 2001), Canada?s pri- mary care system must urgently address the health needs of the home- less population. S.E.D. Shortt is Director, Knowledge Transfer and Practice Policy, Canadian Medical Association, Ottawa, Ontario. Stephen W. Hwang is at the Centre for Research on Inner City Health, St. Michael?s Hospital, Toronto, and teaches in the Faculty of Medicine at the University of Toronto. Heather St u art is in the Department of Community Health and Epidemiology at Queen?s University. Melanie Bedore , Nadia Zurba, and Margaret Darling are associated with the Centre for Health Services and Policy Research, Queen?s University, Kings- ton. This chapter is based on ?Delivering Primary Care to Hom eless Persons: A Policy Analysis Approach to Evaluating the Options,? Health care Policy, vol. 1, no. 4, 2008. This study was made possible by an operating grant from the Canadian Institutes of Health Research. References Aday, L. A., & Begley, C. E. (1993). Evaluating the medical care system: Effectiveness, efficiency, and equity. Ann Arbor, MI: Health Administration Press. Auerswald, C. L., Sugano, E., Ellen, J. M., & Klausner, J. D. (2006). Street-Based STD testing and treatment of homeless youth are feasible, acceptable and effective. Journal of Adolescent Health, 38(3), 208-212. Begin, P., Casavant, L., Chenier, N. M., & Dupuis, J. (1999). Homelessness. Re- trieved June 23, 2008, from http://www.parl.gc.ca/information/library/PRB pubs/prb991-e.htm. Berti, L. C., Zylbert, S., & Rolnitzky, L. (2001). Comparison of health status of children using a school-based health center for comprehensive care. Journal of Pediatric Health Care, 15(5), 244-250. Blewett, D. R., Barnett, G. O., & Chueh, H. C. (1999). Experience with an electron- ic health record for a homeless population. In Proceedings of the American Medical Informatics Association Symposium (pp. 481-485). Bradford, D. W., Gaynes, B. N., Kim, M. M., Kaufman, J. S., & Weinberger, M. (2005). Can shelter-based interventions improve treatment engagement in homeless individuals with psychiatric and/or substance misuse disorders?: A randomized controlled trial. Medical Care, 43(8), 763-768. Bravata, D. M., McDonald, K. M., Shojania, K. G., Sundaram, V., & Owens, D. K. (2005). Challenges in systematic reviews: Synthesis of topics related to the delivery, organization, and financing of health care. Annals of Internal Medi- cine, 142(12, Pt. 2), 1056-1065. Brownell, M. D., Roos, N. P., & Roos, L. L. (2001). Monitoring health reform: A report card approach. Social Science and Medicine, 52(5), 657-670. Buck, D. S., Rochon, D., & Turley, J. P. (2005). Taking it to the streets: Recording medical outreach data on personal digital assistants. Computers, Informatics, Nursing, 23(5), 250-255. Bunce, D. (2000). Problems faced by homeless men in obtaining health care. Nurs- ing Standard, 14(34), 43-45. Bunschoten, B. (1994). Homeless projects show value of electronic records. Health Data Management, 2(3), 51-54. Canadian Health Services Research Foundation, Lamarche, P. A., Beaulieu, M., Pineault, R. Contandriopoulos A., & Denis, J. L. H. J. (2003). Choices for change: The path for restructuring primary health care services in Canada. Re- trieved June 23, 2008, from http://www.chsrEca/final_research/ commis- sioned_research/policy_synthes is/pdf/choices_for_change_e.pdf. Canadian Institute for Health Information (2003). Health Care in Canada. Re- trieved June 23, 2008, from http://secure.cihi.ca/cihiweb/products /hcic2003_e.pdf . Canadian Institute for Health Information (2006). Pan-Canadian primary health care indicator development project. Pan-Canadian Primary Health Care Indi- cator, 1(1-2). Retrieved June 23, 2008, from http://www.cihi.ca/cihiweb/ dis- pPage.jspkw_page=GR_1489_E . College of Family Physicians of Canada (2004). Family medicine in Canada: A vision for the future [Conference Proceedings]. Mississauga, ON: Author. Retrieved June 23, 2008, from http://www.cfpc.ca/local/files/Communications/ Healthpercent20Policy/FAMILY_MEDICINE_ IN_CANADA_English.pdf . Cunnane, E., Wyman, W., Rotermund, A., & Murray, R. (1995). Innovative pro- gramming in a community service center. Community Mental Health Journal, 31(2), 153-161. Cunningham, C. O., Shapiro, S., Berg, K. M., Sacajiu, G., Paccione, G., & Goulet, J. L. (2005). An evaluation of a medical outreach program targeting unstably housed HIV-infected individuals. Journal of Health Care for the Poor and Un- derserved, 16(1), 127-138. DiMarco, M. A. (2000). Faculty practice at a homeless shelter for women and children. Holistic Nursing Practice, 14(2), 29-37. Dixon-Woods, M., Agarwal, S., Young, B., Jones, D., & Sutton, A. (2004). Integra- tive approaches to qualitative and quantitative evidence. National Institute for Health and Clinical Excellence. Retrieved June 23, 2008, from http://www.nice.org.uk:80/niceMedia/pdf/Integrative_approaches_ evidenc e.pdf. Edwards, J. B., Kaplan, A., Barnett, J. T., & Logan, C. L. (1998). Nurse-managed primary care in a rural community: Outcomes of five years of practice. Nursing and Health Care Perspectives, 19(1), 20-25. Farrell, S. J., Huff, J., MacDonald, S. A., Middlebro, A., & Walsh, S. (2005). Tak- ing it to the street: A psychiatric outreach service in Canada. Community Mental Health Journal, 41(6), 737-746. Fiore, D. C. (1995). A homeless shelter medical clinic organized and staffed by family practice residents. Western Journal of Medicine, 163(6), 537-540. Fournier, A. M., Perez-Stable, A., & Greer, P. J., Jr. (1993). Lessons from a clinic for the homeless: The camillus health concern. Journal of the American Medi- cal Association, 270(22), 2721-2724. Gelber, L., Browner, C. H., Lejano, E., & Arangua, L. (2004). Access to women?s health care: A qualitative study of barriers perceived by homeless women. Women and Health, 40(2), 87-100. Gelber, L., Doblin, B. H., & Leake, B. D. (1996). Ambulatory health services pro- vided to low-income and homeless adult patients in a major community health center. Journal of General Internal Medicine, 11(3), 156-162. Gelber, L., Leake, B., Lu, M. C., Andersen, R., Nyamathi, A. M., Morgenstern, H., & Browner, C. (2002). Chronically homeless women?s perceived deterrents to contraception. Perspectives on Sexual and Reproductive Health, 34(6), 278- 285. Hart, J. T. (1971). The inverse care law. Lancet, 1(7696), 405-412. Human Resources and Social Development Canada (2008). Homelessness partner- ing strategy. Retrieved June 23, 2008, from http://www.homelessness.gc.ca/about_us/ understanding_homelessness _e.a sp. Hwang, S. W. (2000). Mortality among men using homeless shelters in Toronto, Ontario. Journal of the American Medical Association, 283(16), 2152-2157. Hwang, S. W. (2001). Homelessness and health. Canadian Medical Association Journal, 164(2), 229-233. Hwang, S. W, Windrim, P. M., Svoboda, T. J., & Sullivan, W. F. (2000). Physician payment for the care of homeless people. Canadian Medical Association Jour- nal, 163(2), 170-171. Kline, E. N., & Saperstein, A. B. (1992). Homeless women: The context of an ur- ban shelter. Nursing Clinics of North America, 27(4), 885-899. Kurtz, S. P., Surratt, H. L., Kiley, M. C., & Inciardi, J. A. (2005). Barriers to health and social services for street-based sex workers. Journal of Health Care for the Poor and Underserved, 16(2), 345-361. Lindsey, A. M. (1995). Physical health of homeless adults. Annual Review of Nurs- ing Research, 13, 31-61. Macnee, C. L., Hemphill, J. C., & Letran, J. (1996). Screening clinics for the home- less: Evaluating outcomes. Journal of Community Health Nursing, 13(3), 167- 177. McMurray-Avila, M., Gelberg, L., Breakey, W. R., & the National Symposium on Homelessness Research (1998). Balancing act: Clinical practices that respond to the needs of homeless people. Retrieved June 23, 2008, from http://aspe.hhs.gov/progsys/homeless/symposium/8-Clinical.htm. Morris, D. W., & Warnock, J. K. (2001). Effectiveness of a mobile outreach and crisis services unit in reducing psychiatric symptoms in a population of homeless persons with severe mental illness. Journal of the Oklahoma State Medical Association, 94(8), 343-346. Morrow, R., Halbach, J. L., Hopkins, C., Wang, C., & Shortridge, L. A. (1992). A family practice model of health care for homeless peopleL Collaboration with family nurse practitioners. Family Medicine, 24(4), 312-326. Morse, G. A., Calsyn, R. J., Miller, J., Rosenberg, P., West L., & Gilliland, J. (1996). Outreach to homeless mentally ill people: Conceptual and clinical consider- ations. Community Mental Health Journal, 32(3), 261-274. Nabors, L. A., Weist, M. D., Shugarman, R., Woeste, M. J., Mullet, E., & Rosner, L. (2004). Assessment, prevention, and intervention activities in a school- based program for children experiencing homelessness. Behavior Modifica- tion, 28(4), 565-578. Nuttbrock, L., McQuistion, H., Rosenblum, A., & Magura, S. (2003).Broadening perspectives on mobile medical outreach to homeless people. Journal of Health Care for the Poor and Underserved, 14(1), 5-16. Ontario Medical Review (1996). Exploring the health impact on homelessness. Re- trieved June 23, 2008, from http://www.oma.org/phealth/homeless.htm. Ontario Women?s Health Council (2002). Health status of homeless women: An in- ventory of issues. Retrieved June 23, 2008, from. http://www.womenshealthcouncil.on.ca/userfiles/page_attachments/library /1/Health_Status_of Homeless_Women_EN _2047_ 695725.pdf . Plescia, M., Watts, G. R., Neibacher, S., & Strelnick, H. (1997). A multidisciplinary health care outreach team to the homeless: The 10-year experience of the Montefiore Care for the Homeless Team. Family and Community Health, 20(2), 58-69. Redlener, I., & Redlener, K. B. (1994). System-based mobile primary pediatric care for homeless children: The anatomy of a working program. Bulletin of the New York Academy of Med icine, 71(1), 49-57. Reuler, J. B. (1991). Outreach health services for street youth. Journal of Adolescent Health, 12(7), 561-566. Rog, D. J., Holupka, C. S., & Combs-Thornton, K. L. (1995). Implementation of the Homeless Families Program: Service models and preliminary outcomes. American Journal of Orthopsychiatry, 65(4), 502-513. Roy, E., Boivin, J. F., Haley, N., & Lemke, N. (1998). Mortality among street youth. Lancet, 352(9121), 32. Spanowicz, M. J., Millsap, G., McNamee, M. J., & Bartek, J. K. (1998). Health problems of sheltered homeless men using a mobile health van: A 4-year study. Clinical Excellence in Nursing Practice, 2(5), 279-285. Statistics Canada (2002). Collective dwellings. Retrieved June 23, 2008, from http://www12.statcan.ca/english/census01/products/analytic/companion/col l/contents.cfm#tble2 . Statistics Canada (2008). Population in collective dwellings by province and territory. Retrieved June 23, 2008, from http://www40.statcan.ca/101/cst01 /famil62a.htm?sdi=collective percent20dwelling. Steele, R. W., & O?Keefe, M. A. (2001). A program description of health care in- terventions for homeless teenagers. Clinical Pediatrics, 40(5), 259-263. Testani-Dufour, L., Green, L., Green, R., & Carter, K. F. (1996). Establishing out- reach health services for homeless persons: An emerging role for nurse managers. Journal of Community Health Nursing, 13(4), 221-235. Tischler, V., Vostanis, P., Bellerby, T., & Cumella, S. (2002). Evaluation of a men- tal health outreach service for homeless families. Archives of Disease in Childhood, 86(3), 158-163. Tollett, J. H., & Thomas, S. P. (1995). A theory-based nursing intervention to in- still hope in homeless veterans. Advances in Nursing Science, 18(2), 76-90. United States Bureau of Primary Health Care (1996). The Working Group on Home- less Outcomes Meeting. Rockville, MD: Author. Wray, N. P., Weiss, T. W., Menke, T. J., Gregor, P. J., Ashton, C. M., Christian, C. E., & Hollingsworth, J. C. (1999). Evaluation of the VA Mobile Clinics dem- onstration project. Journal of Health care Management, 44(2), 133-147. Chapter 3 .1 Homeless Youth: The Need to Link Research and Policy SEAN A. KIDD AND LARRY DAVIDS ON Homeless youths are typically defined as a group of adolescents and young adults, ranging in age from approximately 12 to 24, who live in shelters, on the streets, in abandoned buildings, or who otherwise do not have an adequate place to dwell (stable, with appropriate shelter and amenities) that serves as a permanent home (U.S. Department of Educa- tion, 1989). These young people are concentrated primarily in large ur- ban centres and, though the accuracy of estimates regarding the extent of youth homelessness is questionable, statistics from the United States suggest that numbers are increasing, with at least one million youths thought to be homeless in North America at the present time (Kidd & Scrimenti, 2004). While there may be myriad reasons behind a given youth?s becom- ing homeless, the research literature has highlighted some common fac- tors. These include young people being thrown out of, or running away from, homes in which abuse and neglect are occurring (Maclean et al., 1999; Molnar et al., 1998; Ringwalt et al., 1998). Also frequent are histo- ries of domestic violence, parental criminality and substance abuse, and poverty (Buckner & Bassuk, 1997; Hagan & McCarthy, 1997; Maclean et al., 1999; Ringwalt et al., 1998). On the streets, homeless youth face numerous serious risks. These include high rates of victimization (Whitbeck et al., 2000) and ongoing problems finding shelter and maintaining an appropriate diet (Anto- niades & Tarasuk, 1998). Drug use and dependence are common (Greene & Ringwalt, 1996), as are sex trade involvement (Kidd & Kral, 2002) cri- minality (Hagan & McCarthy, 1997), serious mental illness (Rotheram- Borus, 1993; Whitbeck et al., 2000), poor physical health and high inci- dences of communicable disease (Booth et al., 1999), and suicidal thoughts and attempts (Kidd, 2006). Mortality rates are extremely high, with conservative estimates indicating rates in Canada of up to 11 times that of the general youth population (Roy et al., 1998). Mortality rates up to 40 times greater than the average have been noted in the United King- dom (Shaw & Dorling, 1998). Suicide is the leading cause of death for homeless youth (Roy et al., 2004). Substantially less information is available on the services provided to homeless youths. The few studies that have been conducted generally attest to the difficulty in providing effective services for this group that achieve lasting gains in mental and physical health domains (Barry et al., 2002; Booth et al., 1999; Cauce et al., 1994; Thompson et al., 2002). The numerous and interacting risks accompanied by unstable living circums- tances and structural barriers combine to hamper the effectiveness of any number of innovative and well-conceived interventions. Thus far, there is a lack of any clear distinction between the American and Canadian literature with respect to the etiology, risk, and intervention among homeless youths. Findings to date would not, however, suggest marked differences. Of the range of factors involved in youth homelessness, socio- cultural and policy issues have received the least attention in the mental and physical health literature. This circumstance prevails despite emerg- ing evidence that social stigma at both public (insults, physical assaults, denial of employment and housing due to homeless status) and structur- al (multiple arrests, inadequate funding for services) levels can signifi- cantly heighten levels of risk and hamper intervention efforts (Kidd, 2003; Shissel, 1997). The picture in Canada regarding policy and legislation suggests numerous areas of difficulty. Ongoing problem areas that likely com- pound and contribute to the problem of youth homelessness include: ? difficulty accessing income support (Gaetz & O?Grady, 2002; Raising the Roof, 2001) unemployment insurance and disability payments (City of Toronto, 2001); ? increasing criminalization and disproportionate arrests (Eberle, 2001; Novac et al., 2002; O?Grady & Greene, 2003); ? declines in affordable housing, difficulty accessing socially sup- ported housing, increasing evictions (Kidd et al., 2007); ? breakdown in continuity of care due to child welfare service age cu- toffs (ranging from 16-19) (Novac et al., 2002) and youth services age cutoffs (typically 24) (Kidd et al., 2007); ? a lack of services specific to the needs of homeless youths such as adequate discharge planning from health care and criminal justice systems (Raising the Roof, 2001). Also lacking are employment skills training programs, acute treat- ment centres (e.g., detox programs) and temporary housing (Eberle, 2001), and educational programs such as alternative high schools (Jo- sephson, n.d.). The narratives of homeless youth reflect these barriers, filled as they are with descriptions of multiple arrests due to restrictions on their rights to use public spaces or solicit money, problems finding affordable housing or employment, and difficulties negotiating social assistance services (Kidd, 2003). Given this situation, there would seem to be a compelling need for researchers to launch an examination of larger social processes and social policy as they affect homeless youth. The following represents a propos- al for prevention and intervention efforts that might serve to better ad- dress youth homelessness. These recommendations emphasize linkages between research and policy. Addressing the contexts from which youths run and are thrown out The difficulty in leaving the streets once a youth becomes entrenched in street culture (Barry et al., 2002) suggests that the greatest impact might be made with youths at risk of becoming homeless. Avenues of preven- tion are, however, under-represented in the literature and in service pro- vision initiatives (Karabanow & Clement, 2004). An important direction for future work lies in examining programs that serve to ? identify and intervene with youths at risk of becoming homeless and their families in various contexts such as schools and mental health service provision settings; ? reduce the likelihood of youths becoming homeless following prob- lematic experiences in contexts such as child welfare and criminal justice systems - both common pathways into homelessness (Kara- banow, 2004); ? increase opportunities for impoverished children (e.g., after-school programs). Policy and social stigma Research and public policy both have the potential to substantially ame- liorate the social conditions that contribute to the youth homelessness problem. Some proposed strategies in this area include the following: ? In public statements, government representatives can claim some degree of collective responsibility for the homeless youth situation, citing failures in policy and publicly funded services - rather than individual deficits or behaviours that have led to increasing numbers of children and youth on the streets. ? Challenging legislation that results in the criminalization of home- less youth and hampers their ability to access public assistance and supported housing. These challenges may be undertaken either by legislators or through legal action. ? Research can be used to highlight the extent of the social stigma faced by homeless youth, the erroneous foundations upon which stigma is based, and the damage caused by stigmatization and pre- judicial policy/legislation (O?Grady & Greene, 2003).These kinds of data - such as work which has shown disproportionate arrests of homeless adults for primarily non-violent, minor, and victimless crimes (Snow et al., 1989) and findings that social stigma contributes to suicidality among homeless youth (Kidd, 2004) ? can be used to undermine biased portrayals appearing in media and policy and in- crease understanding of systemic factors as they influence the lives of homeless young persons. ? More important, perhaps, is the direct impact that research can have on policy. In this arena, researchers have the potential to have a ma- jor constructive impact on the lives of homeless youth through care- ful examination of the various impacts of existing policy, analysis of the outcomes of emerging legislation, and evaluation of ways in which existing policy might be altered to allow for improved out- comes. Such an impact has been made previously in areas such as domestic violence, sexual harassment, and prejudice against sexual minorities (Keuhl, 2000). Pathways off the streets As noted above, along with the compelling need to develop an evidence base regarding the effectiveness of interventions for homeless youths, there needs to occur a careful review of funding practices and strategies in a manner informed by research. Current knowledge (Josephson, n.d.; Kidd, 2003) would suggest that the following strategies may represent a significant improvement upon existing practices: ? Funding investigation/evaluation by multidisciplinary task forces comprised of researchers and stakeholders for the purpose of gene- rating solutions ranging from interventions focusing on individuals to changes in legislation. ? Providing funding commensurate with both the extent of the prob- lem in terms of numbers of homeless youth and the complexity of their needs. This will include recognizing, in ongoing evaluation of funded programs, the need to measure ?success? in a way that ac- knowledges the unique challenges of homelessness (i.e., success should not be measured solely by the number of youth who get jobs, return to school, and/or become housed; success may also mean a youth surviving the winter or using fewer harmful substances). ? Providing sustained funding. A key to the effectiveness of interven- tions is the degree to which they do not resemble street existence or problematic home environments (Kidd et al., 2007). Stated different- ly, adequate and sustained funding allows for consistency in service provision and programming and lowers staff turnover, allowing for a thorough assessment of youth needs and the development of stable and trusting relationships with service providers. Conclusion At present, there are major gaps in knowledge transfer with regards to youth homelessness. Beyond publication in academic journals, there would appear to exist few examples of knowledge disseminated in prac- tical formats to youth workers or used to inform policy development at any level. In addition to difficulties surrounding the transfer of existing findings, it is essential that research follows through on increasing calls for the development and examination of primary prevention efforts and attention to the impacts of social stigma and legislation as they impact this population. This need for academic researchers to move beyond the role of ?knowledge gatherer? and engage in more active participation in policy development, evaluation, dissemination of practical and accessi- ble knowledge, and advocacy is increasingly entering the forefront of critical examinations of various fields (Prilleltensky, 1997). Sean A. Kidd is an Assistant Professor in the McMaster Department of Psy- chiatry and Behavioural Sciences. Larry Davidson is with the Program on Recovery and Community Health, Yale University, New H aven, Connecticut. This chapter is drawn from a paper originally published by the Canadian Jour- nal of Public Health, Nov/Dec 2006. References Antoniades, M., & Tarasuk, V. (1998). A survey of food problems experienced by Toronto street youth. Canadian Journal of Public Health, 89(6), 371-375. Barry, P. J., Ensign, J., & Lippek, S. H. (2002). Embracing street culture: Fitting health care into the lives of street youth. Journal of Transcultural Nursing, 13(2), 145-152. Booth, R. E., Zhang, Y., & Kwiatkowski, C. F. (1999). The challenge of changing drug and sex risk behaviors of runaway and homeless adolescents. Child Abuse & Neglect, 23(12), 1295-1306. Buckner, J. C., & Bassuk, E. L. (1997). Mental disorders and service utilization among youths from homeless and low-income household families. Journal of the American Academy of Child and Adolescent Psychiatry, 36(7), 890-900. Cauce, A. M., Morgan, C. J., Wagner, V., Moore, E., Sy, J., Wurzbacher, K., Wee- den, K., Tomlin, S., & Blanchard, T. (1994). Effectiveness of intensive case management for homeless adolescents: Results of a 3-month follow-up. Journal of Emotional and Behavioral Disorders, 2, 219-227. City of Toronto (2001). Toronto report card on homelessness. Toronto: Author. Eberle, M. (2001). The relationship between homelessness and the health, social services and criminal justice systems: A review of the literature. Vancouver: British Co- lumbia Ministry of Social Development and Economic Security. Gaetz, S., & O?Grady, B. (2002). Making money: Exploring the economy of young homeless workers. Work, Employment & Society, 16(3), 433-456. Greene, J. M., & Ringwalt, C. L. (1996). Youth and familial substance use?s associ- ation with suicide attempts among runaway and homeless youth. Substance Use & Misuse, 31(8), 1041-1058. Hagan, J., & McCarthy, B. (1997). Mean streets: Youth crime and homelessness. Cam- bridge: Cambridge University Press. Josephson, G. J. (n.d.). Relating homelessness to education, employment and income support: A review of Canadian Literature. Retrieved June 9, 2005, from http://policyresearch.gc.ca.proxy.lib.uwaterloo.ca. Karabanow, J. (2004). Making organizations work: Exploring characteristics of anti-oppressive organizational structures in street youth shelters. Journal of Social Work, 4(1), 47-60. Karabanow, J., & Clement, P. (2004). Interventions with street youth: A commen- tary on the practice-based research literature. Brief Treatment and Crisis In- tervention, 4, 93-108. Keuhl, S. J. (2000). Seeing is believing: Research on women?s sexual orientation and public policy. Journal of Social Issues, 56(2), 351-359. Kidd, S. A. (2003). Street youth: Coping and interventions. Child and Adolescent Social Work Journal, 20(4), 235-261. Kidd, S. A. (2004). ?The walls were closing in and we were trapped?: A qualita- tive analysis of street youth suicide. Youth & Society , 36(1), 30-55. Kidd, S. A. (2006). Factors precipitating suicidality among homeless youth: A quantitative follow-up. Youth & Societ y, 37(4), 393-422. Kidd, S. A., & Kral, M. J. (2002). Street youth suicide and prostitution: A qualita- tive analysis. Adolescence, 37(146), 411-430. Kidd, S. A., Miner, S., Walker, D., & Davidson, L. (2007). Stories of working with homeless youth: On being ?mind-boggling.? Children and Youth Services R e- view, 29(1), 16-34. Kidd, S. A., & Scrimenti, K. (2004). The New Haven homeless count: Children and youth. Evaluation Review, 28, 325-341. Maclean, M. G., Embry, L. E., & Cauce, A. M. (1999). Homeless adolescents? paths to separation from family: Comparison of family characteristics, psycholog- ical adjustment, and victimization. Journal of Community Psychology, 27(2), 179-187. Molnar, B. E., Shade, S. B., Kral, A. H., Booth, R. E., & Watters, J. K. (1998). Sui- cidal behavior and sexual/physical abuse among street youth. Child Abuse and Neglect, 22(3), 213-222. Novac, S., Serge, L., Eberle, M., & Brown, J. (2002). On her own: Young w omen and homelessness in Canada. Ottawa: Status of Women Canada. O?Grady, S., & Greene, C. (2003). A social and economic impact study of the On- tario Safe Streets Act on Toronto squeegee workers. Online Journal of Justice Studies, 1(1). [This journal is no longer available online.] Prilleltensky, I. (1997). Values, assumptions, and practices: Assessing the moral implications of psychological discourse and action. The American Psycholo- gist, 52(5), 517-535. Raising the Roof (2001). From street to stability: A compilation of findings on the paths to homelessness and its prevention. Toronto: Author. Ringwalt, C. L., Greene, J. M., & Robertson, M. J. (1998). Familial backgrounds and risk behaviors of youth with thrownaway experiences. Journal of Ado- lescence, 21(3), 241-252. Rotheram-Borus, M. J. (1993). Suicidal behavior and risk factors among runaway youths. American Journal of Psychiatry, 150(1), 103-107. Roy, E., Boivin, J. F., Haley, N., & Lemire, N. (1998). Mortality among street youth. The Lancet, 352(9129), 32. Roy, E., Haley, N., Leclerc, P., Sochanski, B., Boudreau, J., & Boivin, J. (2004). Mortality in a cohort of street youth in Montreal. The Journal of the American Medical Association, 292(5), 569-574. Shaw, M., & Dorling, D. (1998). Mortality among street youth in the UK. The Lan- cet, 352(9129), 743. Shissel, B. (1997). Blaming children. Halifax: Fernwood. Snow, D. A., Baker, S. G., & Anderson, L. (1989). Criminality and homeless men: An empirical assessment. Social Problems, 36(5), 532-549. Thompson, S. J., Pollio, D. E., Constantine, J., Reid, D., & Nebbitt, V. (2002). Short-term outcomes for youth receiving runaway and homeless shelter services. Research on Social Work Practice, 12(5), 589-603. U.S. Department of Education (1989). Report to Congress on final reports submitted by States in accordance with Section 724 (b) (3 ) of the Stewart B. McKinney Hom e- less Assistance Act. Washington, DC: Author. Whitbeck, L. B., Hoyt, D. R., & Bao, W. (2000). Depressive symptoms and co- occurring depressive symptoms, substance abuse, and conduct problems among runaway and homeless adolescents. Child Development, 71(3), 721- 732. Chapter 3.2 Whose Safety Counts? Street Youth, Social Exclusion, and Criminal Victimization STEPHEN GAET Z Understanding street youth victimization When homeless youth are discussed during public debates on crime, it is usually with reference to their role as perpetrators. This perspective, rooted in popular and enduring notions of delinquent street urchins, typically characterizes homeless youth as kids who are ?bad? or ?devi- ant? (or, more generously, troubled or misguided) and who leave home for fairly insignificant reasons. Once on the streets, they become involved in delinquent activities and, as a result, put the health and safety of the general public is at risk. It is ?they? who are causing problems for ordi- nary citizens; it is ?they? who are driving away tourists and making the streets unsafe. The persistent public focus on street youth as potential offenders overlooks the real possibility that they may disproportionately be victims of crime. Young people who are homeless experience much higher levels of crimi- nal victimization than other Canadians. An emerging body of literature explores the complex factors that result in higher levels of victimization among the homeless (Baron, 1997, 2003; Fitzpatrick et al., 1999; Tyler et al., 2000; Whitbeck et al, 1997, 2001; In the case of homeless youth, a consensus has emerged suggesting that a majority of street youth in Canada and the United States come from homes characterized by high levels of physical, sexual, or emotion- al abuse and neglect, compared with domiciled youth (Alder, 1991; De- matteo et al., 1999; Whitbeck & Simons, 1990). As Fitz- patrick et al. argue, homelessness is ?a stress-filled, dehumanizing, dan- gerous circumstance in which individuals are at high risk of being wit- ness to or victims of a wide range of violent acts? (1999, p. 439). Much of this research, reflecting the broader findings of sociological and crimino- logical research, identifies the significance of background variables and, in particular, the effects of previous victimization on future occurrences (Lauritsen & Quinet, 1995; Terrell, 1997; Tyler et al., 2000). Gaetz et al., 1999; Janus et al., 1987, 1995; Kufeldt & Nimmo, 1987; Whitbeck & Hoyt, 1999; Whitbeck & Simons, 1993). Rothe- ram-Borus et al. (1996) estimate that street youth are five times as likely as domiciled youth to report having been victims of sexual abuse as child- ren. These young people are likely to experience low self-esteem, an im- paired ability to form affective and trusting relationships with adults, higher rates of depression and suicide attempts, running away, or being kicked out of home (Beitchman et al., 1992; Tyler et al., 2000; Whitbeck et al., 1997). There is also evidence to suggest that an abusive background cha- racterized by coercive and aggressive parenting produces aggression in children and adolescents (Baron, 1997; Baron & Hartnagel, 1998; Fleisher, 1995; Patterson et al., 1984; Patterson et al., 1989; Whitbeck & Hoyt, 1999), who are also more likely to exhibit deviant peer associations and to en- gage in risky behaviours (Kral et al., 1997; MacDonald et al., 1994; Whit- beck et al., 1997, 2001). Youth cultural factors are important here as well. That is, the ?informal rules? that develop on the streets are, in part, a result of such aggressive upbringing and may condition homeless youth to adopt more ?violent? approaches to problem solving (Anderson, 1996; Baron et al., 2001; Terrell, 1997). Similarly, background variables are also correlated with later victi- mization on the streets (Baron, 1997; Browne & Bassuk, 1997; Kipke et al., 1997; Tyler et al., 2000; Whitbeck et al., 1997), in part because the aggres- sive behaviours produced by a violent upbringing may often lead to provocative interactions (Baron, 1997; Fleisher, 1995). In addition, there is evidence that victims of sexual abuse are at increased risk for sexual vic- timization and exploitation when they are older (Janus et al., 1987; Si- mons & Whitbeck, 1991). While background factors help explain deviant and violent beha- viour ? as well as experiences of victimization ? other factors must be taken into account (Whitbeck & Simons, 1990). In making sense of the criminal offending behaviour of homeless youth, Hagan and McCarthy (1997) effectively demonstrate the significance of situational factors. At the same time, lifestyle and routine activities theories highlight the con- textual significance of environmental and situational factors in increasing one?s exposure to the risk of criminal victimization (Cohen & Felson, 1979; Cohen et al., 1981; Hindelang et al., 1978; Miethe & Meier, 1990). Routine activities theory suggests that three conditions increase the opportunity for a crime to occur: a motivated offender, a suitable target, and a lack of capable guardianship (Cohen & Felson, 1979). If one regu- larly frequents dangerous and poorly supervised locations or engages in delinquent behaviours, one?s proximity to other criminal offenders plac- es one at greater risk for victimization (Kennedy & Forde, 1990; Laurit- sen et al., 1991). For young people who are homeless, the implications are clear. Their lives are played out in spaces that bring them into contact with hostile strangers, potential offenders, other homeless people, and people with serious substance abuse issues or mental health problems. Their low level of guardianship (Miethe & Meier, 1994) limits their ability to protect themselves or to be protected, making them suitable targets. An additional lifestyle factor to consider is that street youth, as a group, are more likely to engage in criminal and delinquent activities (Hagan & McCarthy, 1997; Inciardi et al., 1993). Criminological research suggests a link between criminal offending and victimization (Lauritsen et al., 1991). That is, many of the same factors that enable offending be- haviours ? dangerous locations, proximity to other offenders, weak guardianship ? may also lead to victimization (Esbensen & Huizinga, 1991; Rapp-Paglicci & Wodarski, 2000; Rivara et al., 1995). Social exclusion and victimization Lifestyle and routine activities theories suggest that certain social and ecological conditions raise one?s potential risk of personal victimization, both through increased exposure to potential offenders or dangerous situations and through a compromised ability to protect oneself, remove oneself from a dangerous situation, or rely on public safety resources such as the police. Such theories do not, however, explain how and why victimized persons wind up in such circumstances in the first place (Miethe & Meier, 1994; Sampson & Lauritsen, 1990). The notion of ?life- styles? suggests that individuals choose such environments, activities, or associations; and that by making different choices, potential victims could lessen their risk. the process of being shut out, fully or partially, from any of the social, economic, political or cultural systems which determine the social in- tegration of a person in society. Social exclusion may, therefore, be seen as the denial (or non-realization) of the civil, political, and social rights of ci- tizenship. (Walker & Walker, 1997, p. 8) While not dismissing the significance of agency, one must take ac- count of systemic factors that may profoundly limit choice and increase the risk of victimization. The concept of social exclusion allows one to ex- tend routine activities theory by exploring the degree to which the per- sonal histories of individuals intersect with social, political, and eco- nomic conditions that restrict people?s access to spaces, institutions, and practices that reduce risk. Such an account begins with a recognition that marginalized groups and individuals are often socially, economically, and spatially separated from the people and places to which other citi- zens have access within advanced industrial societies (Mandanipour, 1998). Social exclusion is defined as For young people who become homeless, social exclusion is expe- rienced in terms of access to shelter and housing, employment, and a healthy lifestyle, for instance. It is also manifest in their restricted access to (and movement within) urban spaces and their limited social capital. In most cases, the process of social exclusion begins before street youth become homeless, but it intensifies through their experience living on the streets. This experience of social exclusion is cumulative, making it diffi- cult to escape, particularly when constant exposure to risk compromises health, safety, and opportunity. As an outcome of their homelessness, street youth are typically pushed into places and circumstances that im- pair their ability to ensure their safety and security and, consequently, increase their risk of criminal victimization. Being without secure shelter means that the day-to-day lives of homeless youth are played out in a public environment over which they have limited control and within which their freedom of movement is restricted. They spend a large amount of their time on the sidewalks and streets, and in the parks and alleyways, of large cities. Their ?right? to inhabit many of these public spaces is often called into question; street youth regularly report being ?kicked out? of street locations and parks by police in the past year (Gaetz, 2002). Their use of semi-public spaces such as shopping malls is also more constrained than most people?s, as they are often denied service or asked to leave by security staff. The risks of proximity to other offenders cannot be reduced by re- treating to a safe domicile. Even when they are tired, ill, or under the influence of alcohol or drugs, they cannot recover in a secure environ- ment. The alternative is overcrowded social service environments where their health and safety are also jeopardized. Street youth are thus pushed into marginalized spaces where they are exposed to the ongoing risk of assault and property crime. Homeless people are often forcibly removed from safer spaces in the city and relegated to spaces that are potentially more dangerous, where they have less control over whom they interact with. Street youth, whether they are working, resting, or enjoying social interactions, are continually exposed to other potential offenders. The fact that many of their peers are also homeless and more likely to adopt aggressive and violent behaviours as an adaptive strategy for life on the streets may also increase their likelihood of victimization. Recent research (Gaetz & O?Grady, 2002) demonstrates that most homeless youth do not avoid work, but the vast majority face significant barriers to obtaining and maintaining employment. When they do find work, it is often in short-term, dead-end jobs or in unregulated work on the margins of the economy. As a result, they engage in risky money- making strategies, some of them illegal or quasi-legal, including the sex trade, panhandling (begging), squeegeeing (cleaning car windshields), and criminal acts such as theft and drug dealing. The subsistence strategies of the homeless affect their safety and the degree of risk they are exposed to (Russell & Robertson, 1998; Terrell, 1997; Tyler et al., 2000; Whitbeck et al., 1997). Street youth are more likely to be in contact with others who may be deviant or dangerous, and they may place themselves in a more vulnerable position relative to more powerful criminals (pimps, drug suppliers). Because their money- making activities are often highly visible (prostitution, panhandling, squeegeeing) and produce cash-in-hand on a daily basis, street youth present attractive targets, despite their seeming poverty. A final manifestation of the social exclusion experienced by street youth stems from their weak guardianship and lack of protection. Their involvement in delinquent acts increases the likelihood of negative inte- ractions with the police. Potential offenders thus may contemplate com- mitting acts of robbery or violence against homeless youth, knowing that the victim is less likely to seek the involvement of the police (Baron, 1997; Sparks, 1982). Homeless youth, then, experience social exclusion in their inade- quate access to housing and employment, their restricted access to public and semi-public spaces, and their weak social capital. The data presented here highlight some of the consequences of this social exclusion. First, I demonstrate that street youth are much more likely than domiciled youth (aged 15-24) to be victims of a range of personal crimes, and that this cannot be explained merely in terms of their offending behaviour. Second, I argue that when street youth are victims of crime, they general- The ability of street youth to avoid victimization is also limited by their weak social capital. Street youth cannot easily obtain support from authority figures (parents, teachers, the police) to protect them or their property or to assist them when they are victims of a crime. Street youth depend heavily on other street youth (whose social capital is likewise weak, and who may also be potential offenders) and the staff at street youth agencies to provide these resources. Unfortunately, alienation and difficulty in forming attachments and trusting relations with adults ? and with other street youth, for that matter ? may be one consequence of victimization, which, in turn, may increase risk. ly rely on a narrower set of social supports Method to help them deal with the consequences. Third, I explore the degree to which street youth are re- stricted in their ability to effectively engage in strategies to protect them- selves. Finally, an effort will be made to examine how gender shapes the experience of social exclusion of street youth. The data presented here are part of a larger study of legal and justice issues facing street youth involving surveys and interviews with 208 homeless youth living in Toronto. Each person was asked to fill out a structured, self-administered questionnaire consisting of 55 questions. Those with literacy problems were assisted by our research team, which included several current and former street youth. Upon completing the questionnaire, each respondent was asked to sit for a structured inter- view (conducted privately) to provide qualitative data to supplement the survey questionnaire. Where possible, we compared data from this study with recent and broader-based criminal victimization research in Canada ? in this case, Statistics Canada?s General Social Survey (GSS) (Statistics Canada, 1999). The GSS was conducted in 1999 through telephone interviews with ap- proximately 26,000 Canadians, aged 15 or older, living in urban and ru- ral areas across the country. The GSS excluded homeless peo We conducted our research at eight agencies serving street youth throughout the city of Toronto during fall 2001. Those eligible to partici- pate were between 15 and 24 years of age, had been homeless or without shelter during the previous year, and had demonstrated street involve- ment. Respondents who had been homeless for less than 30 days were excluded from analysis. ple from the sample, since they cannot easily be contacted by telephone. Results Service providers estimate that on any given night, the population of homeless and under-housed youth in Toronto ranges between 1,200 and 1,700. Our sample was drawn from the street youth population living in shelters, visiting drop-ins and health services, and living on the streets in the fall of 2001. The average age of young people in our sample was 20.1 years; the mean age at leaving home was 16. Most research on street youth ? whether conducted in Canada or elsewhere ? suggests that certain key demographic features of this group distinguish it from the mainstream youth population. For in- stance, men typically outnumber women, often by a 2:1 ratio (Dematteo et al., 1999; Hagan & McCarthy, 1997). In this survey, 58.6 percent of the respondents were male, 38.7 percent were female, and 2.7 percent were transgendered. (Because the transgendered sample is so small (n = 5), these respondents have been excluded from analysis.) The street youth population is also characterized by the overrepresentation of lesbian, gay, bisexual, and transgendered youth (O?Brien et al., 1994). Of our sample, 29.6 percent defined themselves as ?non-straight?; 5 percent of these were lesbian or gay, and an additional 24.6 percent reported they were ?bisexual,? ?bi-curious,? or ?not sure.? The vast majority (71.4 percent) of street youth in our sample were born in Canada, and more than half were from Toronto; 29.5 percent de- scribed themselves as ?visible minorities,? although the sample as a whole does not demonstrate the range of diverse ethnic origins found in the Criminal victimization broader population of Toronto youth. Aboriginal youth (9.1 percent) and African-Canadian youth (17.7 percent) are overrepresented within the street youth population, while South Asian and East Asian youth are underrepresented. The GSS reports that approximately 25 percent of Canadians are victims of crime in any given year (Statistics Canada, 1999), a figure that has re- mained relatively unchanged over the past decade. Generally, half of these incidents of victimization involve personal crimes (assault, rob- bery, sexual assault, theft) and about 35 percent involve household crimes ? break and enter, motor vehicle/parts theft, theft of household property, and vandalism (Besserer & Trainor, 2000, p. 4). Young people aged 15 to 24 typically report higher levels of victimization (39.7 percent) than do adults and the elderly; 18 percent having been victimized on more than one occasion. In our survey, 81.9 percent of the street youth sampled reported having been victims of crime in the past year, while 79.4 percent re- ported two or more incidents. The vast majority of offences against street youth were personal crimes, since most household crimes (motor vehicle offences, theft of household property) are less likely to be experienced by homeless people with unstable housing and limited property. Table 1 compares the rate of criminal victimization among street youth with that of domiciled youth aged 15 to 24; the latter statistics are drawn from the GSS (Statistics Canada, 1999). The categories and de- scriptions of offences are based on Canadian Criminal Code definitions. In virtually every category, the percentage of street youth who have ex- perienced some form of personal crime is significantly greater than that of domiciled 15- to 24-year-olds in the general population, with respect to both property crime and assault. For instance, higher percentages of street youth (both male and female) report at least one incident of theft, robbery, or vandalism in the past year than do domiciled youth. Al- though homeless people have fewer and less valuable possessions, the experience of being homeless makes them more vulnerable to property crime, since they carry their cash or property with them at all times. It is, however, the high percentage of street youth who report being victims of violent crimes (assault, robbery, sexual assault) that demon- strates most dramatically the extreme nature of their victimization. In particular, 31.9 percent of our street youth sample reported being victims of sexual assault in the past year. Though men in the general population are slightly more likely to be victims of most crimes than women (the exception being sexual assault), the reverse is the case for homeless youth. While young men who are homeless are more likely to report being victims of robbery, female street youth are overall more likely to be victims of crime and, in particular, vandalism, break and enter, and sexual assault. Domestic assault is a particular problem: 25 percent reported being victims of partner abuse in the past year. Nevertheless, as Tanner and Wortley (2002) have noted, male street youth are still much more likely to be victims of sexual as- sault than are domiciled youth either male or female. Offending behaviour Criminological research suggests a linkage between criminal offending and victimization (Lauritsen et al., 1991). Research in Canada has con- sistently shown that the street youth population is generally more likely than domiciled youth to engage in deviant and delinquent behaviours (Hagan & McCarthy, 1997; Tanner & Wortley, 2002). Table 2 shows the frequency of involvement in certain criminal activities, including ?as- sault,? ?theft? (both for personal needs and in order to sell), and ?drug dealing.? The range of delinquent and criminal offences listed here, though limited, represents indicators of degree of criminal involvement. There is clearly a great deal of variation among the population of street youth. While the overall percentage of street youth involved in crime is high, 37 percent of those in our sample reported no involvement in any of the offence categories. The question is whether those street youth who are more criminally involved are also at greater risk of be- coming victims of crime. An analysis of data determines that street youth who report no involvement in the criminal activities listed in Table 2 were only slightly less likely to report experiences of criminal victimiza- tion during the past year. The greatest differences were reported be- tween those who frequently engaged in selling drugs (85 percent were victims of crime) and those who did not deal drugs (76.5 percent). Homeless male youth are more likely than their female peers to be criminally involved, particularly in theft (for purposes of selling) and drug dealing, a difference that also reflects the gendered nature of street youth?s money-making strategies. The relationship between criminal offending and victimization is complex, for female street youth are in general more likely to be victims of crime, but less likely to be offenders than males. Involvement in deviant and delinquent behaviours thus cannot alone explain the high rates of criminal victimization that street youth experience. The complex interplay of gender, crime, and criminal victimization suggests that young women who are homeless may expe- rience social exclusion in profoundly different ways from young men. Worst victimization experience Street youth were asked to identify whom they had told about the most serious episode of criminal victimization they had experienced in the previous year (see Table 3). Although it is not surprising that street youth are most likely to report negative experiences to their friends, giv- en the profound significance of street friendships for homeless people, what is unusual is the number who say that they did not tell anyone about what happened to Street youth were asked what they considered the most serious crime committed against them in the past year. Incidents of assault (22.9 per- cent) and theft of personal belongings (21.1 percent) were mentioned most often. Women were more likely to identify sexual assault (F = 11.3 percent; M = 4.0 percent) and partner assault (F = 15.5 percent; M = 1.0 percent) as the most significant, while men were more likely to identify theft of personal belongings (M = 14.1 percent; F 8.5 percent) and fraudu- lent acts by employers (M = 14.1 percent; F = 7.0 percent). them (33.1 percent), a practice more characteris- tic of homeless men than of homeless women. According to the General Social Survey, on the other hand, only 7 percent of domiciled youth (15?24 years old) chose not to tell anyone when they were victims of crime (Besserer & Trainor, 2000, p. 9). This suggests that although street youth may emphasize the significance of ?street? friendships, often us- ing the language of ?family? to describe such relationships, they often, at the same time, are socially isolated or have weak attachments to others and do not always trust those who are close to them. Few street youth reported incidents of criminal victimization to members of their family or to adult authority figures such as teachers, social workers, or counsellors (including shelter staff), reflecting the es- trangement of young people who are homeless, their weak guardianship, and their limited social capital. Women, however, were much more like- ly to confide in adult authority figures than men. Only 12.2 percent of street youth reported their worst victimization experience to the police; and, in many of these cases, this did not neces- sarily reflect a personal decision (e.g., the police independently arrived at the scene of a crime). Many young people refrain from informing the police of criminal activities that they have experienced because they feel that the incident is minor or there is little the police can do about it (Tan- ner & Wortley, 2002). Young people may also be concerned about being perceived as ?snitches? and about retaliation by the offender. While these explanations may apply to street youth, the responses of a number of street youth reflect their profound alienation from the police, their lack of faith in them, and their desire to avoid them. One of the main reasons our respondents cited for not reporting their victimization to the police was their belief that the police would not believe them anyway (36.5 percent), a view expressed even more strong- ly by male (42.7 percent) than by female youth (21.1 percent). In addi- tion, 20.9 percent reported being unwilling to involve the police because they themselves were committing an illegal act at the time. Safety and preventive strategies The fact that young men (27.4 percent) are more likely than young women (10.5 per- cent) to give this reason, is likely related to their higher levels of criminal involvement. Table 4 shows the range of strategies street youth in our sample reported engaging in to enhance their personal safety, compared with those cited by domiciled youth in the General Social Survey. In some ways, street youth engage in strategies that are typical of adolescents in general. For instance, the most common safety strategy of both street youth and domiciled youth is to change their routines and activities and avoid certain places they consider dangerous. Smaller per- centages of street youth also reported engaging in strategies such as in- stalling new locks (or security bars), taking self-defence courses, or ob- taining a dog, all of which require an investment of resources. However, without the guardianship of parents, street youth rely on safety strategies that more directly reflect their housing instability, their street involvement, and their constant exposure to risk in public spaces. A much higher percentage of street youth (30.4 percent) compared with domiciled youth reported having had to change their residence in order to ensure safety, and 27.8 percent also reported regularly carrying wea- pons to defend themselves. In addition, 44.1 percent of street youth re- ported carrying their possessions with them at all times. This strategy has the disadvantage of restricting their mobility and movement, limit- ing their access to private services (restaurants, stores, shopping malls), and actually making them targets for robbery or assault on the streets. Discussion More than 19 percent of street youth reported altering their appear- ance in order to ?look as tough as possible? in order to ward off would- be attackers. Female youth (29.0 percent) were more than twice as likely as their male peers (12.2 percent) to deliberately adopt this strategy. Being young and homeless in Toronto means many things ?among the most significant being that one?s health and safety are jeopardized on a day-to-day basis and that this is not incidentally related to one?s expe- rience of social exclusion. Street youth are vulnerable to exploitation, whether by petty criminals, sexual predators, unscrupulous landlords or employers, or a whole range of other individuals who can wield power over them, because potential perpetrators recognize that young people who are homeless have few resources to defend themselves and little recourse to challenge them. The circumstances that produce such high levels of victimization among homeless youth cannot be explained simply in terms of these youths? previous history of criminal victimization, nor by their own de- linquent or offending behaviour. The argument here is that the vulnera- bility of street youth to crime is most acutely experienced when multiple dimensions of social exclusion intersect. The problematic backgrounds and difficult home lives of street youth can inhibit their ability to fully participate in society as teenagers and, later, as adults. Once they are on the streets, their exclusionary trajectory intensifies as their inadequate access to housing, limited educational and employment opportunities, and restricted access to public spaces increase their vulnerability to crime. For young women who are homeless, the severity of social exclu- sion and victimization is compounded. The high rate of criminal victimization experienced by street youth means that they are forced to live from day to day with the very real fear of theft and robbery, of being attacked or sexually assaulted. For some, this becomes just another hazard associated with life on the streets; for others, the trauma associated with victimization has a devastating effect and can present yet another barrier to moving successfully off the streets. Street youth adopt subsistence strategies that are quasi-legal (squeegeeing, panhandling, the sex trade) or illegal (theft, drug dealing) and expose them to a range of potentially dangerous and exploitive per- sons. An additional consequence of engaging in risky acts is that the wil- lingness of street youth to turn to police for protection is impaired. Many street youth come to depend on one another for protection. Victimization that occurs as a result of involvement in illegal or quasi-legal activities may lead young people to believe that they have no recourse to the law ? something the perpetrators of crimes against them no doubt consider. Young men and women on the streets have different experiences of homelessness and, consequently, of victimization. The streets are a gen- dered space, one that has historically been colonized and defined as a ?male? space, where particular forms of masculinity and femininity are produced and reproduced (Gardner, 1990; Hatty, 1996). Young women who are homeless face increased vulnerability to specific forms of violent crime, including sexual assault and partner assault (Browne & Bassuk, 1997; Hatty, 1996; Simons & Whitbeck, 1991). Homeless women therefore experience risk differently and adopt gendered personal safety strate- gies. Such risks may, for instance, lead them to establish partnering rela- tionships that may provide shelter and income but also, inevitably, put them at greater risk of assault and exploitation (Maher et al., 1996; Tessler et al., 2001 Street youth, then, are made vulnerable by their limited social capi- tal, their exclusion from adequate housing and employment, their com- promised physical and mental health, and their inability to provide pro- tected spaces for themselves. They are therefore at increased risk for criminal assault or robbery. Alienation, distance, and vulnerability to crime can be considered, then, as both consequences and manifestations of social exclusion. ). Conclusion Our government believes that all people in Ontario have the right to drive on the roads, walk down the street or go to public places without being or feeling intimidated. They must be able to carry out their daily activities without fear. When they are not able to do so, it is time for government to act. (Ontario Legislative Assembly, 1999) Unfortunately, one of the clearest manifestations of this social ex- clusion is the degree to which, in public policy debates concerning safe- ty, street youth and the homeless in general are cast not as real or poten- tial victims (or members of the ?public,? for that matter) but, rather, as criminal offenders. Repressive enforcement measures to contain street youth delinquency are routinely enacted in the name of community and public safety. Street youth are regularly ?moved on? from public spaces; the police are called on by politicians at various levels of government to ?crack down? on squeegeeing and panhandling; and the visible presence of street youth is depicted by the media as having a negative impact on business. This has also resulted in punitive legislation aimed at the homeless. The passage quoted above is taken from a speech by Ontario Attorney General Jim Flaherty introducing 1999?s Safe Streets Act, which essentially targeted street youth by making squeegeeing and most forms of panhandling illegal. The social exclusion of street youth puts them in the contradictory posi- tion of being at increased risk for criminal victimization, on the one hand, and the target of public efforts to control crime and deviance, on the other. One consequence is that street youth have been systematically excluded from discussions of ?community? and public safety, and, by extension, this raises questions regarding citizenship. Many other jurisdictions have passed laws crimi- nalizing homelessness (Foscarinis, 1996; Kalien, 2001; Sossin, 1996). A question to ask during public safety debates is this: To what de- gree are street youth conceptualized as part of the ?community? or as citizens, and thus worthy of public safety measures? Evidence from re- search on homeless youth suggests that much of their criminal involve- ment is a product of their experience of being homeless (Gaetz & O?Grady, 2002; Hagan & McCarthy, 1997). Tactics that intensify the ex- perience of social exclusion of street youth, such as criminalizing home- lessness, should be avoided, as their likely effect is to further marginalize this population, increasing their risk of criminal victimization and creat- ing barriers to their movement away from the streets. A more effective long-term strategy for dealing with street youth criminality should focus on addressing the issues that produce and sus- tain homelessness. Strategies that situate people who are homeless as part of the community ? as persons who share rights and privileges with other citizens ? could ameliorate some of the negative experiences of those whose lives are so profoundly characterized by the process of so- cial exclusion. Public safety strategies, whether developed by govern- ments, community groups, or the police, must thus consider the safety of all citizens, including those who are rightly or wrongly perceived to be dangerous, different, and ?outside? the definition of community, such as the homeless. Stephen Gaetz is the Associate Dean of Research and Field Development in the Faculty of Education, York University. This research was conducted with Jus- tice for Children and Youth, a legal aid clinic for young people in Toronto, Can- ada, and is part of a larger research project on the legal and justice issues of homeless youth. Funding for this project was provided by the Canadian gov- ernment?s Supporting Communities Partnership Initiative and administered by the City of Toronto. References Alder, C. (1991). Victims of violence: The case of homeless youth. Australian and New Zea land Journal of Criminology, 24(1), 1-14. Anderson, R. (1996). Homeless violence and the informal rules of street life. Journal of Social Distress and the Homeless, 5(4), 369-380. Baron, S. W. (1997). Risky lifestyles and the link between offending and vic- timization. Studies on Crime and Crime Prevention, 6(1), 53-72. Baron, S. W. (2003). Street youth violence and victimization. Trauma, Violence and Abuse, 4(1), 22-44. Baron, S. W., Forde, D. R., & Kennedy, L. W. (2001). Rough justice, street youth and violence. Journal of Interpersonal Violence, 16(7), 662-678. Baron, S. W., & Hartnagel, T. F. (1998). Street youth and criminal violence. Journal of Research in Crime and Delinquency, 35(2), 166-192. Beitchman, J. H., Zucker, K., Hood, J., daCosta, G., Akman, D., & Cassavia, E. (1992). A review of the long-term effects of child sexual abuse. Child Abuse and Neglect, 16(1), 101-118. Besserer, S. (2001). Criminal victimization: An international perspective ? re- sults of the 2000 International Crime Victimization Survey. Juristat, 22(4), 1- 18. Besserer, S., & Trainor, C. (2000). Criminal victimization in Canada, 1999. Juristat 20(10), 1-26. Browne, A., & Bassuk, S. S. (1997). Intimate violence in the lives of homeless and poor housed women: Prevalence and patterns in an ethnically diverse sam- ple. American Journal of Orthopsychiatry, 67(2), 261-278. Cohen, L. E., & Felson, M. (1979). Social change and crime rate trends: A routine activity approach. American Sociological Review, 44(4), 588-608. Cohen, L. E., Kluegel, J. R., & Land, K. C. (1981). Social inequality and predato- ry criminal victimization: An exposition and test of a formal theory. American Sociological Review, 46(5), 505-524. Daly, G. ( Dematteo, D., Major, C., Block, B., Coates, R., Fearson, M., Goldberg, E., King, S., Millson, M., O?Shaughanessy, M., & Read, S. (1999). Toronto street youth and HIV/AIDS: Prevalence, demographics, and risks. Journal of Adolescent Health, 25(5), 358-366. 1996). Homeless: Policies, strategies, and lives on the street. New York: Routledge. Esbensen, F.-A., & Huizinga, D. (1991). Juvenile victimization and delinquen- cy. Youth and Society, 23(2), 202-228. Fitzpatrick, K. M., LaGory, M. E., & Ritchey, F. J. (1999). Dangerous places: Expo- sure to violence and its mental health consequences for the homeless. Amer- ican Journal of Orthopsychiatry, 69(4), 438-447. Fleisher, M. S. (1995). Beggars and thieves. Madison: University of Wisconsin Press. Foscarinis, M. (1996). Downward spiral: Homelessness and its criminaliza- tion. Yale Law and Policy Review , 14, 1-63. Gaetz, S. (2002). Street justice: Homeless youth and access to justice. Toronto: Justice for Children and Youth. Gaetz, S., & O?Grady, B. (2002). Making money: Exploring the economy of young homeless workers. Work, Employment and Society, 16(3), 433-456. Gaetz, S., O?Grady, B., & Vaillancourt, B. (1999). Making money: The Shout Clinic Report on Homeless Youth and Employment . Toronto: Central Toronto Community Health Centres. Gardner, C. B. (1990). Safe conduct: Women, crime and self in public places. Social Problems, 37(3), 311-328. Hagan, J., & McCarthy, B. (1997). Mean streets: Youth crime and homelessness. New York: Cambridge University Press. Hagedorn, J. M. (1998). The business of drug dealing in Milwaukee. Wisconsin Policy Research Institute Report, 11(5), 1-30. Homeless Youth, Social Exclusion, and Criminal Victimization 45 I Hatty, S. E. (1996). The violence of displacement: The problematics of survival for homeless young women. Violence Against Women, 2(4), 412-428. Hindelang, M. J., Gottfredson, M. R., & Garofalo, J. (1978). Victims of personal crime: An empirical foundation for a theory of personal victimization. Cambridge, MA: Bal- linger. Inciardi, J. A., Horowitz, R., & Pottieger, A. E. (1993). Street kids, street drugs, street crime: An examination of drug use and serious delinquency in Miami. Belmont, CA: Wadsworth Publishing. Janus, M.-D., Archambault, F. X. , Brown, S. W., & Welsh, L. A. (1995). Physical abuse in Canadian runaway adolescents. Child Abuse and Neglect, 19(4), 433-447. Janus, M.-D., Burgess, A., & McCormack, A. (1987). Histories of sexual abuse in ado- lescent male runaways. Adolescence, 22(86), 405-417. Kallen, L. (2001). Invading the ?homes? of the homeless: Is existing right-of- privacy/publicity legislation adequate?. Cardozo Arts and Entertainment Law Journal, 19(3), 405-432. Kennedy, L. W., & Forde, D. R. (1990). Routine activities and crime: An analysis of victimization in Canada. Criminology, 28(1), 137-152. Kipke, M. D., Simon, T. R., Montgomery, S. B., Unger, J. B., & Iverson, E. F. (1997). Home- less youth and their exposure to and involvement in violence while living on the streets. Journal of Adolescent Health, 20(5), 360-367. Kral, A. H., Molnar, B. E., Booth, R. E., & Watters, J. K. (1997). Prevalence of sexual risk behaviour and substance use among runaway and homeless adolescents in San Francisco, Denver and New York City. International Journal of STD and AIDS, 8(2), 109-117. Kufeldt, K., & Nimmo, M. (1987). Youth on the street: Abuse and neglect in the eigh- ties. Child Abuse and Neglect, 11(4), 265-291. Lauritsen, J. L., & Quinet, K. F. D. (1995). Repeat victimization among adolescents and young adults. Journal of Quantitative Criminology, 11(2), 143-166. Lauritsen, J. L., Sampson, R. J., & Laub, J. H. (1991). The link between offending and victimization among adolescents. Criminology, 29(2), 265-291. Logan, R. (2001). Crime statistics in Canada, 2000. Juristat, 21(8), 1-21. MacDonald, N. E., Fisher, W. A., Wells, G. A. Doherty, J.-A. A., & Bowie, W. R. (1994). Canadian street youth: Correlates of sexual risk-taking activity. Pediatric Infectious Disease Journal, 13(8), 690-697. MacDonald, R. (1997). Youth, the ? underclass? and social exclusion. New York: Routledge. MacDonald, R. (1998). Youth, transitions and social exclusion: Some issues for youth research in the UK. Journal of Youth Studies , 1(2), 163-176. Maher, L., Dunlop, E., Johnson, B. D., & Hamid, A. (1996). Gender, power and al- ternative living arrangements in the inner city crack culture. Journal of Research in Crime and Delinquency, 33(2), 181-205. Madanipour, A. (1998). Social exclusion and space. In A. Madanipour, G. Cars, & J. Allen (Eds.), Social exclusion in European cities (pp. 75-94). London: Jessica Kingsley. Miethe, T. D., & Meier, R. F. (1990). Criminal opportunity and victimization rates: A structural choice theory of criminal victimization. Journal of Research in Crime and De- linquency, 27(3), 243-266. Miethe, T. D., & Meier, R. F. (1994). Crime and its social context: Toward an inte- grated theory of offenders, victims and situations. Albany: State University of New York Press. Ontario Legislative Assembly, Official Report of Debates (Hansard), 13 (2 November 1999) Hon. Jim Flaherty. O?Brien, C.-A., Travers, R., & Bell, L. (1994). No safe bed. Toronto: Central Toronto Youth Services. Patterson, G. R., Dishion, T., & Bank, L. (1984). Family interaction: A process model of deviancy training. Aggressive Behavior, 10(3), 235-267. Patterson, G. R., DeBaryshe, B. D., & Ramsay, E. (1989). A developmental perspec- tive on antisocial behavior. American Psychologist, 44(2), 329-335. Rapp-Paglicci, L. A., & Wodarski, J. S. ( Rivara, F. P., Shepherd, J. P., Farrington, D. P., Richmond, P. W., & Cannon, P. (1995). Victim as offender in youth violence. Annals of Emergency Medicine, 26(5), 609-614. 2000). Antecedent behaviors of male youth victimization: An exploratory study. Deviant Behaviour: An Interdisciplinary Journal, 21(6), 519-536. Rotheram-Borus, M. J., Mahler, K. A., Koopman, C., & Langabeer, K. (1996). Sexual abuse history and associated multiple risk behaviour in adolescent runaways. American Journal of Orthopsychiatry, 66(3), 390-400. Russell, L. A., & Robertson, M. (1998). Child maltreatment and psychological distress among urban homeless youth. New York: Garland. Sampson, R. J., & Lauritsen, J. L. (1990). Deviant lifestyles, proximity to crime and the offender-victim link in personal violence. Journal of Research in Crime and Delinquency, 27(2), 110-139. Simons, R. L., & Whitbeck, L. B. (1991). Sexual abuse as a precursor to prostitution and victimization among adolescent and adult homeless women. Journal of Family Issues 12(3), 361-379. Sossin, L. (1996). The criminalization and administration of the homeless: Notes on the possibilities and limits of bureaucratic engagement. New York Uni versity Review of Law and Social Change, 22(3), 623-700. Sparks, R. F. (1982). Research on victims of crime. Washington, DC: Government Print- ing Office. Statistics Canada (1999). General social survey: Victimization, cycle 13 . Ottawa: Author. Tanner, J., & Wortley, S. (2002). The Toronto Youth Leisure and Victimization Sur- vey: Final report. Toronto: University of Toronto. Terrell, N. E. (1997). Street life: Aggravated and sexual assaults among homeless and runaway adolescents. Youth and Society , 28(3), 267-290. Tessler, R., Rosenheck, R., & Gamache, G. (2001). Gender differences in self-reported reasons for homelessness. Journal of Social Distress and the Homeless, 10(3), 243- 254. Tyler, K. A., Hoyt, D. R., & Whitbeck, L. B. (2000). The effects of early sexual abuse on later sexual victimization among female homeless and runaway adoles- cents. Journal of Interpersonal Violence, 15(3), 235-250. Walker, A., & Walker, C. (Eds.). (1997). Britain divided: The growth of social ex- clusion in the 1980s and 1990s . London: Child Poverty Advocacy Group. Whitbeck, L. B., & Hoyt, D. R. (1999). Nowhere to grow. New York: Aldine de Gruyter. Whitbeck, L. B., Hoyt, D. R., & Ackley, K. A. (1997). Abusive family back- grounds and later victimization among runaway and homeless adoles- cents. Journal of Research on Adolescence, 7(4), 375-392. Whitbeck, L. B., Hoyt, D. R., Yoder, K. A., Cauce, A. M., & Paradise, M. (2001). Deviant behaviour and victimization among homeless and runaway adolescents. Journal of Interpersonal Violence, 16(11), 1175-1204. Whitbeck, L. B., & Simons, R. L. (1990). Life on the streets: The victimization of runaway and homeless adolescents. Youth and Society , 22(1), 108-125. Whitbeck, L. B., & Simons, R. L. (1993). A comparison of adaptive strategies and patterns of victimization among homeless adolescents and adults. Vi- olence and Victims, 8(2), 135-153. Table 1: Experiences of criminal victimization, co mparing domi- ciled youth in the general public (15 to 24) with street youth Total Male Female Total Male Female Assault (an a ttack, a face - to- face threat, or an inci- dent with a weapon) 62.3% (18 6) 61.1% (10 8) 64.4% (73 ) 12.0% (3,384 ) 15.1%44 * (1.645 ) 8.9%* 1 (1,739) Theft (theft of personal or hous e- hold pro p erty) (186) 50.3% (108) 49.3% (73) 52.8% (3,386) 10.0% (1,646) 9.8% (1,740) 10.2% Robbery (face - to- face theft in which perpetr ator uses force or threat of force) 36.1% (18 6) 45.4%* " (10 8) 23.39'0 '4 (73 ) 3.0% (3.386 ) 3.7%* (1,646 ) 2.3%* (1,740) Sexual assault (forced sexual activity; an a t- tempt at forced sexual activity, or unwanted sexual touching. gra b- bing, kissing, or fo ndling 31.9% (18 6) 18.9%* "" ` (10 6) 51.4%* + * (72 ) 3.8% (3,384 ) 0.9%* ** ` (1,645 ) 6.6%*** (1,739) Vandalism ( willful damage of pe r- sonal pro p erty) (186) 30.4% (108) 25.9% (73) 35.6% (3.386) 5.7% (1.646) 6.3% (1,740) 5.2% Break and enter (ille gal entry of household pro p er- ty) (186) 15.5% (108) 12.0% (73) 19.2% (3,386) 4.6% ( I ,646) 4.3% (1,740) 4.9% TOTAL re porting at least one crime incident (186) 81.9% (108) 76.6%* (73) 91.5%* (3,421) 39.7% (1.660) 42.1% (1,761) 37.5% X2 Statistics relating to domiciled youth are derived from Statistics Canada?s General Social Survey (1999). (significance of gender): * p < 0.05: ** p < 0.01; *** p < 0.001 Table 2: Street youth involvement in delinquent and criminal activities (frequency = 3 or more times in the past 12 months) Frequency In the past 12 months, have you engaged in any of the following? Total Male Female Assault (for reasons other than self- defence) (n: male = 104; female = 69; total = 173) Never 58.1% 51.9% 66.7% Once 10.5% 11.5% 10.1% More than once 31.4% 36.5% 23.2% Shoplifting (stealing something for your own use) (n: male = 107; female = 68; total = 175) Never 47.1% 42.1% 54.4% Once 13.8% 12.1% 16.2% More than once 39.1% 45.8% 29.4% Theft (stealing goods for the purpose of sell- ing them) (n: male = 106; female = 69; total = 175)* Never 62.6% 52.8% 76.8% Once 9.2% 9.4% 10.1% More than once 28.2% 37.7% 13.0% Drug dealing (n: male = 101; female = 68; total = 170)* Never 50.0% 40.6% 61.8% Once 8.3% 9.9% 7.4% More than once 41.7% 49.5% X 30.9% 2 (significance of gender): * p < 0.05 Table 3: Street youth reporting of criminal victimization W h o did you tell about the incident? Total (n) Male (n) Female (n) I didn?t tell anyone (151) 33.1% (90) 43.3%** (61) 18.0%** I told a friend 41.7% (151) (90) 37.8% (61) 47.5% I told my partner (boyfriend, girlfriend. etc.) (151) 17.2% (90) 7 . 8 % (61) I told a social worker, teacher. or counsellor (151) 12.6% (90) 3.3%*** (61) 26.2%** I talked to a lawyer about it (151) 9.9% (90) 6.7% (61) 14.8% I told a member of my family (151) 15.9% (90) 7.8%** (61) 27.9%** I told the police (148) 12.2% (87) 4.6%** (61) 23.0%** X2 (significance of gender): * p < 0.05; ** p < 0.01; ***p < 0.001 Note: Some respondents gave multiple answers. Table 4: Strategies to increase safety: Comparing street youth to dom i- ciled youth (1 5 -24) 1 In order to protect yourself or your property from crime, do you or have you done any of these things in the last 12 months? Homeless youth Domiciled Youth Total Male Female Total Male Female (a) Changed your routine, activities or avoided ce rtain places? 52.9% (17 0) 47.4% (97 ) 60.6% (71 ) 24.7% (3,391 ) 31.7%** * (1.646 ) 40.3%*** (1.745) (b) Ca rried your posse ssions with you at all times? 44.1% (17 0) 41.2% (97 ) 49.3% (71) N/A N/A N/A (c) I nstalled new locks or security bars? 15.8% (16 9) 13.3% (98 ) 18.3% (71 ) 13.7% (3,390 ) 11.8%** (1,645 ) 15.4%** (1,745) (d) Taken a self- defence course? 16.0% (16 9) 16.5% (97 ) 15.7% (70 ) 9.2% (3,396 ) 7.8%* (1.648 ) 10.5%* (1,748) (e) Tried to look as tough as po ssible so people would leave you alone? 19.5% (16 9) 12.2%* (98 ) 29.0%* (69) N/A N/A N/A (f) Changed your phone number? 12.9% (17 0) 8.2% * (97 ) 19.7%* (71 ) 3.2% (3,396 ) 3.6%* * (1, 648 ) 4.7% (1,748) (g) O b tained a dog? 7.6% (17 0) 10.2% (98 ) 4.3% (70 ) 3.2% (3,395 ) 2.6%* (1,647 ) 4.0%* (1,748) (h) Carried a we apon reg ular- ly? 27.8% (16 9) 27.8% (97 ) 27.1% (70 ) 16.5% (3,396 ) 10.4%* (1,648 ) 22.2%*** (1,748) (1) Changed residence or moved? 30.4% (17 1) 26.5% (98 7) 36.6% (71 ) 2.5% (3,396 ) 1.4%* ** (1,647 ) 3.5%*** (1,749) X2 1 Questions (b) and (e). not included in the original General Social Survey, wer e added to the survey of street youth to reflect strategies employed by homeless people. (significance of gender): * p < 0.05 ; p < 0.01; p < 0.00 1 Statistics relating to domiciled youth are derived from Statistics Canada's General Social Survey (199 9). Chapter 3.3 Social Housing Policy for Homeless Canadian Youth SHIRLEY B.Y. CHAU AND MIKE GAWLIUK The responsibility to provide social care outside the family is assigned to the child welfare system. However, young people between the ages of 16 to 24 are in a ?twilight zone,? in which they can receive only basic social care, including housing, if they are willing to put themselves in the care of the child welfare system. Yet there is abundant evidence that many young people are homeless because neither the family home or the child welfare system was able to provide a safe and adequate environment (Janus et al., 1987; Mathews, 1989; Powers et al., 1990; Simons & Whit- beck, 1991; Dadds et al., 1993; Fitzgerald, 1995; Hagan & McCarthy, 1997; Whitbeck et al., 1997; McCaskill et al., 1998; Ringwalt et al., 1998; Gaetz et al., 1999; Hoyt et al., 1999; MacLean et al., 1999; Ryan et al., 2000; Hyde, 2005). The Effects of Not Having Safe, Stable Housing Being homeless at this critical life stage of development increases young peoples? risk of negative health trajectories, and adversely affects their developmental path into adulthood. As well, homelessness may prevent them from completing their education, and thereby limit their ability to secure stable employment. Stress may compound their risks of develop- ing mental health problems, such as depression and anxiety, and may lead to the use of substances for relief. In turn, dependence on substances aggravates other risks to their physical health and overall well-being. Without access to affordable, safe, and stable housing, these individuals are faced with an endless array of complications. When young people are homeless, the stress of finding a safe shelter to sleep and rest every night is overwhelming. Many end up living on the streets and in places and spaces not designed or fit for rest (Deisher et al., 1992; Rew, 1996; Ensign & Santelli, 1997; Unger et al., 1998; Klein et al., 2000). Prolonged exposure to poor and unsanitary living conditions, such as sleeping on the streets and exposure to the elements wears down the health of young people. The social conditions in which homeless young people find them- selves add to their risks of declining health because of a lack of access or poor access to nutritious food (Dachner & Tarasuk, 2002; Tarasuk et al., forthcoming), exposure to diseases and pestilence such as fleas, lice, or bedbugs (Hwang et al. 2003, 2005), physical and sexual violence (Whit- beck & Simons, 1990; McCarthy & Hagan, 1992; Hagan & McCarthy, 1997; Kipke et al., 1997; Whitbeck et al., 2001; McCarthy et al., 2002; Stewart et al., 2004; Whitbeck et al., 2004), and the chronic experience of social stigma and alienation in the communities where they reside. This is not an exhaustive list of the conditions and outcomes that re- searchers and service providers have documented in their work with homeless young people; it is only a beginning, considering these indi- viduals experience health disparities early in life that will negatively af- fect their adult life with the added risk of premature mortality (Hwang, 2000; Cheung & Hwang, 2004). Considering that the numbers of young people who are homeless in Canada are rising each year and the social and individual costs are increasing as well, the solution is clear: housing for young people in this circumstance is badly needed. Housing Policy and Child Welfare in Canada Young people under the age of majority, usually age 18 or 19 (depending on the province), are considered ?minors,? and are therefore under the care and responsibility of their parents or a legal guardian, which may be the child welfare system. This means that minors are socially constructed in the category of ?children,? unable to make decisions about what is good or bad for themselves, by themselves. These children therefore lack access to essentials on their own because they are ?owned? and cared for by their caregivers; and their caregivers have legal responsibility over minors until they reach the age of majority. Child protection is the responsibility of provincial governments; each province operates slightly differently regarding how teenagers are handled by the child welfare system. When a family breakdown involv- ing minors occurs, the provincial agent responsible for protecting the rights of any minors is the child welfare system. In practice, however, minors who enter the child welfare system at age 15 or 16 are not consid- ered as vulnerable and as much in need of care as young children. At this age, young people are often discouraged from entering the child welfare system, and encouraged to return home through the principle of family preservation. Entry into the child welfare system apparently occurs only when individuals can demonstrate that they are in critical need of care and protection by the state. Often, young people at this age receive limited assistance from the system and are forced to choose between returning to their family environment or fending for themselves. Homelessness among youth is linked to chronic family conflict and involvement in the child welfare system during childhood. Young people become homeless as a result of ?aging out of care? ? that is, leaving government care with- out appropriate skills and provisions to ensure they are in stable, living circumstances with sufficient supports ? or by running away from the child welfare system. When young people between 12 and 19 are not cared for by the child welfare system or are unwilling to stay in the child welfare system, they turn to the streets. For many young people, returning home is not an option for numerous reasons (e.g., emotional abuse, sexual abuse, neglect, family conflict, etc.). Homelessness at this life stage is com- pounded by the fact that other systems, such as provincial housing sys- tems, do not recognize the plight of youth who become homeless as their responsibility, but as that of the child welfare system. The stories in this chapter illustrate the complexity of the systems with which youth become involved. They demonstrate how circum- stances, social policies, and discrimination towards youth in the housing market exacerbate the problem and plight of youth homelessness. The names in the case studies are not real, but the situations are. John?s Story John, now 20, lives in the Central Okanagan area of British Columbia. His parents are divorced. His father runs a successful business. His mother has a history of mental health concerns and drug abuse. John?s father is remarried, but John does not get along with his stepmother. John turned to drugs as a teenager to escape the problems at home. He developed symptoms of mental illness, including hallucinations, de- lusions, and intense paranoia. His behaviour at home frightened his fa- ther and stepmother. His father decided that he could no longer keep John at home and turned to the mental health and child welfare system for help. Mental health professionals, including a child psychiatrist, be- lieved John was experiencing a psychotic break and he was placed on antipsychotic medication. Child welfare was hesitant to respond, but after much advocacy on the part of John?s mental health team, he was taken into the care of the government. When John was 17, he was placed in a group home for adolescents in the care of the child welfare system. He was housed with younger adolescents. He took his prescribed medication, but continued to use drugs, and did not participate in the behavioural program offered. Be- fore long, he was discharged after physical assaulting a staff member, and consequently developed a reputation within the system as being a safety risk. Foster care was not an option, and there were no housing options available to address his mental health problems. At 18, John found himself at a local emergency shelter for youth. The professional supports in John?s life focused on helping him find rental housing and making the transition to adult services, as he would soon be 19 ? too old for the child welfare system. John found housing in the form of a travel trailer located next to a house. It was substandard, but the only option available to him, so his social worker reluctantly ap- proved. Within a month, John was evicted and returned to the shelter. In total, John spent the better part of six months at the shelter. The transition to adult services did not go smoothly. John would not attend scheduled appointments with his adult mental health worker and was labelled as difficult. As his mental illness was attributed to his drug use, the adult system blamed John for his circumstances. He was treated as less deserving of service than others who came into the sys- tem. Nevertheless, he continued to go with his social worker every other week to see his doctor and get his medication. John?s 19th birthday was on the horizon and his professional sup- port team turned to his father one more time to see if a return home was an option. John?s father was not prepared to take him back into the fam- ily home, but John did do some work for his father from time to time. Staff continued to try and find housing for John. John?s father agreed to contribute to John?s rent so that he could find something better than an- other trailer. John found a suite in a local hotel that was prepared to rent on a month-to-month basis during the off-season. John moved into the hotel shortly after his 19th Within a few weeks of moving into the hotel, John took in a fellow 19-year-old who was struggling with a cocaine addiction. The hotel suite slowly turned into a drug house. John, who had used marijuana regu- larly, started to use crack cocaine. The police came to the hotel on more than one occasion. John was eventually evicted, as the situation did not change. He found himself on the streets with nowhere to stay. He is now 20, and has been in and out of adult shelters, stayed with friends, and been on the streets. His future is uncertain. birthday. Youth services could no longer serve John. The adult mental health system was made aware of his living situa- tion, but did not provide any services to him. Child Welfare System: The Reality In British Columbia, the child welfare system is responsible for serving children and youth in need of protection, outlined by the Child, Family and Community Services Act. While the Act outlines the criteria that de- termines when a child is in need of protection, the systemic responses do not always protect those whom the system is designed to serve. The child welfare system serves young people until they reach the age of ma- jority (19) in British Columbia. For young people who come into the care of the government, the Ministry of Children and Family Development (MCFD) is charged with providing a place for them to live. The options include: ? foster homes; ? group homes; ? independent living suites; ? emergency shelters and transition houses; ? safe houses; ? financial support to find housing in the private rental stock. Housing options available to children and youth look very different dependant on the area of the province where a young person lives. John accessed the government system of care after advocacy on the part of a child psychiatrist. He was fortunate to have access to residential services, but when his behaviour and mental health issues became prob- lematic, his housing options were limited. That John was able to find housing was unusual, as residential ser- vices are often hard to come by, especially for teenagers. The few foster parents available are seldom interested in taking in teens, since they do not have the skills or abilities to handle the behaviours that someone like John presents. Group homes can support high-risk young people, but they are expensive and there are fewer beds available than the number of young people who require them. Foster homes are at a premium, with foster parent recruitment and retention an issue. Priority for residential care is given to young people in the continuing care of MCFD, especially those under 12. However, life in care is not always what it is cut out to be. Place- ment breakdowns, moves to different homes, and frequent upheaval is sometimes common. Young people in the child welfare system have been known to have as many as 30 moves into and out of foster homes and other placements. Some young people successfully make the transi- tion to a life of independence, but those with the greatest challenges are often the ones that the system cannot support. The residential resources available to young people cannot address the problems such as addic- tions, mental ill-health, learning disabilities, and other challenging be- haviours. While John was able to access care in the child welfare system, teens who are having problems at home and may be in need of protection do not always get the help they are looking for and have limited options. Only those who face the most extreme situations within their family home are considered for care. Parent-Teen Conflict or Family Abuse? Sarah?s Story The line between parent-teen conflict and physical abuse where a young person is deemed as in need of protection can be very blurry, as the fol- lowing example illustrates. Sarah was 16 when she was referred to a lo- cal community service agency after running away from home. By all ac- counts, Sarah had a functional family, with two loving parents and siblings. She attended school regularly and had excellent grades. When Sarah was contacted by a local outreach worker, she disclosed that leav- ing home was the result of the ongoing physical abuse by her father, which had come to a head when her father found out that Sarah was dat- ing a boy at school. Her father limited her contact with outside world, and as a function of the family?s religious affiliation, she was only al- lowed to associate those within the religion. The outreach worker re- ferred Sarah to a mental health professional because she was having sui- cidal thoughts. The physical abuse was reported to child protection authorities, who commenced an investigation. Sarah?s father admitted to the abuse and promised the child protection social worker that he would not abuse his daughter again. MCFD considered the situation one of parent-teen conflict, closing the investigation with no further action. Sarah?s father contacted the outreach worker who had been involved with his daughter and explained that such services were no longer required, and that if contact continued with the worker, he would proceed with legal action against the worker and organization. Sarah was told by the child protec- tion social worker to go home, since there was nothing further that the child welfare system could offer her. She had no money and no other options, so she reluctantly returned home. At this point, her father withdrew Sarah from school, to prevent her from having contact with the outside world. He kept the promise made to MCFD and did not physically abuse his daughter again. Instead, he chose to have Sarah?s siblings lock her in a closet and physically assault her. She fled for the second time and contacted the mental health profes- sional whom she had been introduced to earlier by the outreach worker and through this contact was able to find a place to live and receive gov- ernmental financial assistance. Sarah had a friend from school whose family offered to house her. Like Sarah, youth who find themselves in an unsafe family situation and turn to the system may not always find the response they are hoping for. The child welfare system must weigh the stories of both the parents and the young person, and the young person may have often has less credibility than the parent. Teenagers are considered capable of protect- ing themselves in an unsafe family situation and told to contact the RCMP should they be subjected to abuse at home. As a result, young people who may require protective intervention are not taken into gov- ernment care and become alienated from the very system designed to help them. Youth come to learn that discussing what is really happening at home might place them at further risk if they are made to return home. The next time something happens, they are less likely to seek out formal help, placing them at greater risk for homelessness. Youth Agreements: A Non- Care Option for High-Risk Youth To address the needs of high risk young people, who are seldom taken into MCFD care, the government has created a program specifically for high-risk young people 16 to 18 years of age. Known as Youth Agree- ments, the program involves: Legal agreements between youth and MCFD is available to youth ages 16 to 18 who cannot return home to their family for reasons of safety, and youth who have no parent or guardian willing to take responsibility for them. It provides financial assistance for youth to live independently, as well as structure and access to services and support to help them gain in- dependence and self-confidence, develop life skills, return to school and/or gain work experience, and deal with concerns such as mental health and addiction issues. Aims to protect the rights of youth to be healthy and independent, to receive guidance and support, and to be pro- tected from abuse, neglect, and harm. Program is accessed through local MCFD offices. (www2.vpl.vancouver.bc.ca/DBs/Redbook/orgPgs/1/12939.html) As with the formal system of care, the use of youth agreements varies based on the region of the province young people live. Some communities have the resources and make extensive use of the program whereas other communities do not. How the program is delivered also looks different, depending on where in B.C. the young person lives. Youth Agreements are a cost-effective way to support high-risk and homeless young people and can prevent them from coming into gov- ernment care. However, young people can be punished for the very things that brought them to ask for help in the first place. In practical terms, young people placed on a youth agreement must meet certain high-risk criteria in order to be accepted into the program. These include: ? unable to live with family; ? street involved or homeless; ? have untreated addictions or mental health concerns; ? not involved in a day program such as school or employment; ? may be involved in or at high risk of sexual exploitation. Youth first attend a triage meeting and are told whether they are a candidate for the program. If accepted, they must complete a three-week trial period where they are asked to work on the issues that brought them to the program in the first place. Once the youth agreement is signed, the young person agrees to follow through with specific expecta- tions and in exchange, the government provides financial and relational support to live independently in the community. The youth agreement is reviewed on a regular basis, and if the young person is not following the terms of the agreement, he or she is given a warning and can be dropped from the program. Although the government is responsible for young people in their care until the age of 19, and therefore has a legal responsibility to provide them with housing and supports, regardless of their situation, a youth agreement is different as the young person is not in the care of MCFD. In many cases, a Youth Agreement works well. Since youth agree- ments provide financial support (up to $450 a month for housing) for young people to live independently, they must find housing in the community. In a tight rental market, however, it can be very difficult to find a place. Youth Agreements are a cost-effective way to support high- risk and homeless young people and can prevent high-risk youth from coming into care. But within the program, young people can be punished for the very things that brought them to ask for help in the first place. As sole responsibility for follow-through rests with young people them- selves, they have limited options if they are dropped from the program. Although the government is responsible for young people in their care until the age of 19, and therefore has a legal responsibility to provide them with housing and supports, regardless of their situation, a youth agreement is different. In some cases, young people who are homeless and the highest risk youth in the community simply are often unable to secure a youth agreement. If they do manage to enter the program, finding a place to rent and achieving stability can be is difficult, if not impossible. To ex- pect a young person to make a go of it on $675 a month in a tight hou s- ing market, when they lack the necessary knowledge and abilities, sets them up for certain failure. While Youth Agreements fill a gap for high- risk youth between 16 and 18, the reality that teens are not taken into government care has created a new gap in service. There are limited op- tions There is nothing available for those between 13 and 16 who cannot or will not stay at home. Too Old for Care, Too Young for a Youth Agreement: Shannon?s Story Shannon, 15, has a history of conflict with her parents. She has been in and out of the family home numerous times since she was 13. Her mother finally told Shannon that she is no longer welcome at home. Shannon turned to her outreach worker for support. Shannon had no extended family she could live with, so she turned to MCFD. Shannon?s mother told the social worker that she could come home, but she would have to follow a long list of expectations in order to return. The outreach worker knew that she had no intention of taking Shannon back and made unreasonable requests to make sure that this did not happen. But the social worker told Shannon there was nothing she could do, since Shannon?s mother had made the offer, and there are no foster homes available for 15 year olds. Shannon was told that she could apply for a Youth Agreement in 10 months? time, when she would be 16, but until then, she was on her own. Shannon stayed with a friend at first, and over the course of the next several months, moved from friend to shelter to the street. She began to use drugs and as a result, her schoolwork suffered. She was expelled for lack of attendance. Shannon took a part-time job, but could not hold it for long ? she had trouble getting to and from work, because she never knew where she would be sleeping from night to night. She became de- pressed and felt suicidal. Shannon became involved with a 25-year-old man who provided her with a place to stay. She became pregnant, but miscarried. The relationship with her boyfriend broke down. She went back three times before finally deciding to leave him. She was willing to see a mental health worker, who helped her. Shannon?s 16th birthday is approaching. In the last year, she has lived in 15 different places, has dropped out of school, and has experi- enced a miscarriage. With her outreach worker, Shannon approaches MCFD about a Youth Agreement. She does what is asked of her and signs on to the program. It takes her two months to find a place to rent, and she has to move twice after entering the program. Paradoxically, Shannon would have been eligible for the program at 15 if she had been pregnant or parenting. In the absence of other options, she was left to fend for herself and she experienced a year of instability. Youth Aging Out of the System: Falling Through the Cracks While Shannon was too young to receive services, John was too old. In British Columbia, services are terminated for young people in govern- ment care upon their 19th birthday, whether they are ready to live on their own or not. The youth services system operates quite differently from the adult system. Young people may experience culture shock when they move into a different system of care. In John?s case, he was at a disadvantage to begin with, as he had a reputation for being difficult, which followed him into the adult system. John was also seen as causing his own mental health problems. If John had been seen as ?ready? to receive help, the adult system could give him the limited help that was available to him, provided he did what he was asked to do. Unfortunately, John was not able to do this, so he was labelled as ?resistant? to service. Many young people today live with their families until they are well into their twenties. They may not possess either the financial capac- ity or the life skills to live on their own. Those who do move out can turn to their families for material, financial, or emotional support. By contrast, for young people in government care, 19 is the age at which they are expected to go it alone. Whether they can take care of themselves is not a consideration. Yet addictions, mental health issues, street involvement, a traumatic history, and other circumstances can im- pede their developmental trajectory. It is no surprise then, that the litera- ture shows such poor outcomes for young people exiting care. Fortunately, as this article was being written, the provincial gov- ernment announced the implementation of Agreements for Young Adults, whereby those that turn 19 years of age and in the care of the government or on a youth agreement are eligible to receive financial support for up to 24 months to enrol in schooling or rehabilitative pro- gramming. A Lack of Safe and Affordable Housing for Youth: Brad?s Story John?s story demonstrated some of the barriers facing all young people when they move out on their own. The Central Okanagan has one of the hottest real estate markets in Canada. With the average home costing about $500,000, few young families can afford to purchase a home. This increases the number of people looking for rental accommodation, thereby pushing down vacancy rates and raising rents. Since 2006, the vacancy rate in the area has been between 0 and 0.6%, and a verage rent for a one-bedroom apartment is $800 a month. By contrast, young people on income assistance receive $375 a month for rent and those on youth agreements receive $450 a month. With such a limited supply of rental housing, landlords in the community can pick and choose their tenants. It is not uncommon for a landlord to hang up the phone upon hearing that the applicant is 17 or 18. For someone like John, finding housing is next to impossible. He has never rented before, has no housing references, and has limited life skills and an active addiction. His options include small travel trailers with insufficient amenities or hotels that rent to marginalized young people in the off-season. These hotels charge high rates and the tenant must move out in the spring when tourists come to town. There is very little low- income housing in the area, and wait lists are up to five years for the few units that do exist. Priority is given to single-parent families, seniors, and low-income working singles. Brad, 18, has a history of crystal meth use. He was kicked out of his mother?s home and lived in a local emergency shelter for several months. He struggled with his addiction, and while he did get addictions coun- selling, he did not follow through with a plan of care. However, he was determined to find work and housing. He found work, as he has a back- ground in the food service industry. Working a 40-hour week, he made 10 dollars per hour plus some tips. His take-home pay is roughly $1,300 a month. While he stayed at the shelter, he was able to save $1,600 to put towards a place to rent. Brad took the better part of three months to find rental housing. Finding a landlord that would consider renting to him was a challenge. When he approached a potential landlord with his housing support worker, Brad found someone willing to take him on. The challenge: Brad would have to pay $1,000 a month for rent, or 80 percent of his take- home pay. He also has to pay for a cellular phone and transportation to and from work. He has limited income to purchase food and his apart- ment is bare, with little furniture and few household items. If his work hours were reduced, he could find himself homeless. He cannot save any money and lives from cheque to cheque. He uses the food bank when necessary, and brings food home from work. His apartment is far from the downtown core, and as a result, Brad has reduced his drug use con- siderably. By all accounts, he is doing extremely well, but he would be defined as ?precariously? housed. One bad choice or unfortunate cir- cumstance would put him back where he started. What Happens When a Young Person Burns B ridges and Has Nowhere to Turn: Lisa?s Story Lisa, 22, has been on and off of the streets since she was 12. Her mother chose a never-ending procession of boyfriends over Lisa, and there was a family history of drug and alcohol abuse and mental illness. Lisa left home and came to the attention of child welfare, but her mother told so- cial workers that Lisa was welcome home, so she could get no assistance. When she did attempt to return, one of her mother?s boyfriends at- tempted to have sex with her. When Lisa told her mother about it, Lisa?s mother did not believe her. At that point, Lisa left home for the last time. Lisa quickly fell in with the wrong crowd. She experimented with alcohol and marijuana at an early age, and an older man introduced her to cocaine when she was 13. It was not long before Lisa was standing on a street corner, and at the age of 13, selling herself to feed her addiction. Lisa stayed in a number of different places, including the homes of adult sex trade workers and known drug houses. Lisa began injecting drugs when she was 15. She has been sexually assaulted on several occasions, has had to stab johns who tried to assault her, and has collected debts for drug dealers. She has had a series of ?boyfriends? who have physically abused her and lived off her earnings. At 15, Lisa came into government care. She lived in hotels and re- connected with a former foster parent. Lisa stayed with this person for over a year, taking prescribed medication for a diagnosed mental illness and returning to school. She said that she wanted to finish school and become a youth worker, so that she could help young people who have been in the same situation as her. Lisa?s situation took a turn for the worse when her foster parent found drugs in her possession and needles in the home. With younger children in the home, she was unable to keep Lisa. Lisa quit school and found herself downtown once again using drugs and working the street. She made several suicide attempts, and overdosed on more than one oc- casion. She has been in and out of detox several times and has stayed at shelters time and again, only to return to the streets. At 19, Lisa aged out of the child welfare system. Her life has contin- ued down the same path, with one exception, a period of months during which she found stable employment in a restaurant and left the streets. But she was let go when the tourist season ended and found herself back on the streets, using drugs heavily. She was introduced to the criminal justice system after being charged for assault and prostitution, and has been in and out of jail. Lisa remains in contact with an outreach worker she met when she was 14, and still wants to finish school and become a youth worker. She struggles with her addiction and mental health and has not had stable housing for several months. Successful Community Responses to Homeless Young People: Jake?s and Jesse?s Stories While the stories told so far illustrate the systemic, community, and in- dividual challenges that homeless youth face, some have successfully made the transition from the streets to longer-term housing. These sto- ries identify some of the strategies that have been effective in supporting high-risk and homeless young people. The young people involved had social supports and were open to working on their situations. Jake, 17, has a history of family abuse. Raised by his father, Jake was severely beaten throughout his childhood. At 12, Jake was brought to the attention of child protection authorities by a concerned school counsel- lor. Jake had symptoms of attention-deficit disorder and had difficulty managing in school, getting into fights with his classmates. An investiga- tion revealed that Jake was being mistreated at home. Jake was taken into government care and experienced foster care, group care, and independent living. Jake did not manage well in group care, where he faced conflict with staff and youth. Each care option he tried ended with his being discharged due to an inability to follow pro- gram expectations. He was not welcome at his social worker?s office, where he acted out, and he could only attend in the company of a youth worker. Some organizations refused to work with Jake, saying he pre- sented a safety risk to staff. He found himself in the shelter system. His social worker sought the assistance of a community organiza- tion to provide Jake with intensive support, so that he could find hous- ing in the community. The support worker found a hotel room for Jake to rent in the off-season and attempted to build a working relationship with him. Initially, problems arose when Jake invited his friends to his room. However, the support worker, Jake, and the landlord were able to work on these issues and Jake became a good tenant. When the hotel was nearing busy season, the landlord offered him an extra month to stay and offered Jake a reference for his future housing searches. The support worker, focusing on Jake?s strengths, including his in- dependence and his desire to be free of the child protection system and to find suitable employment, developed a positive working relationship with Jake. He was referred to a local employment program, and was soon successful in finding employment. When he was laid off during the off-season, he was able to find employment again. Through this full-time job, Jake made friends who offered him a place to stay when their roommate moved out. Jake moved in with the group and remained there for close to a year. Jake soon found himself facing eviction when the owner of the home sold the house. He was extremely discouraged and ended up quit- ting his job. He ended up back in the shelter system, but still wanted to achieve stability. A pilot project had just started, in which Jake could en- ter a transitional housing suite temporarily. He did so and remained in the program for several months. Initially, he was used marijuana and was not motivated to find work, but eventually found a full-time job and registered for school. He has been a good neighbour in the program and has recently met a new roommate. They are getting along well and have talked about moving out together when they are ready. Jake is not out of the woods yet, but his resilience and determination have shown through over the past few years. Jesse, 19, comes from a family with significant mental health issues. He did not enter government care, because he found himself in a bail hostel after being charged with a crime. He was 15 at the time, and al- though a return home was not an option, MCFD was not prepared to take him into care, preferring to wait until he turned 16 and to offer him a Youth Agreement. Luckily for Jesse, he developed a good connection with the bail hostel family and stayed for over a year, without breaching his bail conditions. Like Jake, Jesse had the support of a youth worker who was charged with helping him find and maintain housing, attend to his health issues, find employment and connect him with disability cover- age, so that he had financial assistance. Jesse eventually found a room for rent in the community. The path to independent living was not straight, as he struggled with alcohol abuse. This created problems with his roommates and Jesse was evicted from the home. Luckily for him, an- other residence was available and he made a transition from one home to the other. This time, he made an effort to work on the things that had caused problems in the first place. Jesse was deemed unemployable by his family physician due to his mental health issues. He did odd jobs and volunteered at a local youth centre helping out with community clean-up and graffiti eradication. Is There a Housing Policy for Young People? The Homelessness Partnering Strategy replaced the National Homeless- ness Initiative on April 1, 2007. However, it does not include a specific initiative that addresses the needs of homeless youth. Many policies cover children and adolescents as long as they remain attached to their families and remain the responsibility of their legal guardians. But once young people step outside the structures designed to keep them inside their homes, there are few courses available to them. At this time, limited money is available to address youth homeless- ness, and these funds are temporary and project-based, rather than long- term core funding. The fragmentation of social policies contribute to the creation and maintenance of youth homelessness. One solution is the implementation of social policy that leads to the unification of various systems that affect at-risk youth to reduce the gaps and develop and implement a new policy to house at-risk youth. Col- laboration between service systems and organizations is essential. The challenges in this work are many, but some things are absolutely neces- sary to help young people succeed. The following must be in place: ? A supply of affordable, safe and stable housing designated specifi- cally for the youth population ? Income supports that provide young people with the financial re- sources to pay their rent and meet basic needs ? Social supports that assist young people in developing the skills, knowledge and abilities to achieve self-sufficiency The old saying that ?it takes a village to raise a child? holds true as communities across Canada work to eliminate youth homelessness. Shirley Chau is assistant professor in the School of Social Work at the Univer- sity of British Columbia. Her interests and experience are in the areas of work and wellness, crisis intervention, psychological trauma, and individual counsel- ling. Mike Gawliuk is Area Director of Youth and Family Services for the Central Okanagan Boys and Girls Clubs References Cheung, A. M., & Hwang, S. W. (2004). Risk of death among homeless women: A cohort study and review of the literature. Canadian Medical Association Jour- nal, 170(8), 1243-1247. Dachner, N., & Tarasuk, V. (2002). Homeless ?squeegee kids??: Food insecurity and daily survival. Social Science & Medicine , 54(7), 1039-1049. Dadds, M. R., Braddock, D., Cuers, S., Elliott, A., & Kelly, A. (1993). Personal and family distress in homeless adolescents. Community Mental Health Journal, 29(5), 413-422. Ensign, J., & Santelli, J. (1997). Shelter-based homeless youth: Health and access to care. Archives of Pediatrics & Adolescent Medicine, 151(8), 817-823. Farrow, J. A., Deisher, R. W., Brown, R., Kulig, J. W., & Kipke, M. D. (1992). Health and health needs of homeless and runaway youth. Journal of Adoles- cent Health, 13(8), 717-726. Fitzgerald, M. D. (1995). Homeless youth and the child welfare system: Implica- tions for policy and service. Child Welfare League of America, 74(3), 717-730. Gaetz, S., O?Grady, B., & Vaillancourt, B. (1999). Making money: The Shout Clinic report on homeless youth and employment (pp. 1-52). Toronto: Central Toronto Community Health Centres. Hagan, J., & McCarthy, B. (1997). Mean streets: Yout h crime and homelessness. To- ronto: Cambridge University Press. Hoyt, D. R., Ryan, K. D., & Cauce, A. M. (1999). Personal victimization in a high risk environment: Homeless and runaway adolescents. Research in Crime and Delinquency, 36(4), 371-391. Hwang, S. (2000). Mortality among men using homeless shelters in Toronto, On- tario. Journal of American Medical Association, 283(16), 2152-2157. Hwang, S., Martin, R. E., Tolomiczenko, G. S., & Hulchanski, J. D. (2003). The relationship between housing conditions and health status of rooming house residents in Toronto. Canadian Journal of Public Health, 94(6), 436-440. Hwang, S. W., Svoboda, T. J., De Jong, I. J., Kabasele, K. J. & Gogosis, E. (2005). Bed bug infestations in an urban environment. Emerging Infectious Diseases, 11(4), 533-538. Hyde, J. (2005). From home to street: Understanding young people?s transitions into homelessness. Journal of Adolescence, 28(2), 171-183. Janus, M.-D., Burgess, A. W., & McCormack, A. (1987). Histories of sexual abuse in adolescent male runaways. Adolescence, 22(86), 405-417. Kipke, M. D., Simon, T. R., Montgomery, S. B., Unger, J. B., & Iversen, E. F. (1997). Homeless youth and their exposure to and involvement in violence while living on the streets. Journal of Adolescent Health, 20(5), 360-367. Klein, J. D., Woods, A. H., Wilson, K. M., Prospero, M., Greene, J., & Ringwalt, C. (2000). Homeless and runaway youths? access to health care. Journal of Ado- lescent Health, 27(5), 331-339. MacLean, M. G., Embry, L. E., & Cauce, A. M. (1999). Homeless adolescents? paths to separation from family: Comparison of family characteristics, psy- chological adjustment, and victimization. Journal of Community Psychology, 27(2), 179-187. Mathews, F. (1989). Familiar strangers: A study of adolescent prostitution. (Revised ed.). Toronto: Central Toronto Youth Services. McCarthy, B., & Hagan, J. (1992). Surviving on the street: The experiences of homeless youth. Journal of Adolescent Research, 7(4), 412-430. McCarthy, B., Hagan, J., & Martin, M. J. (2002). In and out of harm?s way: Violent victimization and the social capital of fictive street families. Criminology, 40(4), 831-865. McCaskill, P. A., Toro, P. A., & Wolfe, S. M. (1998). Homeless and matched housed adolescents: A comparative study of psychopathology. Journal of Clinical Child Psychology, 27(3), 306-319. Powers, J. L., Eckenrode, J., & Jaklitsch, B. (1990). Maltreatment among runaway and homeless youth. Child Abuse & Neglect , 14(1), 87-98. Rew, L. (1996). Health risks of homeless adolescents: Implications for holistic nursing. Journal of Holistic Nursing, 14(4), 348-359. Ringwalt, C. L., Green, J. M., & Robertson, M. J. (1998). Familial backgrounds and risk behaviors of youth with throwaway experiences. Journal of Adolescence, 21(3), 241-252. Ryan, K. D., Kilmer, R. P., Cauce, A. M., Watanabe, H., & Hoyt, D. R. (2000). Psy- chological consequences of child maltreatment in homeless adolescents: Untangling the unique effects of maltreatment and family environment. Child Abuse & Neglect , 24(3), 333-352. Simons, R. L., & Whitbeck, L. B. (1991). Running away during adolescence as a precursor to adult homelessness. Social Service Review, 65(2), 225-247. Stewart, A. J., Steiman, M., Cauce, A. M., Cochran, B. N., Whitbeck, L. B., & Hoyt, D. R. (2004). Victimization and posttraumatic stress disorder among home- less adolescents. Journal of American Academy of Child & Adolescent Psychi a- try, 43(3), 325-331. Tarasuk, V., Dachner, N., & Li, J. (2005). Homeless youth in Toronto are nutri- tionally vulnerable. Journal of Nutrition, 135(8), 1926-1933. Tarasuk, V., Dachner, N., Poland, B., & Gaetz, S. (forthcoming). Food deprivation is integral to the ?hand to mouth? existence of homeless youth in To- ronto. Public Health Nutrition. Unger, J. B., Simon, T. R., Newman, T. L., Montgomery, S. B., Kipke, M. D., & Albomoz, M. (1998). Early adolescent street youth: An overlooked popula- tion with unique problems and service needs. Journal of Early Adolescence, 18(4), 325-348. Whitbeck, L. B., Chen, X., Hoyt, D. R ., Tyler, K. A., & Johnson, K. D. (2004). Men- tal disorder, subsistence strategies, and victimization among gay, lesbian, and bisexual homeless and runaway adolescents. The Journal of Sex Research, 41(4), 329-342. Whitbeck, L. B., Hoyt, D. R., & Ackley, K. A. (1997). Families of homeless and runaway adolescents: A comparison of parent/caretaker and adolescent perspectives of parenting, family violence, and adolescent conduct. Child Abuse & N eglect, 21(6), 517-528. Whitbeck, L. B., Hoyt, D. R., Yoder, K. A., Cauce, A. M., & Paradise, M. (2001). Deviant behavior and victimization among homeless and runaway adoles- cents. Journal of Interpersonal Violence, 16(11), 1175-1204. Whitbeck, L. B., & Simons, R. L. (1990). Life on the streets: The victimization of runaway and homeless adolescents. Youth & S ociety, 22(1), 108-125. Chapter 3.4 Street Survival: A Gendered Analysis of Youth Homelessness in Toronto BILL O ?G RADY AND STEPHEN GAET Z Research on homeless youth1 Most research on gender and employment, examining issues such as the earning ratio and occupational segregation, has used samples of formal labour market participants (for example, Davies et al., 1996; Hughes & Lowe, 1993; Kaufman, 2002). However, much of the ?work? that homeless youth engage in falls outside the boundaries of (or in Canada has grown over the past two decades, resulting in some important findings that address central ques- tions about the social characteristics and lifestyles of this economically and socially marginal group. In recent years, more attention has been paid to the heterogeneity of homeless people, and in particular the signi- ficance of gender (i.e., Carlen, 1996; Fitzpatrick, 2000; Novac et al., 2002; Wardhaugh, 1999). Yet, besides anecdotal and journalistic accounts, rela- tively little is systematically known about the varied subsistence strate- gies of homeless young men and women. 1 For our purposes, homeless youth (also referred to as street youth) include: ?? young people up to the age of 24 who are absolutely, periodically or temporarily without shel- ter, as well as those who are at substantial risk of being in the street in the immediate fu- ture? (Daly, 1996, p. 24). minimally, on the margins of) the formal labour market. Previous re- search has indicated that, in order to survive, homeless youth must adopt flexible and diverse money-making strategies (Gaetz & O?Grady, 2002). This ranges from working at paid jobs in the formal economy (usually on a short-term basis), to engaging in informal economic activi- ties associated with homelessness in Canada, such as begging and squeegee cleaning, sex trade work, and illegal or criminal activities (drug dealing, theft). In this chapter, we explore how gender shapes the experiences of street youth in Toronto, and in particular how street youth make mon- ey. One cannot make sense of the distinctive lives of young homeless men and women without reference to the ?streets??the range of public and semi-public spaces that homeless people frequent?as a gendered space where notions of masculinity and femininity are shaped and re- produced. An examination of how the gendered experiences of young homeless men and women shape their subsistence strategies will en- hance our understanding of how gender affects earning production for marginal groups in society, and will better inform policy-making. The goals of this work are threefold. First, our findings show that income generation does vary on the basis of gender; young women typi- cally report lower incomes and are, to some extent, involved in different economic activities. Second, these differences can be generally explained on the basis of the living conditions reported by young homeless men and women. Finally, we frame our analysis in terms of a broader discus- sion of how homelessness is gendered within the spaces and places that homeless youth inhabit. Literature Review The literature on gender and work has traditionally focused on how the participation and experiences of women in the labour force differs from those of men (for example, Marini, 1989). Despite gains in women?s em- ployment opportunities and rewards, work-related gender inequalities still exist, including the persistence of occupational gender segregation (Crompton, 1997), that is, the tendency of the majority of both men and women workings in occupations that are largely defined in terms of ?male? or ?female? jobs. One of the major effects of such segregation concerns wages. In 2000, for example, the female-male earnings ratio for all full-time work- ers in Canada was 71.7 percent; the figure has not changed substantially over the past two decades. The situation for younger workers is some- what less pronounced. For example, in 2001 in Ontario the weekly earn- ings ratio for male and women youth aged 15-24 who worked full time was 78 percent (Statistics Canada, 2002). Scholars have generally relied upon four factors to account for em- ployment segregation and women?s lower earnings: skill deficits, worker preferences, economic and organizational structure, and sex stereotyping (Kaufman, 2002). The data used to test these explanations rely on the census or population surveys. Marginal groups, like the homeless, are therefore excluded from this research because of their tenuous links with the labour market and lack of stable housing. We do know that homeless youth are not entirely excluded from the labour market. Research has demonstrated that many homeless young people report a history of for- mal labour-market participation (Baron, 2001; Ennew & Milne, 1997; Gaetz et al., 1999). However, obtaining and maintaining what are mostly low-skilled, poorly paying, service-sector jobs on the margins of a labour market and competing with youth and adults with more settled backgrounds is an extremely challenging task. Although these supply- and demand-side factors are important for understanding gendered employment differences in the formal econo- my, such an approach may not be adequate to explain the subsistence strategies of the homeless. Research on street youth and crime shows that one of the consequences of their disadvantaged backgrounds, stress- ful current life events, and labour-market marginality is the lure that is provided by money-making activities outside the formal economy (Ba- ron & Hartnagel, 1998; Carlen, 1996; Greene et al., 1999; Hagan & McCar- thy, 1991; Stephens, 2001). For homeless youth, most work takes place in informal, unregulated economic spheres, and includes begging and squeegeeing, quasi-legal activities (in the sex trade, for instance), and criminal activities. For many years these behaviours have been regarded as a commonplace means of survival for the poor in developing countries (see Stephens, 2001), but they are only beginning to be fully explored within developed countries such as Canada (Gaetz & O?Grady, 2002). Since the subculture of the ?streets? produces a context whereby the opportunities to earn money through activities such as drug dealing and prostitution are often transparent, many youths without a stable and re- liable food source who lack safe shelter are open to participation in these deviant social networks (see Stephens, 2001). Such subsistence strate- gies?whether legal, quasi-legal or illegal?generate ?cash in hand? each day, a benefit for those who must focus their efforts on meeting imme- diate needs (for food and shelter, among other things). The challenge, then, is to explore the different dimensions of work (both in the formal and informal economy) to understand the dynamics involved in the gendered nature of subsistence among this population. We contend that the culture of the streets and the ways in which masculinity and femininity are organized in this marginal arena need to be considered to appreciate the broad range and flexible nature of the economic activity associated with being homelessness. Space, Gender , and Making Money Research on the use of space by young people in general focuses, first, on conflicts that emerge from the presence and visibility of young people in public spaces (Sibley, 1995; Valentine, 1996; White, 1994;), and second, on how such spaces are implicated in adolescent identity formation (Massey, 1998; Robinson, 2000). Urban spaces may be ?colonized? by young people, who actively negotiate the meaning of such spaces among themselves, other members of the public, and authorities, while nurtur- ing and exploring individual and group identities. Space, place, and identity thus are bound in a way distinct from more structured family, community, and institutional spaces under the greater control of adults. For young homeless people, the relationship between space and identity is even more complex. In constituting identities as homeless per- sons, street youth are also engaged in negotiating space not only with members of the general public (passersby, other youth, customers) and agents of social control (the police, security guards), but also with other street youth. Here, much of the informal economic work that young people engage in?begging, squeegeeing, sex work, or dealing drugs? plays a role in helping homeless youth stake out urban space not only for economic activities, but also for recreation, eating, and sleeping. Such space is also used tactically in the negotiation of gender identities. Ever since McRobbie and Garber?s (1975) pioneering work, it has become necessary when examining youth cultural phenomenon to ac- count for the way in which roles and options in the home, school, workplace, and on the streets (the site of much youth cultural activity) are structured, organized, and experienced on the basis of gender. More recently, urban geographers have focused on how institutional spaces (i.e., schools, religious institutions, the workplace, and the community) affect the process of creating gender identities in which definitions of masculinity, femininity, and sexuality are constructed (Hanson & Pratt, 1995). For adolescents, ?the streets? constitute one such space, although one often defined more clearly in masculine terms. One consequence is that, until relatively recently, research on public youth cultures (and, indeed, much of the research on street youth) has rendered young wom- en practically invisible. Female involvement in such spaces, however, should not be considered as marginal to that of men; rather, it is structu- rally different in terms of how young women exercise independence, nurture friendships and attachments, and explore youth cultural options and economic opportunities. A central feature of the distinctiveness of homeless young women is that detachment from home and family situates women not only within a largely male-defined category of homelessness, but also outside more traditional environments for girls. The streets are a social and economic arena where men have more power and control than women. As Wardhaugh (1999) has argued, the streets are the quintessential male space; one where women, even those who are ?streetwise,? are never fully comfortable. These notions are in many ways in keeping with re- search on street gangs, and the gendered differences in criminal activity, where gender divergence is in part related to the social construction of gendered dominance and subordination (Messerschmidt, 1995). Hatty (1996) suggests that young women who are homeless (or fac- ing the prospect of homelessness) experience different opportunities and risks as a result of becoming physically and cognitively displaced into male spaces. She cites Gardner (1990) who, in writing on women, safety, and public places, suggests, ?Women regularly are judged and discrimi- nated against in such places; further, women fear physical and sexual assaults? (p. 417). Since the ?streets? have traditionally been defined as male space, the money-making opportunities available to homeless youth are likely to be structured accordingly. Talking to Homeless Youth As Canada?s largest city, Toronto has the largest numbers of homeless youth in the country. The estimates on any given night vary, and al- though no accurate census data exist, we believe the number to be about 2,000 (City of Toronto, 1999). They may be temporarily living in hostels, staying with friends, living in squats, or actually on the streets, and inva- riably it is the chronic instability?defined in terms of housing, relation- ships, income, and health?that most clearly characterizes their lives. Our study included a self-administered questionnaire and open- ended/semi-structured interviews. A total of 360 youths completed the questionnaire, and 20 also participated in tape-recorded interviews. The information collected from these interviews provided rich accounts about the challenges involved in surviving street life. All participants were given a $10 honorarium. Considering the nature of our population, selecting a statistically random sample was not possible. However, to capture a sample that we felt was representative of the Toronto homeless youth population, 360 surveys were purposively administered at six street youth serving agen- cies (n = 178) and eight youth shelters (n = 145) that were spread throughout the inner city of Toronto and in two suburbs. We also soli- cited young people for interviews on the streets to ensure that the views of those who are not connected to youth serving agencies were represented (n = 37). We used a Participatory Action Research approach in our research, which involved including those who are intended as the subject of the research in the design and implementation of the project. In this case, six Peer Outreach workers, who were all street-involved and included a cross-section of the homeless youth in the community in terms of age, length of time on the streets, gender, sexual orientation, and primary economic activity, were hired and trained to assist in administering the surveys. They helped select research sites to administer questionnaires, locate youth who did not normally use service agencies, explain the project to young people, and assist them in filling out the survey if lan- guage or literacy were issues. An overview of homeless youth The body of research on homeless youth demonstrates clearly the degree to which background variables are implicated in the pathways to home- lessness, and subsequently have an impact on the experiences of young people once on the streets. For instance, research suggests that a dispro- portionate number of street youth have experienced domestic physical, sexual and emotional abuse (Dematteo et al., 1999; Gaetz et al., 1999; Ja- nus et al., 1987; Kufeldt & Nimmo, 1987; Whitbeck & Simons, 1993). Vic- tims of sexual abuse are more likely to engage in risky sexual behaviours as adults, and to participate in the sex trade, for instance (Beitchman et al., 1992; Tyler et al., 2000; Whitbeck et al., 1997). Although we found that men outnumber women by two to one, consistent with other literature on street youth in Canada (Hagan & McCarthy, 1997), there no striking differences in our sample between men and women in terms of age (men = 21.3 years; women = 20.3 years), nationality (men = 83 percent Canadian; women = 7 9 percent Canadian), length of time on the street (men = 5.8; women = 5.3), the age when the respondent first left home (men = 15.3 years; women = 14.9 years) and levels of high school completion (men = 57 percent; women = 60 percent). However, while most young people in the sample came from family backgrounds characterized by problematic relations with parents and caregivers, young women were more disadvantaged in a number of ways. They (50 percent) were more likely than men (40 percent) to report interventions by child welfare authorities, and to have spent time in fos- ter care. In their reasons for leaving home, women were more likely to identify parental conflict (women = 71.1 percent; men = 62.4 percent), physical abuse (women = 45.3 percent; men = 27.1 percent), sexual abuse (women = 34.9 percent; men = 15.2 percent), and mental health issues (women = 26.4 percent; men = 13.1 percent) as significant, while men were more likely to identify independence-seeking variables such as ?looking for work? (men = 3 5 percent; women = 24.4 percent) as well as trouble with the law (men = 30.4 percent; women = 21.3 percent). Subsistence strategies Homeless youth face challenges in entering the formal job market, large- ly because they lack the skills and education, their health is compro- mised, and their inadequate housing makes it difficult for them to suc- ceed. However, contrary to popular perceptions that homeless youth do not want to work, our data suggest that an overwhelming majority of those youth are interested in finding paid employment (83.4 percent of men and 87.8 percent of women). Despite the challenges homeless youth face in gaining employment, an examination of the incidence of labour-market economic activity en- gaged in by Toronto homeless youth (Table 1) suggests they are not completely excluded from the regular economy. Table 1: Labour market activities by gender Men Women Currently employed 31 percent (n = 54) 27 percent (n = 24) Number of jobs during past year 2.4 (n = 129) 2.1 (n = 72) Modal weekly income $310 (n = 56) $201 (n = 26) Employed (n = 128) (n = 71) Full time 67 percent 48 percent Part time 33 percent 52 percent Paid cash in hand 33 percent (n = 51) 50 percent (n = 24) Gender does not appear to be related to current employment status, as men are only slightly more likely to report being employed (31 per- cent) than women (27 percent). The mean number of jobs held over the previous year is also similar for both groups. However, men were more likely than women to report earning higher weekly wages and to have full-time jobs, indicating in relative terms greater labour market success. Nevertheless, when exploring the work that street youth obtain in the formal economy, one must bear in mind that when they do get work, it tends to be at the margins of the formal economy, in jobs that are often informally organized (and therefore fall outside of regulated employ- ment and safety standards), provided in many cases by unscrupulous employers, who may feel little if any commitment to the young person they hire. A substantial number report being paid ?cash in hand,? a cir- cumstance that is more likely to be reported by women than men. While this form of payment means that they do not pay taxes on income, it also means that they typically are paid at rates below minimum wage, and that they are otherwise vulnerable to abuse by employers. I?ve had under-the-table [cash -in-hand] jobs. I w as promised $100 to clean some offices. I did the job, went back at the end of the week, and they gave me $50. There? s a lot of bad under-the-table stuff. (Maria, age 18) Many street youth find such jobs through temp agencies or labour exchanges. Sometimes, small business operators approach them directly on the streets and ask them to work for a day or two, for cash. In this economic context, exploitation is rife. Thus, while over one-quarter of our sample reported ?having a job,? the meaning assigned to such em- ployment should not be compared with what work means for workers who are housed and live in more stable and supportive environments. Participation in the labour market is not the only way homeless youth generate income. Some received some form of state assistance, such as general welfare payments, disability benefits, or employment insurance. The percentage of homeless youth who rely on social assis- tance as their main source of income is quite low (15 percent), consider- ing the high unemployment that characterizes this impoverished group. This reflects the barriers to obtaining?and maintaining?such benefits for people who are young, out of school and without shelter.2 While this percentage includes relatively equal numbers of men and women, one-half of the women who were claiming benefits were young mothers living with their dependent children. Pregnancy (and the risk of it) and young motherhood are of course salient features of life on the streets for young women (Greene & Ringwalt, 1998). The fact that home- less youth are more likely to engage in sex at a younger age, and with 2 In Toronto, young people under the age of 18 who are not ?legally emancipated? are not eligible to receive welfare benefits unless they are enrolled in school full time and with the permission of their parents. more partners (in many cases for subsistence reasons), puts young wom- en at greater risk (Kral et al., 1997). Although only a minority of homeless youth earn regular income through continuous participation in the labour market or through social assistance, it cannot be said that they are idle or without a ?job.? In fact, most homeless youth engage in flexible and diverse money-making strategies. In some cases, this includes legal activities that are part of the informal economy and are identified with homelessness in Canada (squeegeeing, panhandling); in other cases, this refers to activities that are quasi-legal (the sex trade) or illegal (theft of stolen goods and/or drug dealing). One advantage of such income-generating strategies for social- ly and economically marginal people is that they provide cash on a day- to-day basis, allowing young people to meet immediate needs. I pan [beg] until I get what I need and then I get out of there ? [It] d e- pends ? usually I?ll come here for breakfast and then I?ll go and pan for the day to get something to eat for dinner ? it?s usually just for food or whatever ? if my friends are going out we?ll pan to go out for a drink or whatever? (Dani, age 19) Table 2 shows the prevalence of begging and squeegee cleaning re- ported by the sample. Squeegeeing (the unsolicited act of cleaning car windshields for a donation at intersections mainly in downtown Toron- to3 Panhandling (also called ?panning? or ?begging?) is the act of ask- ing people for money in public environments, including busking, in which some sort of entertainment or service is exchanged for money. While the prevalence of panhandling, as with squeegeeing, is not greatly demarcated on the basis of gender, there are notable differences in inci- dence levels. More men (22 percent) reported squeegee cleaning on a daily basis, compared with women (13 percent). While a similar pattern emerges in the prevalence of panhandling?as approximately one-half of the overall sample reported to have panhandled at least once in the past ), was an activity engaged in by 40 percent of male and 36 percent of female respondents (?sometimes? or ?daily?). 3 When data were collected for this project squeegee cleaning was legal in Ontario. As of 31 January 2000, the Ontario Safe Streets Act outlawed squeegee cleaning and ?aggres- sive? panhandling. six months?women were more likely than men to regularly engage in this behaviour on a daily basis (women = 17 percent; men = 12 percent). Table 2: Level of participation in panhandling and squeegee cleaning by sex in past six months (percentages) Never Sometimes Daily n Squeegeeing Men 60 18 22 187 Women 64 23 13 108 Panhandling Men 48 41 12 189 Women 52 31 17 108 According to our interview data, squeegeeing and panhandling are typically engaged in by small groups of two or three (squeegeers some- times operate in larger groups). The proceeds of such economic activities are typically shared, and in many cases money is used to purchase goods that are collectively consumed (such as food). In mixed groups of pan- handlers (although this is less likely to be the case with squeegeers), women play the more active role in soliciting. Many homeless youth ar- ticulate this as a strategy to engage what they believe to be the greater sympathy the public has to the plight of homeless women. People offer more (to girls) than they do to guys ? I can make more money panhandling than any guy, because they say, ?Oh, it?s a girl? ? they don?t think I can manage. (Mandy, age 18) While female panhandlers may evoke more public sympathy than men who beg, the monetary rewards of such compassionate acts are not particularly lucrative, relative to other ways in which street youth make money. Also of importance here is the fact that it is men who are more likely than women to engage in these activities independently, in isola- tion from other homeless youth. This reflects the different risk factors that men and women on the streets face. Nevertheless, this also gives young men the advantage of retaining all the income from their work. I know a lot of girls who will go off and pan by themselves, which is stu- pid. How easy is it for someone to say, ?She?s pretty ? she?s gone ? ? you hear about it every day. (Donna, age 19) Involvement in the sex trade provides another avenue for income generation for homeless youth. The sex trade includes a broad range of activities, including street prostitution, working in strip clubs, escort ser- vices, or computer/telephone sex. Many homeless youth exchange sex for money or other goods, including food, shelter or drugs (Kral et al., 1997; Webber, 1991). In Table 3, the involvement of homeless youth in some aspects of the sex trade is explored. Overall, similar percentages of men and women were involved in the sex trade, although some interesting differences emerge relating to specific types of sex work. For instance, women are more likely to work as exotic dancers in strip clubs, in part reflecting the structure of the sex trade economy that provides more employment opportunities for wom- en in these areas. Table 3: Sex work over the past six months (percentages) Never Sometimes Daily n Had sex with someone for money Men 75 20 5 177 Women 73 22 5 107 Escort service work Men 90 7 3 173 Women 88 8 4 104 Exotic dancing Men 87 12 1 175 Women 78 16 6 105 Male and female patterns of street prostitution are quite different. Typically women?unlike men?work under the control of a pimp, re- sulting in restrictions on their personal freedom and their ability to keep the money they earn. In fact, there are few opportunities in the sex trade where women are able to operate independently. Our final category of money-making is criminal activity, which here refers to income generated largely through theft (breaking and entering, selling stolen goods) and drug dealing. Research (for example, Baron, 2001; Hagan & McCarthy, 1997) has established that homeless youth are, as a whole, more criminally involved than are domiciled youth. It is use- ful to make a distinction between those who engage in criminal beha- viour in order to survive, from those whom criminal activity is not dri- ven by deprivation. Clearly, many homeless youth will report having stolen or hustled food, clothing or other items to meet immediate needs from time to time. However, there is a segment of the homeless youth population that relies on criminal activities for subsistence purposes. Ta- ble 4 presents rates of homeless youth participation in criminal activities. Table 4: Crime over past six months (percentages) Never Sometimes Daily n Property crime (theft or break and enter) Men 55 40 5 181 Women 67 32 1 103 Selling drugs Men 44 41 15 186 Women 57 34 9 105 Two key statistics stand out. First, while crime and other forms of ?deviant? activity are not uncommon, theft and drug dealing are by no means dominant money-making activities. Second, crime is clearly de- fined as a male activity, a finding that should not come as a surprise since, along with age, gender has long been recognized as a key correlate to street crime in criminology. Moreover, men are more likely to report that they have been arrested in the past than are women. Nevertheless, there is evidence that female homeless youth are in general more crimi- nally involved than either male or female domiciled youth (Tanner & Wortley, 2002). In some cases, this means active involvement in drug dealing or other crime in a manner that is indistinguishable from young men. In other cases, women may adopt more narrowly defined ?sup- port? roles in such criminal activities, for instance as drug runners: I help my friends who are drug dealers?I get them customers and i n- stead of giving me drugs ? ?cause I don?t do them ? they give me money in exchange ? (Michelle, age 18) Earnings Overall, then, the data presented thus far reveal that, as with the formal labour market, the informal economy of homeless youth is characterized by a degree of gender-based, work-related segregation. We also found some evidence that there are different monetary awards associated with the various economic activities of homeless youth, whether legal, quasi- legal or illegal.4 As seen in Table 5, panhandlers and squeegeers (both legal at the time) earned considerably less than did youth engaged in the sex trade and crime. Men and women reported to earn identical amounts of mon- ey ($27 a day) for panhandling. However, men out earned women as squeegeers; for every dollar earned by men, women earned 75 cents. In drug dealing, men also made more money than women. The only eco- nomic activity where women reported earning made more money than men was in sex work. However, as previously noted, many women in the sex trade work under the control of pimps, while men are more likely to work independently. This difference may inflate the earnings reported by women (gross as opposed to net). Ratios were also calculated for Theft and Break and Enter. Since only two women reported to have in- comes in this category, it is not included in Table 5. 5 For the 13 men who reported money from this category, the mean was $262.00. Activity Table 5: Earnings for men and women from different activities Mean daily income ($) Men Women Earnings ratio (per- cent) Panhandling 27 (n = 31) 27 (n = 13) 100 Squeegee cleaning 75 (n = 26) 56 (n = 9) 75 Sex work 166 (n = 18) 233 (n = 20) 140 Selling drugs 407 (n = 21) 142 (n = 6) 34 4 These figures represent self-reported income and their accuracy cannot therefore be verified. It should also be noted that the number of respondents who answered this question was rather low. 5 The mean calculated for the two women was $350.00. From these data we get the impression that, with the exception of the sex trade, the economic activities that offer the most lucrative re- wards for homeless youth tend to be activities that are engaged in by men. Disadvantage, Gender and Income Generation The spaces that street youth occupy?to eat, sleep, ?hang out,? and make money?are gendered. Gender disparity is manifest across the range of activities that street youth engage in to survive, and is reflected in terms of differences in earning power and, to some extent, work-related segre- gation. With perhaps the exception of some work related to the sex trade, male homeless youth appear to be engaged in the more financially lucra- tive sectors of the street economy, such as crime. Even when homeless youth report being engaged in similar money-making activities, gender segregation is often manifest in terms of differences in roles played (the drug trade), opportunities for independent activity (panhandling and the sex trade), and control over earnings. How to account for such differences? While supply- and demand- side factors are commonly used to explain gender inequality in the for- mal labour market, a more useful starting point for accounting for eco- nomic inequality among homeless youth is to consider how gendered identities are negotiated in the streets. As space is negotiated with the public and other street youth, economic opportunities become structured in particular ways that reflect both the youth and general public?s under- standing of gender and homelessness. Street Youth Backgrounds The background variables presented earlier suggest that women, as a group, are more disadvantaged than men, and that this has an impact on their experience of homelessness. For instance, consistent with other re- search (Novac et al., 2002), women in our sample were more likely to implicate experiences of physical and sexual abuse in their reasons for leaving home. This may go some way towards explaining why there are typically one-half as many homeless young women as there are young men (Ha- gan & McCarthy, 1997). Because the streets have historically been colo- nized by (and defined as) ?male? space, it may not be as obvious an op- tion for some young girls experiencing family difficulties, or for those interested in seeking independence. As a result, many young women choose?or are forced?to endure family difficulties for longer, or to seek alternative living arrangements (moving in with relatives or partners, for instance). Those who cannot remain at home or find alternative ar- rangements (or who are in foster care and have no home to return to) invariably end up on the streets. The Context of the Streets Health has long been considered a factor that has an impact on employa- bility. A range of factors associated with being homeless, including lack of sleep, poor nutrition, repeated injuries, and inability to maintain good hygiene, compromise one?s ability to keep healthy and to recover from illness or injury. Homeless youth are also vulnerable to debilitating ill- nesses, sexually transmitted diseases, substance abuse, and trauma (De- matteo et al., 1999; MacDonald et al., 1994; Wang et al., 1991), all of which can impair their ability to obtain and maintain stable employment. There are also significant gendered differences in health status. Women were more likely to describe their current health status as ?un- healthy? (women = 24.5 percent; men = 18.5 percent), to report being de- pressed once a week or more (women = 77 percent; men = 60 percent), and were also more likely than men to report having to go without food for a day one or more times per week (men = 37 percent; women = 51 percent). This may be a reflection of the fact that they were less likely to be in shelters (where food is provided) but, perhaps more so, that they were more impoverished or had less control over personal resources, including what they earned. There is also evidence that women are more likely than men to use health services,6 6 For instance, Shout Clinic (a community health centre for street youth in Toronto) re- ports consistently that over one-half of their visits are by young women, in spite of the fact that women make up only one-third of the street youth population. which reflects not only differences in health status, but also the different issues young women face regarding reproductive health. Pregnancy (and the risk of it) presents perhaps one of the greatest challenges homeless young women face and puts them in a position of considering shelter options, relationships (sexual and otherwise), inde- pendence, and safety in profoundly different ways from young men. Pregnancy also physically limits what young women can (or choose to) do to earn money, and adds additional risk to the act of making money. The experience of criminal victimization, whether in terms of prop- erty crime, assault, or sexual assault, affects health and well-being. Con- sistent with lifestyle-exposure theory (Hindelang et al., 1978), many of the income-generating activities of homeless youth increase their risk of criminal victimization, as such activities routinely take place in unsafe spaces, involve physical risks, and expose young people to dangerous adults and peers. While research indicates that the rates of criminal vic- timization among homeless youth in Toronto (both male and female) is indeed much higher than among the general population (Gaetz, 2004; Tanner & Wortley, 2002), homeless women are more vulnerable to certain types of violent and sexual crimes than are adolescent men (Si- mons & Whitbeck, 1991). As a result, women both perceive and expe- rience personal safety?whether on the streets or in terms of interper- sonal violence?differently from the way men do. This affects their mobility, their choices of action, and their comfort levels in different en- vironments and their choices regarding they generate income. The gendered nature of the streets means that the various spaces that street youth colonize?to sleep, to occupy at night, to walk alone within, to eat, to meet friends, to drink or take drugs, to rest in or to oth- erwise exist within?carry different risks for men and women. These risks help shape the options and choices that street youth make about living arrangements, interactions with others, establishing significant relations, and independence. While our data showed that a similar per- centage of men and women report being absolutely without shelter, men are much more likely to be shelter or hostel users, and women are more likely to report that they are staying in ?their own place.? The difference in shelter usage reflects the fact that women are more likely to see shel- ters as having ?too many rules? (women = 59 percent; men = 4 5 percent), and because they are less likely to see shelters as safe (women = 55 per- cent; men = 43 percent).7 The fact that women are more likely to report staying in their ?own place? does not necessarily mean that they are living independently. One-third of this group were currently caring for their own children and therefore eligible for housing support. Slightly more than one-quarter were living with male partners. This suggests the need to more closely examine the living arrange- ments of homeless youth, and the degree to which this may reflect gen- dered differences in personal independence. Smith and Gilford (1998) argue that relative or hidden homelessness is more common for women than absolute homelessness (compared with men),8 Other research suggests that absolutely homeless women are more likely to experience violence by their partners than those who are housed (Browne & Basuk, 1997), and may in fact experience similar difficulties in exerting control over their choice of work, and income derived from it. ?Many women find themselves in interpersonal relationships in which in large part due to the dangers they face (including sexual assault). Because young women face unique challenges?and risks?on the streets, they will often move quickly to secure shelter or establish partnering relationships that pro- vide shelter, even if these relationships are problematic and regardless of whether the choice represents a safe or healthy decision. For instance, Maher, Dunlap, Johnson, and Hamid (1996) have demonstrated how im- poverished, crack cocaine-using women who lack independent living arrangements often wind up living ?in the household of an older male with a dependable income for a period of time? (p. 194). In exchange for shelter, women provide sex, companionship, domestic service, and drugs. They are not visibly homeless, but they sacrifice independence to boyfriends who exercise considerable control over their activities? including how they make money. Drug dealers, for example, use the la- bour power of their ?girlfriends? in exchange for the provision of shelter. 7 At the time of this survey, there were 11 shelters for youth in Toronto. Two were exclu- sively for women, one for men, and the rest were co-ed. 8 By ?absolute? homelessness, we mean sleeping in spaces that are unfit for human habi- tation (e.g., rooftops, doorways, parks, under bridges). ?Relative? homelessness refers to environments such as short-term rentals and temporarily staying with friends. they are dependent on another persona or persons for their (and their children?s) survival? (Tessler et al., 2001, p. 251). Gender and Subsistence on the Streets The factors that lead to homelessness, and the rigours of life on the streets?affecting health and well-being, personal safety, and social and living arrangements?have consequences for employment and money- making. Explanations of gender differences that rely on educational at- tainment and personal motivation as the primary predictors of employ- ment status (as is often the case with human capital theory) overlook other important factors. For instance, young people who have suffered sexual and physical abuse and may be experiencing mental health stresses will find it difficult to compete economically. This argument is supported by recent research on youthful victimization where victimiza- tion during adolescence has been shown to have negative effects on oc- cupational status and earnings in young adults (Macmillan, 2000). While the economic activities in which homeless youth engage carry many risks, young women are more likely to experience stress, depres- sion, and lower levels of work satisfaction. In our survey, young men were more likely to report that they are currently satisfied with their main way of making money than women (men = 72 percent; women = 5 2 percent). Men typically reported features of their work such as ?being their own boss? as significant. Conversely, women were more likely to report the experience of abuse and humiliation as reasons for not liking their current work. Women were also more pessimistic about the possi- bility of finding better work in the future, with 20.8 percent reporting they were not very hopeful compared with 8 percent of men. The experience of stress, fear, and depression may also influence young women?s occupational choices. For instance, for safety reasons, young women will be more likely to engage in activities that take place in public spaces that are open, well-travelled and well-lit. In general, it appears that they are less likely to be drawn to criminal economic activi- ties that rely on aggression, violence, or intimidation. In addition, it is clear that decisions about economic activity are complex, and are not always made independently. Because the streets are a male space, young women are less likely to operate independently when working, and are also more likely to find themselves engaging in economic activities (sex trade, drug dealing) where they are forced to surrender independence?and earnings?to others (usually men). Further research must be carried out on how interactions with (and the perceptions of) the general public shape the gendered nature of work on the streets. We do know, for instance, that street prostitution is orga- nized spatially, so that specific urban areas become identified as ?tracks? where particular groups (men, women, transgendered persons) operate. There are also some indications that street youth, in panhandling or squeegeeing, may think tactically about the relationship between gender and ?giving.? The identities of homeless youth, as expressed and articu- lated through work, are not merely the product of street youth interac- tions and constructions, but also involve a broader negotiation of their identity with other segments of society. Female homeless youth, then, occupy economic niches on the street not just because the work itself is gendered, but also because their expe- rience of being young homeless women shapes what is possible. Home- less youth, unlike other youth, spend much time in the public realm. Their money-making activities are often conflated with social and leisure activities. In this public context, masculinities and femininities are pro- duced and reproduced in various ways. This is consistent with research in Britain that suggests women are unable to claim a place on the streets in ways that men can, and thus their survival strategies differ. According to Wardhaugh (1999), ?women must ? disappear? in order to survive, while men have the additional option of seeking safety in numbers, by claiming the city streets as their own? (p. 103). Conclusion The money-making activities of homeless youth are clearly gendered. The differences need to be understood in terms of the living arrange- ments and risks associated with being homeless. Since much of the ?work? of homeless youth falls outside the formal labour market, de- mand-side factors associated with occupational segregation do not apply to the economic opportunities available to homeless youth. Much of the demand for female labour in this unregulated environment is on the ba- sis of a barter system?domestic labour, sex, or storing and running drugs in exchange for shelter. Unlike women who are adequately housed and participate in the formal labour market, the private and public spheres of domestic work and paid employment for homeless women are not separate spaces. Fu- ture theorizing on the survival strategies of marginal women must go beyond dealing with gender-role socialization, the family division of la- bour, the operation of the formal labour market, and attitudes and beha- viours of employers. Initiatives designed to improve the conditions of these youth?health services, housing, employment training and place- ment and support programs?must take into account the degree to which the experience of homelessness is gendered. Bill O?Grady works in the Department of Sociology and Anthropology at the University of Guelph. Stephen Gaetz is the Associate Dean of Research and Field Development in the Faculty of Education, York University. This research was conducted on behalf of the Shout Clinic (Central Toronto Community Health Centres), a community health centre for homeless youth in Toronto, and funded by The City of Toronto, Human Resources and Development Canada, and Shout Clinic. This article was originally published in the Journal of Youth Studies, Volume 7, Issue 4, 200 4 , pages 397 ? 416. References Baron, S. (2001). Street youth labour market experiences and crime. Canadian Review of Sociology and Anthropology, 38(2), 189-215. Baron, S., & Hartnagel, T. F. (1998). Street youth and criminal violence. Journal of Research in Crime and Delinquency, 35(2), 166-192. Beitchman, J., Zucker, K., Hood, J., Dacosta, G., Akman, D., & Cassavia, E. (1992). A review of the long-term effects of child sexual abuse. Child Abuse & Neg- lect, 16(1), 101-118. Browne, A., & Bassuk, S. (1997). Intimate violence in the lives of homeless and poor housed women: Prevalence and patterns in an ethnically diverse sam- ple. American Journal of Orthopsychiatry, 67(2), 261-278. Carlen, P. (1996). Jigsaw: A political criminology of youth homelessness. Buckingham: Open University Press. City of Toronto (1999). Taking responsibility for homelessness: An action plan for To- ronto. Toronto: Mayor?s Homelessness Action Task Force, Author. Crompton, R. (1997). Women and work in modern Britain. Oxford: Oxford Universi- ty Press. Daly, G. (1996). Homeless: Policies, strategies, and lives on the street. New York: Routledge. Davies, S., Mohser, C., & O?Grady, B. (1996). Educating women: Gender inequali- ties among Canadian university graduates. Canadian Review of Sociology and Anthropology, 33(2), 125-142. Dematteo, D., Major, C., Block, B., Coates, R., Fearon, M., Goldberg, E., King, S. M., Millson, M., O?Shaughnessy, M., & Read, S. (1999). Toronto street youth and HIV/AIDS: Prevalence, demographics and risks. Journal of Adolescent Health, 25(5), 358-366. Ennew, J., & Milne, B. (1997). Methods of research with street and working children: An annotated bibliography. Stockholm: Radda Barnen. Fitzpatrick, S. (2000). Young homeless people. London: Macmillan. Gaetz, S. (2004). Safe streets for whom?: Homeless youth, social exclusion and criminal victimization. Canadian Journal of Criminology and Criminal Jus- tice, 46(4), 423-455. Gaetz, S., & O?Grady, B. (2002). Making money: Exploring the economy of young homeless workers. Work, Employment and Society, 16(3), 433-456. Gaetz, S., O?Grady, B., & Vaillancourt, B. (1999). Making money: The shout clinic report on homeless youth and employment. Toronto: Central Toronto Commu- nity Health Centres. Gardner, C. B. (1990). Safe conduct: Women, crime and self in public places. So- cial Problems, 37(3), 311-328. Greene, J. M., Ennet, S. T., & Ringwalt, C. L. (1999). Prevalence and correlates of survival sex among runaway and homeless youth. American Journal of Public Health, 89(9), 1406-1409. Greene, J. M., & Ringwalt, C. L. (1998). Pregnancy among three national samples of runaway and homeless youth. Journal of Adolescent Health, 23(6), 370-377. Hagan, J., & McCarthy, B. (1991). Homelessness: A criminogenic situation?. Brit- ish Journal of Criminology, 31(4), 393-410. Hagan, J., & McCarthy, B. (1997). Mean streets: Youth crime and homelessness. Cam- bridge: Cambridge University Press. Hanson, S., & Pratt, G. (1995). Gender, work and space. London: Routledge. Hatty, S. (1996). The violence of displacement: The problematics of survival for homeless young women. Violence Against Women, 2(4), 412-428. Hindelang, M. S., Gottfredson, M., & Garogalo, J. (1978). Victims of Personal Crime. Cambridge: Ballinger. Hughes, K., & Lowe, G. (1993). Unequal return: Gender differences in initial em- ployment among university graduates. Canadian Journal of Higher Educa- tion, 23(1), 37-55. Janus, M., McCormack, A., Burgess, A., & Hartman, C. (1987). Adolescent runa- ways: Causes and consequences. Lexington, MA: Lexington Press. Kaufman, R. L. (2002). Assessing alternative perspectives on race and sex em- ployment segregation. American Sociological Review, 67(4), 547-572. Kufelt, K., & Nimmo, M. (1987). Youth on the street: Abuse and neglect in the eighties. Child Abuse & Neglect, 11(4), 265-291. Kral, A. H., Molnar, B. E., Booth, R. E., & Waters, J. K. (1997). Prevalence of sex- ual risk behaviour and substance use among runaway and homeless ado- lescents in San Francisco, Denver and New York City. International Journal of STD and AIDS, 8(2), 109-117. MacDonald, N. E., Fischer, W. A., & Wells, G. A. (1994). Canadian street youth: Correlates of sexual risk-taking activity. Pediatric Infectious Disease Jour- nal, 13(8), 690-697. MacMillan, R. (2000). Adolescent victimization and income deficits in early adulthood: Rethinking the costs of criminal violence from a life course perspective. Criminology, 38(2), 553-587. Maher, L., Dunlap, E., Johnson, B., & Hamid, A. (1996). Gender, power and alter- native living arrangements in the inner city crack culture. Journal of Research in Crime and Delinquency, 33(2), 181-182. Marini, M. (1989). Sex differences in earnings in the United States. Annual Review of Sociology, 15, 343-380. Massey, D., (1998). The spatial construction of youth culture. In T. L. Skelton & G. Valentine (Eds.), Cool places: Geographies of youth cultures (pp. 121- 129). London: Routledge. McRobbie, A., & Garber, J. (1975). Girls and subcultures: An exploration. In S. Hall & T. Jefferson (Eds.), Resistance through rituals: Youth subcultures in post - war Britain (pp. 209-222). London: Hutchinson. Messerschmidt, J. (1995). From patriarchy to gender: Feminist theory, criminolo- gy and the challenge for diversity. In N. H. Rafter & F. Heidensohn (Eds.), International feminist perspectives in criminology: Engendering a discipline (pp. 167-188). Philadelphia: Open University Press. Novac, S., Serge, S., Eberle, M., & Brown, J. (2002). On her own: Young w omen and homelessness in Canada. Ottawa: Canadian Housing and Renewal Associa- tion. Robinson, C. (2000). Creating space, creating self: Street-frequenting youth in the city and the suburbs. Journal of Youth Studies , 3(4), 429-443. Sibley, D. (1995). Geographies of exclusion: Society and difference in the west. London: Routledge. Simons, R., & Whitbeck, L. (1991). Sexual abuse as a precursor to prostitution and victimization among adolescent and adult homeless women. Family Is- sues, 12(3), 361-379. Smith, J., & Gilford, S. (1998). Not only street sleepers are homeless: Issues of gender and ethnicity in researching youth homelessness. Youth and Pol i- cy, 59, 70-85. Statistics Canada (2002). Labour force historical review (Rev. ed.) [CD -ROM]. Otta- wa: Author. Stephens, S. (2001). Street children in Moscow: Using and creating social capital. The Sociological Review, 49(4), 530-542. Tanner, J., & Wortley, S. (2002). The Toronto youth leisure and victimization survey final report. Toronto: University of Toronto. Tessler, R., Rosenbeck, R., & Gamache, G. (2001). Gender differences in self- reported reasons for homelessness. Journal of Social Distress and the Home- less, 10(3), 243-254. Tyler, K., Hoyt, D., & Whitbeck, L. (2000). The effects of early sexual abuse on later sexual victimization among female homeless and runaway adoles- cents. Journal of Interpersonal Violence, 15(3), 235-250. Valentine, G. (1996). Children should be seen and not heard: The production and transgression of adult?s public space. Urban Geography, 17(3), 65-89. Wang, E., King, S., & Goldberg, E. (1991). Hepatitis B and human immunodefi- ciency virus infection in street youth in Toronto, Canada. Pediatric Infectious Disease Journal, 10(2), 130-133. Wardhaugh, J. (1999). The unaccommodated woman: Home, homelessness and identity. The Sociological Review, 47(1), 91-109. Webber, M. (1991). Street kids: The tragedy of Canada?s runaways. Toronto: Univer- sity of Toronto Press. Whitbeck, L., Hoyt, D., & Ackley, K. (1997). Abusive family backgrounds and later victimization among runaway and homeless adolescents. Journal of Re- search on Adolescence, 7(4), 375-392. Whitbeck, L., & Simons, R. (1993). A comparison of adaptive strategies and pat- terns of victimization among homeless adolescents and adults. Violence and Victims, 8(2), 135-198. White, R. (1994). Street life: Police practices and youth behaviour. In R. White & C. Alder (Eds.), The police and young people in Australia. Cambridge: Cam- bridge University Press. Chapter 3.5 Social Stigma and Homeless Youth SEAN A. KIDD The family histories of most homeless youth are troubled, often consist- ing of disrupted and abusive home environments. High rates of drug and alcohol abuse are found among the parents of street youth, as is pa- rental criminality (Hagan & McCarthy, 1997; Maclean et al., 1999). Addi- tionally, a high percentage of the families are on social assistance (Ring- walt et al., 1998), and disrupted families are common, with few homeless youth reporting having lived with both biological parents (Hagan & McCarthy, 1997). Also common are reports of marital discord (Dadds, et al., 1993), domestic violence, and household moves involving frequent changes of school (Buckner & Bassuk, 1997). The majority of the research into the backgrounds of street youth has focused on physical and sexual abuse, rates of which are consistently high (MacLean et al., 1999; Molnar et al., 1998; Ringwalt et al., 1998). His- tories of emotional abuse (Ringwalt et al., 1998) and neglect (Dadds et al., 1993) are also frequently reported. These negative home experiences are associated with a host of other problems, including poor performance in school, conflict with teachers, and conduct problems (Feitel et al., 1992; Hagan & McCarthy, 1997; Rotheram-Borus, 1993). All of these phenomena are understood to be different or deviate from the ideals of the ?social norm,? and having such experiences has the effect of placing the individual outside of the cultural models of ?normalcy.? This is supported by a vast literature on the topic of child abuse, neglect and other dysfunctional backgrounds indicating that children who have suffered in these ways feel isolated and ostracized, and perceive others as dangerous and rejecting ? see Wagner (1997) and Kendall-Tackett, Williams, and Finkelhor (1993). Having such abusive and disrupted childhoods initiates a process of stigmatization in which children are identified and labelled as different, and as their opportuni- ties, social and otherwise, narrow due to the beliefs and actions of others (Tomlin, 1991). For many homeless youth, having these types of early experiences likely leaves them more vulnerable to negative experiences associated with social stigma on the streets, given research showing that stigmatization has a greater impact upon the self-esteem of persons who have been abused in childhood (Coffey et al., 1996; Crocker & Major, 1989). Street youth face many dangers and sources of stress in their lives on the street. To support themselves, they engage in activities such as trying to find work, seeking money from family or friends, panhandling, prostitution, survival sex (sex for food, shelter etc.), drug-dealing, and theft (Greene et al., 1999; Hagan & McCarthy, 1997; Kipke et al., 1997). The difficulty of surviving on the streets is highlighted by the large number of homeless youth who regularly lack shelter and go hungry (Antoniades & Tarasuk, 1998). Moreover, street life presents numerous dangers and stresses in the form of physical and sexual assaults and oth- er types of victimization (Whitbeck et al., 2000). Drug abuse is a common way of coping with these stressors (Adlaf et al., 1996) and addiction is a major problem (Greene & Ringwalt, 1996). There is a high incidence of mental disorders among homeless youth, such as depression, post- traumatic stress disorder, and suicidal behaviour (Kidd, 2004; Whitbeck et al., 2000; Yoder, 1999). Mortality rates for homeless youth have been found to be 12 to 40 times those of the general population (Shaw & Dorl- ing, 1998), and suicide is the leading cause of death (Roy et al., 2004). Despite the powerful and pervasive social stigma faced by homeless youth, it remains an overlooked topic in the research literature, with a few exceptions (Schissel, 1997; Kidd, 2003, 2004). More commentary can be found in the adult homeless literature (Boydell et al., 2000; Lankenau, 1999; Phelan et al., 1997). The perception of discrimination based upon negative stereotypes is related to feelings of worthlessness, loneliness and social alienation, and suicidal thoughts. The linkages between social stigma, depression, and suicidality have been previously found among other groups of adolescents who face high levels of discrimination (e.g. gay, lesbian, and bisexual youth) (Radkowsky & Siegel, 1997) and among adult homeless among whom stigma has been related to social isolation and a devalued sense of self (Boydell et al., 2000). Looking for connections This study examined the relationship between greater levels of perceived social stigma and lower self-esteem, loneliness, suicidal thoughts, and the feeling of being ?trapped? (Kidd, 2004, 2006). In particular, we looked at stigma and three street demographic factors. Prior work that has suggested that youths who are involved in the sex trade (Kidd & Kral, 2002) and homeless persons who engage in pan handling (Lanke- nau, 1999) are stigmatized to a greater extent. It was expected, therefore, that the degree of involvement in pan handling and sex trade activities would be positively associated with perceived stigma. The third factor ? the total amount of time the youth had been homeless ? was also ex- pected to be positively related to perceived stigma. While such a rela- tionship has not been examined previously, it would seem logical to suggest that the longer a youth had been homeless, the greater his or her exposure to stigmatizing circumstances and perception of stigma. Gender, sexual orientation, and ethnicity were also expected to be related to the degree of stigma perceived by the participants in this study. There is a large body of literature suggesting that gay, lesbian, and bisexual adolescents both experience and perceive markedly higher levels of socially oppressive views and practices (see Radkowsky & Sie- gel, 1997, for review). Similar findings have been consistently hig- hlighted for female adolescents (Leadbeater & Way, 1996) and ethnic minority adolescents (Comer, 1995). While perceived stigma among homeless youths having these cha- racteristics has not been evaluated, there is evidence that females and gay/lesbian homeless youth face greater adversity and victimization on the street (Cauce et al., 2000; Cochran et al., 2002; O?Grady & Gaetz, 2004). Given these findings it was expected that, for homeless adoles- cents, female gender, gay/lesbian sexual orientation, and non-white eth- nicity would be positively related to perceived social stigma. Lastly, of the aspects of stigma measured (understanding of public perception, actions of public against self, self-blame or guilt due to stig- matized status, struggles against larger society), feelings of shame and guilt were expected to have the greatest adverse impact upon mental health, a finding noted among other populations (e.g. HIV-positive per- sons) (Berger et al., 2001). Self-blame was also expected to be higher among sex trade-involved youth and gay, lesbian, and bisexual partici- pants given previous findings of generally higher self-blame/guilt among those groups (Kidd & Kral, 2002; Kruks, 1991). Examination of self blame as it was related to gender, ethnicity, pan handling, and total time on the streets was exploratory, however, given a lack of relevant previous find- ings to inform hypotheses. Talking to young homeless people about stigma To participate in the study, youth had to be 24 years of age or younger and have had no fixed address or be living in a shelter at the time of the survey. The study took place in agencies and on the streets of New York City and Toronto, and 208 youths participated. Street interviews were done in a range of locations where homeless youths congregated or pan- handled (e.g., sidewalks of streets with heavy pedestrian traffic, public parks). Agency interviews included a youth agency in New York provid- ing ranging services for disadvantaged youth and two agencies in Toron- to which provide a similar range of drop-in services targeting homeless youth, one of which focused on providing services for youth involved in the sex trade. In New York, we interviewed 100 youths (39 agency, 61 street) and in Toronto 108 youths (31 at each agency, 46 street). Although inter- viewed in different cities, the youths? narratives and survey responses did not suggest any notable variation on the basis of their geographical location. Youths from New York City and Toronto are not considered separately, given the lack of any significant differences between these groups on any of the independent or dependent variables measured. Participants were reimbursed with restaurant coupons and 97 percent of those approached agreed to participate. The data used in the present study were derived from the quantitative survey component of an inter- view including both qualitative and quantitative elements. Of the 208 participants, 122 (59 percent) were male, 84 (40 percent) were female, and 2 (1 percent) were transgendered (male to female). With respect to ethnicity, 56 percent were White, 12 percent Black, 12 percent Hispanic, 5 percent Aboriginal, 14 percent of mixed ethnicity, and the remainder varied. The ages ranged from 14 to 24, with a median age of 20. The average age of the youths? first experience of leaving/being thrown out of home was 15, with a mean level of education of 10.6 years. A substantial proportion (57 percent) reported having been homeless for more than two years, with 33 percent reporting continuous homelessness and 40 percent having had conventional housing 25 percent of the time. Most youth resided in street and/or squat locations (47 percent), with 26 percent ?couch-surfing? (temporarily living with others), and 14 percent lived in shelters. Most youths reported some combination of income sources with pan handling (45 percent), dealing drugs (23 percent), a job (23 percent), and sex trade involvement (15 percent) appearing with the most frequency. We developed a 12-item survey to measure the sense of social stig- ma. Each item response was formatted as a 4-point Likert-type scale (strongly disagree, disagree, agree, strongly agree) relating to the expe- rience of being stigmatized: ? ?I have been hurt by how people have reacted to me being home- less.? ? ?I feel that I am not as good as others because I am homeless.? ? ?I feel guilty and ashamed because I am homeless.? ? ?People seem afraid of me because I am homeless.? ? ?Some people act as though it is my fault that I am homeless.? ? ?Homeless people are treated like outcasts.? ? ?Knowing that you are homeless, people look for things wrong about you.? ? ?I have been insulted by strangers because I am homeless.? ? ?Most people think that homeless people are lazy and disgusting.? ? ?Homeless people can?t get jobs because they are homeless.? ? ?I have to fight against the opinions and values of society.? ? ?Homeless people are harassed by the police because they are home- less.? Group stigma vs. individualized stigma We distinguished between a generalized stigma factor, relating to the idea that ?Homeless people are treated like outcasts,? indicative of an understanding of public perceptions and actions by the public based upon stigma (insults, harassment, biases in hiring), being hurt by social stigma, and having to ?fight? against social stigma. The second, more personal component focuses on self-blame (feeling guilty, ashamed or not as good as others). We examined the effects of gender, ethnicity, sexual orientation, panhandling, sex trade involvement, and total time on the streets upon perceived social stigma, hypothesizing that these variables would all have significant relationships with the experience of stigma. Contrary to our hypothesis, female gender was not found to have a significant impact on either the self-blame component of stigma nor the general stigma component. For the purposes of analyzing the relation- ship between ethnicity and stigma, two groups were created, composed of White youth (n=117) , and youth of other ethnicities (n=91; including Hispanic [ n=24] ; Black, [ n=25] ; Asian, [ n=3 ] ; Aboriginal, [ n=11] ; mixed, [ n=28]. Contrary to our hypothesis, we found that White youths reported greater general perceived stigma, but there was no significant difference among ethnicities in reports of self-blame. In relation to sexual orientation, we created four groups: straight (n=115), some degree of bisexual orientation (n=40), bisexual (n=31), and primarily gay/lesbian (n=22). Our analysis indicated that while sexual orientation was not significantly related to general stigma, it was, as we expected, significantly related to self-blame. Examination of panhandling and sex trade involvement as they related to stigma indicated that, consistent with our hypotheses, youth who panhandle report higher levels of general stigma, but lower levels of self- blame, while degree of sex trade involvement was not associated with general stigma, but was related to higher levels of self-blame. Lastly, the hypothesis that total time on the streets would be related to perceived stigma was supported with respect to general stigma, but not self-blame. Simultaneous multiple regression analyses were used to examine the relationship between general stigma and self-blame and four depen- dent variables: self-esteem, loneliness, feeling trapped, and suicidal thoughts or attempts.1 The burden of stigma We found that both general stigma and self-blame were predictors of these four variables. Our analysis indicated that the experience of social stigma varied de- pending on ethnicity, sexual orientation, subsistence activity, and total time on the streets. The experience of stigma was also found to have sig- nificant relationships with low self-esteem, loneliness, feelings of being trapped, and suicidal thoughts or attempts. Since previous work that has indicated that female adolescents ex- perience greater amounts of discrimination (Leadbeater & Way, 1996) and that homeless girls are more disadvantaged financially (O?Grady & Gaetz, 2004) and more frequently victimized on the streets (Cauce et al., 2000), we expected that female participants would report greater levels of social stigma relative to males. This was not the case for either general or self-blame aspects of social stigma. Similar hypotheses were made with respect to ethnicity and sexual orientation, based on evidence that ethnic minority adolescents (Comer, 1995) and gay, lesbian, and bisexual adolescents (Radkowsky & Siegel, 1997) face higher levels of stigma relative to the general population and gay, lesbian, and bisexual homeless youth are more frequently victi- mized on the streets (Cochran et al., 2002). We found the inverse of the hypothesized relationship between stigma and ethnicity, with white youths reporting greater general stigma, though no significant difference in degree of self-blame. Sexual orientation did, however, emerge as hav- ing a significant relationship with stigma. As predicted, a linear relation- 1 Of those we interviewed, 46 percent reported making at least one suicide attempt in home or street environments, and of these, 78 percent reported that they had made more than one attempt. Differentiated by gender, 55 percent of females and 40 percent of males reported at least one attempt. Variations among the different ethnicities were mi- nor. Methods including overdosing (42 percent), cutting with a sharp object (32 percent), hanging (15 percent), jumping from a height (7 percent), with miscellaneous remainders. ship was found between the degree of bisexuality and gay/lesbian sexual orientation and the amount of guilt and self-blame as it related to stigma. It seems that gay and bisexual youth engage in more self-blame in reaction to stigma based upon homeless status, perhaps reflecting pre- vious evidence of their having poorer psychological and physical health relative to straight homeless youth (Cochran et al., 2001). A similar rela- tionship was not found with general stigma. The findings that gender, ethnicity, and sexual orientation are not related strongly with a general sense of stigma might be explained by the design of the survey instrument, which connected stigma solely with homelessness. As one participant noted while filling out the survey: ?People aren?t afraid of me because I am homeless. People are afraid of me because I am Black.? It may be that, with the exception of sexual orientation and self-blame, stigma associated with these demographic variables is occurring in an additive fashion, in which homelessness- specific stigma is not perceived as substantially different among these groups. Females and minorities, however, may experience additional challenges. Future work may serve to better delineate the implications of multiple sources of stigma. We confirmed our hypotheses on the impact of panhandling, sex trade involvement, and total time homeless upon perceived stigma. Pan- handlers publicly display their status as homeless persons and regularly face humiliating interactions with strangers and authorities. Having panhandling as a primary source of income was strongly related to per- ception of general social stigma. Conversely, panhandling had a signifi- cant negative relationship with self-blame, confirming Lankenau?s ob- servation that panhandlers find constructive ways of managing their stigmatized identities (Lankenau, 1999). Sex trade involvement was related to self-blame but, contrary to our hypothesis, not general stigma, possibly reflective of the additional stig- ma ascribed to prostitution (Brock, 1998). Sex trade work may, however, be similarly affected by the question of multiple forms of discrimination, unlike panhandling, which is more closely associated with homelessness. Lastly, the total amount of time spent homeless was significantly re- lated to general perceived stigma, but not self-blame. This is likely simi- lar to the relationship between stigma and panhandling in that youth who are homeless for longer are exposed to greater amounts of social stigma, with the greater amount of time potentially allowing for adjust- ment to discrimination, such that guilt and self-blame in response to stigma are reduced. Consistent with previous analyses (Kidd, 2004), perceived stigma was found to have a significant relationship with low self-esteem, loneli- ness, feelings of being trapped, and suicidal thoughts and attempts. Feeling trapped, a construct of helplessness and hopelessness, has emerged in previous work as being central to suicide attempts among homeless youth (Kidd, 2004, 2006). Of the variables noted above, per- ceived stigma was most strongly associated with loneliness and feeling trapped. While the cross-sectional nature of this study does not allow for an examination of direction of these relationships, these findings suggest that the well-documented tendency of society to blame homeless persons for their predicament (Phelan et al., 1997) may further compromise the mental health of youth already grappling with myriad risks and chal- lenges. The potential influence of stigma in the lives of these youth may extend to mortality, given the relationships between the above variables and suicide, and the recent finding that suicide is the leading cause of death for this population (Roy et al., 2004). As hypothesized, self-blame caused by a stigmatized status emerged as having the strongest relationships with the mental health variables measured in this study. It is likely that self-blame reflects the degree to which these youths? stigmatized status is internalized, with the implication that the degree to which homeless youths internalize socie- ty?s negative view of them is a central aspect of the process through which discrimination affects mental health. Such a finding has been noted among other populations (Lee et al., 2002). The association between stigma and low self-esteem is not, contrary to common belief, typically found among most stigmatized groups (Crocker, 1999; Crocker & Major, 1989; Pinel, 1999). Recently, theorists have emphasized within-group variability (Crocker, 1999; Pinel, 1999), the meanings people give to situations in which stigmatization might be occurring (Crocker, 1999), and the protective coping strategies such indi- viduals employ (Crocker & Major, 1989). Crocker and Major (1989) proposed mediating factors that may ex- plain the strong impact social stigma appears to have on street youth. These mediating influences include the consideration that since these youth have not had the stigma of homelessness since birth, they have likely not had as many opportunities to adapt and develop coping strat- egies related to that stigma. They are also a group that has been exposed to many of mainstream society?s beliefs about drug addiction, poverty, prostitution, etc., before they identify with these characteristics. Thus, negative stigmatizing evaluations are more salient since they may to a certain extent have internalized those beliefs and more readily apply them to themselves. Homeless youth are stigmatized for reasons that are largely thought to be the responsibility of the person ? poverty, drug addiction etc. (Schissel, 1997). Such groups are stigmatized to a greater extent than those not thought responsible for their conditions, such as those with a developmental disability). With respect to coping with stigma, homeless youth may have diffi- culty putting in place the protective mechanism of devaluing the stan- dards against which they are criticized. Beliefs about physical appear- ance, being drug-free, financial success, and education are central to Western culture and difficult to ignore. Additionally, the abusive pasts of many street youth have likely left them more vulnerable to negative so- cial and emotional consequences of stigmatization (Coffey et al., 1996). Lastly, the stigma and social oppression experienced by homeless youth appears to be occurring, for most, at a very high level. Homeless youth speak of a multi-levelled and institutionalized dis- crimination that is probably one of the more extreme forms to be found in North America, and a constant condition that cannot be escaped by ?going home? (Kidd, 2004). These findings indicate that homeless youths? experience of stigma plays a major role in their mental health status and suicide risk level. It will likely prove important for interventions to address social stigma as it is perceived and experienced by these young people, exploring how these perceptions are affecting their mental health in various domains, and helping them to find ways to insulate themselves from the discrimi- nation that they face. This may involve working on ways to replace in- ternalized messages of guilt and shame with a more empowering under- standing of the various factors underlying stigma and systemic discrimi- nation. Finally, counsellors should be aware of the effects of discrimination and stigma on some groups, including gay, lesbian, bi- sexual, and transgendered persons, youth involved in the sex trade, and of how perceptions of stigma change over time as youth are exposed to ongoing discrimination. Sean Kidd is an assistant professor with the McMaster Department of Psy- chiary and Behavioural Neurociences. His primary research areas are suicide and resiliency among homeless youth and the application of qualitative metho- dologies in psychological research. This chapter is a condensation of an article published in the Journal of Youth and Adolescence, vol. 36, 200 7. References Adlaf, E. M., Zdanowicz, Y. M., & Smart, R. G. (1996). Alcohol and other drug use among street-involved youth in Toronto. Addictions Research & Theory , 4(1), 11? 24. Aish, A., & Wasserman, D. (2001). Does Beck?s Hopelessness Scale really meas- ure several components?. Psychological Medicine, 31, 367? 372. Antoniades, M., & Tarasuk, V. (1998). A survey of food problems experienced by Toronto street youth. Canadian Journal of Public Health, 89(6), 371? 375. Beck, A. T., Weissman, A., Lester, D., & Trexler, L. (1974). The measurement of pessimism: The hopelessness scale. Journal of Consulting and Clinical Psychol- ogy, 42(6), 861? 865. Berger, B. E., Ferrans, C. E., & Lashley, F. R. (2001). Measuring stigma in people with HIV: Psychometric assessment of the HIV stigma scale. Research in Nursing & Health , 24(6), 518? 529. Boydell, K. M., Goering, P., & Morrell-Bellai, T. L. (2000). Narratives of identity: Re-presentation of self in people who are homeless. Qualitative Health Re- search, 10(1), 26? 38. Brock, D. (1998). Making work, making trouble: Prostitution as a social problem. To- ronto: University of Toronto Press. Buckner, J. C., & Bassuk, E. L. (1997). Mental disorders and service utilization among youths from homeless and low-income household families. Journal of the American Academy of Child and Adolescent Psychiatry, 36(7), 890? 900. Cauce, A. M., Paradise, M., Ginzler, J. A., Embry, L., Morgan, C., Lohr, Y., & Theofolis, J. (2000). The characteristics and mental health of homeless ado- lescents: Age and gender differences. Journal of Emotional and Behavioral Dis- orders, 8(4), 230? 239. Cochran, B. N., Stewart, A. J., Ginzler, J. A., & Cauce, A. M. (2002). Challenges faced by homeless sexual minorities: Comparison of gay, lesbian, bisexual, and transgender homeless adolescents with their heterosexual counterparts. American Journal of Public Health, 92(5), 773? 777. Coffey, P., Leitenberg, H., Hennig, K., Turner, T., & Bennett, R. T. (1996). Media- tors of the long-term impact of child sexual abuse: Perceived stigma, be- trayal, powerlessness, and self- blame. Child Abuse & Neglect, 20(5), 447? 455. Comer, J. P. (1995). Racism and African American adolescent development. In C. V. Willie, P. P. Rieker, B. M. Kramer, & B. S. Brown (Eds.), Mental health, rac- ism, and sexism (pp. 151? 170). Pittsburgh: University of Pittsburgh Press. Crocker, J. (1999). Social stigma and self-esteem: Situational construction of self- worth. Journal of Experimental Social Psychology, 35(1), 89? 107. Crocker, J., & Major, B. (1989). Social stigma and self-esteem: The self-protective properties of stigma. Psychological Review, 96(4), 608? 630. Crocker, J., Major, V., & Steele, C. (1998). Social stigma. In D. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychology. Vol. 2 (4th ed., pp 504? 553). New York: McGraw-Hill. Dadds, M., Braddock, D., Cuers, S., Elliott, A., & Kelly, A. (1993). Personal and family distress in homeless adolescents. Community Mental Health Journal, 29(5), 413? 422. Feitel, B., Margetson, N., Chamas, J., & Lipman, C. (1992). Psychosocial back- ground and behavioral and emotional disorders of homeless and runaway youth. Hospital and Community Psychiatry, 43(2), 155? 159. Gorsuch, R. L. (1983). Factor analysis (2nd ed.). Hillsdale, N.J.: Lawrence Erlbaum. Greene, J. M., Ennett, S. T., & Ringwalt, C. L. (1999). Prevalence and correlates of survival sex among runaway and homeless youth. American Journal of Public Health, 89(9), 1406? 1409. Greene, J. M., & Ringwalt, C. L. (1996). Youth and familial substance use?s associ- ation with suicide attempts among runaway and homeless youth. Substance Use & Misuse, 31(8), 1041? 1058. Hagan, J., & McCarthy, B. (1997). Mean streets: Y outh crime and homelessness. Cam- bridge: Cambridge University Press. Kendall-Tackett, K. A., Williams, L. M., & Finkelhor, D. (1993). Impact of sexual abuse on children: A review and synthesis of recent empirical studies. Psy- chological Bulletin, 113(1), 164? 180. Kidd, S. A. (2003). Street youth: Coping and interventions. Child and Adolescent and Social Work Journal, 20(4), 235? 261. Kidd, S. A. (2004). ?The walls were closing in and we were trapped?: A qualita- tive analysis of street youth suicide. Youth & Society, 36(1), 30? 55. Kidd, S. A. (2006). Factors precipitating suicidality among homeless youth: A quantitative follow-up. Youth & Society , 37(4), 393? 422. Kidd, S. A., & Kral, M. J. (2002). Street youth suicide and prostitution: A qualita- tive analysis. Adolescence, 37(146), 411? 430. Kipke, M. D., Unger, J. B., O?Connor, S., Palmer, R. F., & LaFrance, S. R. (1997). Street youth, their peer group affiliation and differences according to resi- dential status, subsistence patterns, and use of services. Adolescence, 32(127), 655? 669. Kruks, G. (1991). Gay and lesbian homeless/street youth: Special issues and con- cerns. Journal of Adolescent Health, 12(7), 515? 518. Lankenau, S. E. (1999). Stronger than dirt: Public humiliation and status en- hancement among panhandlers. Journal of Contemporary Ethnography, 28(3), 288? 318. Leadbeater, B. J. R., & Way, N. (1996). Urban girls: Resisting stereotypes, creating identities. New York: New York University Press. Lee, S. L., Kochman, A., & Sikkema, K. J. (2002). Internalized stigma among people living with HIV-AIDS. AIDS and Behavior, 6(4), 309? 319. Lester, D. (2001). An inventory to measure helplessness, hopelessness, and hap- lessness. Psychological Reports, 89(3), 495? 498. Lewinsohn, P. M., Rohde, P., & Seeley, J. R. (1996). Adolescent suicidal ideation and attempts: Prevalence, risk factors, and clinical implications. Clinical Psychology: Science and Practice, 3(1), 25? 46. Maclean, M. G., Embry, L. E., & Cauce, A. M. (1999). Homeless adolescents? paths to separation from family: Comparison of family characteristics, psycholog- ical adjustment, and victimization. Journal of Community Psychology, 27(2), 179? 187. Meyer, I. H. (2003). Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: Conceptual issues and research evidence. Psychologi- cal Bulletin, 129(5), 674? 697. Molnar, B. E., Shade, S. B., Kral, A. H., Booth, R. E., & Watters, J. K. (1998). Sui- cidal behavior and sexual/physical abuse among street youth. Child Abuse & Neglect, 22(3), 213? 222. Noell, J. W., & Ochs, L. M. (2001). Relationship of sexual orientation to substance use, suicidal ideation, suicide attempts, and other factors in a population of homeless adolescents. Journal of Adolescent Health, 29(1), 31? 36. O?Grady, B., & Gaetz, S. (2004). Homelessness, gender and subsistence: The case of Toronto street youth. Journal of Youth Stud ies, 7(4), 397? 416. Phelan, J., Link, B. G., Moore, R. E., Stueve, A. (1997). The stigma of homeless- ness: The impact of the label ?homeless? on attitudes towards poor persons. Social Psychology Quarterly, 60(4), 323? 337. Pinel, E. C. (1999). Stigma consciousness: The psychological legacy of social ste- reotypes. Journal of Personality and Social Psychology, 76(1), 114? 128. Radkowsky, M., & Siegel, L. J. (1997). The gay adolescent: Stressors, adaptations, and psychosocial interventions. Clinical Psychological Review, 17(2), 191? 216. Ringwalt, C. L., Greene, J. M., & Robertson, M. J. (1998). Familial backgrounds and risk behaviors of youth with thrownaway experiences. Journal of Ado- lescence, 21(3), 241? 252. Rosenberg, M. (1989). Society and the adolescent self-image. Middletown, CT: Wes- leyan University Press. Rotheram-Borus, M. J. (1993). Suicidal behavior and risk factors among runaway youths. American Journal of Psychiatry, 150(1), 103? 107. Roy, E., Haley, N., Leclerc, P., Sochanski, B., Boudreau, J., & Boivin, J. (2004). Mortality in a cohort of street youth in Montreal. The Journal of the American Medical Association, 292(5), 569? 574. Russell, D. W. (1996). UCLA Loneliness Scale (Version 3): Reliability, validity, and factor structure. Journal of Personality Assessment, 66(1), 20? 40. Shaw, M., & Dorling, D. (1998). Mortality among street youth in the UK. The Lan- cet, 352(9159), 743. Schissel, B. (1997). Blaming children: Youth c rime, moral panics and the politics of hate. Halifax: Fernwood. Simons, R. L., & Whitbeck, L. B. (1991). Sexual abuse as a precursor to prostitu- tion and victimization among adolescent and adult homeless women. Jour- nal of Family Issues, 12(3), 361? 379. Tomlin, S. S. (1991). Stigma and incest survivors. Child Abuse & Neglect, 15(4), 557? 566. Wagner, B. M. (1997). Family risk factors for child and adolescent suicidal beha- vior. Psychological Bulletin, 121(2), 246? 298. Whitbeck, L. B., Hoyt, D. R., & Bao, W. (2000). Depressive symptoms and co- occurring depressive symptoms, substance abuse, and conduct problems among runaway and homeless adolescents. Child Development, 71(3), 721? 732. Yoder, K. A. (1999). Comparing suicide attempters, suicide ideators, and nonsui- cidal homeless and runaway adolescents. Suicide and Life-Threatening Beha- vior, 29(1), 25? 36. Chapter 3.6 How Young People Get off the Street: Exploring Paths and Processes JE F F KARABAN OW The vast majority of literature concerning homeless young people has focused on street engagement and street culture. Although this focus is vital to understanding etiology and street life experiences, there has been a surprising neglect on the part of the academic community to complete the analysis of street youth career patterns. The literature has provided an impressive grasp on the causes and consequences of street life ? in- cluding family dysfunction, abuse and trauma, exploitation and aliena- tion, poverty, addiction, and mental health and child welfare inadequa- cies ? but little acknowledgment of how some of these young people complete the cycle and move away from street culture (Alleva, 1988; Edelbrock, 1980; Ensign, 1998; Karabanow, 2003; Kufeldt & Nimmo, 1987; Kurtz et al., 1991). This study highlights the paths and processes involved in ?getting off the street,? told from the perspectives of young people and service providers. Funded by a Canadian National Homelessness Initiative grant, this study involved in-depth, semistructured interviews with 128 young people (90 males, 38 females) and 50 service providers in six Canadian cities (Toronto, Montreal, Halifax, Calgary, Ottawa, and Vancouver). Whereas the majority of interviews were conducted as one-on-one dis- cussions, several interviews were carried out as mini-focus groups (made up of two or three participants). Interview questions probed participants? experiences prior to street life, street experiences, and ways in which they have attempted to ?exit? this life. Service providers were asked about their experiences working with young people on the street and exiting street life. Purposive sampling in each site allowed for enhanced diversity within the participant arena ? young people in various stages of their street life career were chosen (approximately 20 participants per stage) as well as various service provider settings (such as drop-in clinics, emer- gency shelters, detoxification services, job training outlets, health centres, mobile crisis units, educational services, and supportive housing/second- stage offerings). Participants were recruited through advertisements placed in local newspapers, local hangout areas (such as parks and coffee shops), and social service agencies. Two young people (in different phases of their street career) were hired as research assistants and conducted interviews alongside the re- search team. The involvement of these young assistants in the research was important in recruiting participants and building trust, especially with hardcore street youth. Regular discussions/informal interviews with the two research assistants shed light on the research process and analysis of data. Moreover, their own life experiences became important data sources for the research team. Complementing this analysis were 15 brief case study portraits of a diverse set of organizational structures serving street youth across Cana- da. Data analysis involved content analysis and constructivist grounded theory foundations (Charmaz, 2004; Strauss & Corbin, 1990) through the implementation of open, axial, and selective coding schemas. The analy- sis involved locating common and dissimilar themes, building thematic narratives that surfaced from the construction, and linking core catego- ries. This article is organized using the core themes or narratives that emerged from the data.1 1 The researchers used the qualitative software package Atlas-TI in organizing and making sense of the large data set. The research was conducted between May 2004 and August 2005. This article explores the complex pathways used by young people in their attempts at street disengagement and sheds light on the strate- gies and obstacles involved in moving away from street culture. The findings suggest several interrelated dimensions to the exiting process ? including contemplation, motivation to change, securing help, transition- ing from the street, changing daily routine, and redefining one?s sense of self. Throughout these dimensions, street youth organizational structures play significant roles in supporting young people?s disengagement from street culture. Who Are Street Youth? A primary finding from this research is that the street youth population is diverse, complex, and heterogeneous. Although this notion has been previously highlighted (e.g., Karabanow, 2004a), it is pivotal here in or- der to comprehend the myriad avenues young people take to disengage (or attempt to disengage) from street culture. The generic term ?street youth? is made up of a number of subcul- tures (by no means mutually exclusive) including hardcore street- entrenched young people, squatters, group home kids, child welfare kids, soft-core ?twinkies,? ?in-and-outers,? punks, runaways, throwa- ways, refugees and immigrants, young single mothers, and those who are homeless because their entire family is homeless (Karabanow, 2003; Kufeldt & Nimmo, 1987; McCarthy, 1990; Michaud, 1989; Morrissette & McIntyre, 1989). Within these makeshift categories are numerous de- scriptors that signal street activities such as gang bangers, prostitutes, drug dealers, drug users, panhandlers, and squeegeers. Although these labels may denote some of the actions of young people on the street, for the purposes of this research, street youth are defined as young people between the ages of 16 and 24 who do not have a permanent place to call home and who, instead, spend a significant amount of time and energy on the street (e.g., in alleyways, parks, store- fronts, dumpsters, etc.); in squats (usually in abandoned buildings); at youth shelters and centres; and/or with friends (?couch surfers?). Such a broad description functions as a framework for the overall analysis that attempts to thread common themes and stories that emerge from the ex- periences of a diverse group of young people. Getting on the Street Exploring street engagement provides important context to street disen- gagement. For the majority of the sample, family life prior to street en- trance was characterized by physical, sexual, or emotional abuse; vi- olence and substance abuse within the home; and family instability, including numerous transitions and moves (i.e., divorce, separation, in- troduction of stepparents and stepchildren, moving residencies, chang- ing cities, and shifting living arrangements). Family life was seen as chaotic, disruptive, and inconsistent, with a lack of love, care, interest, and support from caregivers. For the most part, young people experienced loneliness, boredom, alienation, and neglect (in addition to such traumas as being witnesses or victims of violence, abuse, and substance misuse) within their family settings. Thus, it is hardly surprising that most of the young people in- terviewed viewed the street as a safer and more stable environment than home: That was the whole reason I would never try to live back home: in the last day/night that I slept there, my dad grabbed me by my throat and put me up against the wall ?cause I was thinking about leaving. So that was his answer ?cause my dad?s very short-tempered and high-fused . . . I would rather stay on the street than move back there. (Lisa, age 24, Halifax) An equally important (and alarming) factor pushing young people to the street involves problematic child welfare placements. More than half of our sample came to the street after having lived in a group home or in foster care. These experiences were most often described as uncar- ing, exploitative, and unstable. Numerous moves from group home to group home (or foster care to foster care), coupled with feelings of being treated as ?criminals,? ?delinquents,? or ?unwanted,? shaped young people?s transition to street life. Child welfare settings were described by participants as unresponsive to their needs and perceived as ?prisons? rather than loving and home-like structures. Street life became an entic- ing option for young people who either experienced episodes of ?run- ning away? from or graduating from child welfare placements: I was a runaway and started to be on the street. I liked more being on the street than at a youth centre and get brainwashed or something. . . . It?s [youth centre] like prison, it?s like a jail. (Nick, age 17, Vancouver) Many participants spoke of choosing street life. They did not de- scribe themselves as passive actors or victims of circumstance; rather, they talked about their own involvement in the street engagement process. Some spoke about being equally responsible for problematic family or child welfare experiences, whereas others saw the street as the only option when home or child welfare settings became unbearable. Still others equated street life with a ?time-out? period to reflect on their particular situation while experiencing camaraderie with other youth in similar situations. In the end, whatever the reasons for street engage- ment, young people are active participants in making transitions and building street identities (e.g., Green, 1998; Karabanow, 2004a). The Process of Disengagement The findings of this study suggest that the exiting process for the majori- ty of street youth is made up of layers or dimensions of various activities (see Figure 1). These layers are by no means mutually exclusive, nor are they meant to portray a purely linear path. Rather, the vast majority of youth participants described repeated attempts (on average, about six tries) at street disengagement. This study highlights the significant ele- ments and characteristics commonly experienced by those who have at- tempted to move out of homelessness. Contemplation Layer 1 includes precipitating factors that initiate thinking of street dis- engagement. In general, street youth recontemplate their street careers in the face of traumatic street experiences (such episodes included physical and sexual assault, drug and/or alcohol overdoses, involvement with the criminal justice system, and the witnessing of street violence); addressing their disenchantment with street culture; and/or experiencing grave boredom with street survival activities: I was in Montreal and a lot of really bad stuff happened. . . . I went in- sane, like, my last 5 days in Montreal, I stayed at the Bunker [youth se r- vice], walked around, I didn?t even do drugs. . . . It was either like, man, I?m going to let this guy take control . . . and go do smack and just die on Mount Royal [area in Montreal], or go home [to Halifax], and I went home and I think it?s like the best decision I?ve ever made. (Heidi, age 19, Halifax) Figure 1 I looked at my life and realized, where am I going? I wasn?t happy with how things were, so I decided to try and change it. . . . I was, like, I can?t do this anymore. I can?t just do nothing. I?m going to have to make a change. (William, age 20, Toronto) The freedom that had initially attracted young people to the streets (or pulled them away from other problematic situations) grew into aim- lessness and boredom, and the result was a desire for something more. Interwoven within this notion is the struggle young people face with day-to-day street survival ? securing shelter, finding money, seeking food and clothing, and staying safe: I mean, everything gets boring after a while. . . . Just really bored sitting on the street asking for money or trying to shine shoes or read poetry or whatever, you know, I?m just really tired of it, so it?s like, I?m going to get a job and get off the streets for a while because it?s boring. . . . I?m tired of this, you know? (Roger, age 21, Halifax) For other youth, heavy drug and alcohol use combined with grow- ing older wore them down both physically and mentally, to the point where they decided to make a change: Now, I?m just trying to get the fuck out of this city because it?s starting to, like, eat me alive and the drug thing is, like, too much. (Jordan, age 21, Vancouver) Although the majority of youth cite boredom, fatigue, heavy drug and alcohol use, and growing older as impetuses to exiting the street, the following narrative describes how numerous young people arrived at a point at which they no longer perceived street life as viable, without any particular reason or explanation beyond ?something clicking? in their heads or ?enough was enough?: I?m proud of myself. That was after a year and a half of using. I finally de- cided that enough was enough and I did a 28-day program [detoxific a- tion]. . . . One night, I said, ?Enough is enough,? and I went into the rehab program. (Daniel, age 22, Calgary) Supporting the claim that exiting the street is a complex process, numerous young people are unable to accurately detail or explain their particular exiting process. As such, it appears that street exiting involves tangible or perceptible paths as well as intangible or elusive dimensions. A service provider postulated that although it is virtually impossible to truly decipher the reason that these youth made such choices when they did, the simple explanation they usually provided is that they were ?fi- nally prepared?: I have seen some kids, like, it amazes me, they?ll be in it for five years and then boom, one day [they?re off], and then I always ask them what made that difference and they?re just like, ?I was ready.? It?s always a simple answer, I was ready. I was just ready. So, I think, a lot of times it has to come deep from within them about being at their breaking point or what- ever it is for them then. But yeah, then some people just never hit that and then, like, why is it that there are people that never get to that place? I don?t know, that?s a question I always ask myself, what makes that differ- ence? (service provider, Vancouver) What becomes evident is that exiting street life is a challenging and non-linear process. The decision to disengage from the street is intricate and demands varying degrees of courage. The themes outlined above are presented as the most common avenues to youth deciding to get off the street and do not represent the sole reasons for street disengagement. Furthermore, even once the decision to disembark has been made, young people continue to face many barriers to becoming an ?ex-street youth.? Motivation to Change Layer 2 involves mustering the courage to change, which tends to be heightened through increased responsibilities (such as becoming preg- nant or having an intimate partner); gaining support through family and friends; having an awareness that someone cares for them; and building personal motivation and commitment toward changing one?s lifestyle. With street culture commonly described by participants as ?exploit- ative,? ?uncaring,? ?ruthless,? and ?dangerous,? it is remarkable that some young people maintain a sense of hope for a better future, which can inspire motivation to change one?s lifestyle: I think drive has a lot to do with it, seeing hope. Some people have been hurt so much that they don?t think anything good will ever come out of anything, so why try? (service provider, Toronto) Most young people spoke of needing a ?desire? to exit street life or having ?motivation? and ?strong will power? to combat impressive ob- stacles such as drug addictions, personal trauma, lack of housing and employment, and few support mechanisms: Mostly, the only resource that will get the person off the street is the per- son themselves. They have to [want] to get off, they have to be wanting something. They want to be able to grasp something. If they don?t want to grasp anything or want to move on, they?re not going to move on. They have to have the willpower to do it. (Randall, age 20, Toronto) Findings suggest that young people who believed they had support from family or friends or that there was someone in their lives who truly cared for them were able to build motivation for street disengagement: The way [I got off] is . . . since I had a job, my mom said that I could stay with her for a couple of weeks. Then when my boyfriend got out of jail, my mom let us stay with her for another couple of weeks. We were pay- ing her rent and buying food and all that stuff. That?s how we got off the street, I guess, with the help from my mom and working. (Heather, age 23, Calgary) Just knowing that somebody cares and you have that extra support and they want to see you succeed. I think that?s really important for kids to understand that there are people out there that care. Like, sometimes they don?t have that support from their families but at least there?s, like, re- sources that they can go to where they really can get help and whatnot. (Joanna, age 17, Toronto) Within street culture, asking for and seeking help proved to be a struggle for the majority of participants, but, at the same time, an integral part of the disengagement process: What didn?t work was doing it on my own and relying on my friends that were in the same position because, I mean, it?s a cycle and you just get dragged back into it again and again if you don?t have outside help. (Ahmed, age 23, Vancouver) Participants stressed the liberating quality associated with overcom- ing one?s reluctance to ask for or accept help: I kind of had a problem with my pride, where I didn?t want help. I thought I should have to do things for myself because I thought that my situation was my fault. But I kind of had to take my pride and put it in my back pocket and take some help. . . . I?m paranoid that I?m always be- ing a burden, so I just had to push that aside and actually take some help. And actually, it only took the one time for my buddy to say, ?You know, why don?t you stay here?? So I did. (Charles, age 20, Halifax) In addition, for some young people, having a child or being in a se- rious romantic relationship was an important motivator for making changes in their lives. Recognizing that someone else was depending on them helped to increase their sense of self- worth: Getting pregnant got me to think about it. It was someone else to be re- sponsible for, so it increased my determination to get off. . . . To get off, you really have to want to change your life. (Lindsey, age 20, Toronto) [My boyfriend] always felt kind of bad because he always sort of blamed himself for me [running away from home to live with him on the streets]. So he was always like, okay, I?ve got to get into shape, I?ve got to find an apartment so that she can be happy. (Rose, age 17, Toronto) The motivation for youth to move away from the street inevitably rests on a multitude of internal and external factors. Youth participants often spoke of having to overcome personal barriers, such as lacking in- ner drive and motivation, a bruised sense of self, uncertainty about out- side passions and interests, and an inability to ask for help. At times, young people can overcome such obstacles on their own, but the majority suggest an urgent need for support and guidance. Of- ten, it was enough to feel responsible for another person or to know that somebody believed in them and would be supportive even in failure. Youth with strong personal support systems tended to demonstrate few- er struggles with street disengagement. However, none of these dimen- sions is mutually exclusive, and they often intertwine and intersect with one another. Young people with a strong desire to get off the street may be more willing to ask for help. Feeling responsible for a new baby might inspire a young mother to leave street life. These motivations and supports may change throughout a youth?s exiting process, and the study?s findings suggest that without continued forms of support, a return to homeless- ness is most probable. Furthermore, even with strong supports in place and keen motivation, youth face numerous hurdles in their move(s) away from the street. Securing Help Layer 3 involves seeking support for the initial stages of getting off the street. This layer tends to include the use of available services; searching for formal employment and stable housing; and some form of formal institutional involvement (such as returning to school or entering sup- portive housing or structured program entities). Within this layer, it be- came evident that service providers play a significant role in supporting young people to regain or rebuild a sense of self. Most participants de- scribed diverse service provisions as ?surrogate families? and ?brokers? between street culture and mainstream living. Young people on the street struggle with numerous interrelated is- sues. Within a culture of personal and environmental trauma, street youth deal with daily survival; experience physical, mental, and spiritual health concerns; maintain a lack of life and employment skills; and have little in terms of what Jacqueline Wiseman (1970) termed ?social mar- gin.? Service providers such as shelters, drop-in centres, health clinics, second-stage independent living resources, mobile care units, and out- reach programs not only provide basic needs (such as food, clothing, shower facilities, and shelter) and life and employment skills training (such as how to manage a budget, cook, search for employment, and car- ry out a job interview), but often forge community spaces where young people can regain confidence and self-esteem within a ?culture of hope? (Karabanow, 2003, 2004b). Service providers have been credited by participants for support in seeking employment possibilities, housing options, and educational op- portunities within an environment of care, safety, and learning. Many organizational structures even succeed in carving out community envi- ronments where young people can regain a sense of self, begin to work out personal dilemmas, build a critical consciousness as to why they are on the street, join or initiate advocacy strategies to fight structural injus- tices that maintain their homeless status (such as a lack of affordable housing or meaningful employment opportunities for youth), and reins- tall a sense of hope and a better future. For the majority of participants, not being judged for their homeless status and feeling as if someone un- derstands and empathizes with their struggles are key ingredients to service delivery satisfaction and engagement. Transitioning From the Street Layer 4 deals with transitioning away from the street and, in the study, proved to be a complex and difficult stage of street disengagement. Mov- ing away from the street entails physically leaving the downtown core, reducing ties with street culture and street friends, and constructing (or reconstructing) relationships with mainstream society. Cutting street ties meant leaving friends, surrogate families, and a culture associated with the downtown core. For many young people, friends and surrogate families were forged as a result of, or during, very stressful survival situations. Survival is the paramount objective on the street, and many young people join or de- velop tight-knit community bonds with other street colleagues. The data elicited a strong positive relationship between the length of time on the street and the difficulty of leaving the downtown core/friends ? the long- er on the street, the deeper the relationships one would have to the street and the harder it would be to disconnect from street culture: But it kind of compounds itself ? the longer you?re on the street, the hard- er it is to get off because you get more entrenched in the culture and you have more of the problems that come with that. (service provider, Cal- gary) Participants spoke about the street lifestyle as more than a physical space and associated leaving the street with disconnecting from friends. Breaking ties with street-involved peers was different for each youth but was generally seen as a slow and gradual process. The majority of youth stressed how disconnecting from friends who they perceived as a bad influence was an essential part of the exiting process: Most of them come by and ask me, ?Could you help me for two days, like sleep at your house?? I don?t have the choice [but] to say no, because if I help them, they?ll come back and see me and they won?t help themselves, and since I need to help myself first of all, I don?t have a choice either. (Mohamad, age 23, Montreal) Breaking ties with friends and drugs was highly intertwined. Ad- dictions were described as interwoven into the fabric of street culture and street families. Youth who had moved into a more stable living envi- ronment spoke about the difficulty of dealing with their drug addictions. Youth also expressed that ending drug or alcohol misuse was a signifi- cant step in getting off the street and helped to improve self-esteem Yeah, it helped quite a bit and made me feel a lot better about myself. I think that was the biggest thing. It?s all about really not using drugs. I think it?s a lot about how you feel about yourself. If you feel good about yourself, then you don?t really need the drugs. (Chester, age 24, Calgary) Youth openly commented about the difficulties and challenges of leaving behind street friends, often exploring feelings of confusion, guilt, abandonment, disloyalty, resentment, and loneliness. For some young people, street friends and street families were communities where they experienced security, acceptance, and love, often for the first time in their young lives. Although the majority of participants agreed that breaking street ties was necessary to becoming more stable, it was undeniable that the process and the actions associated with breaking such ties were emo- tionally difficult: I found my biggest [obstacle] was leaving the crowd that I was with, like my friends, the situation with my friends, because they were all like, ?No, don?t go, stay down here and hang with us, go do this and go do that,? and that was probably my biggest crutch, was getting away from my friends because I?d been friends with them my whole life, and for me to just push them away and just say, ?No, I?m getting away from this, I?m getting out of this.? It was a big step for me. (Chris, age 21, Calgary) Although participants were clear that breaking ties with street cul- ture and friends was essential to the transitioning stage, it is also clear from the findings that a majority of young people re-enter street life to visit street friends and street communities, interact with street youth or- ganizations (predominantly located in downtown areas), and supple- ment (primarily through panhandling and squeegeeing) their often mea- gre minimum-wage earnings from formal sector employment. It is not surprising that young people leaving the street experienced mixed feelings; there were unmistakable feelings of pride, hope, and self- confidence coupled with deep emotions such as loneliness, guilt, and disloyalty. Such confusion was typically directed towards street culture and friends; nevertheless, there were also comments made about service providers, who were commonly perceived as surrogate parents. For some youth, moving away from street culture also entailed breaking ties with the services that had supported them: Because if I have to go downtown even for a few minor services, it still puts me in that scene and makes me, like, face-to-face with a lot of stuff that I don?t need to be involved with. (Jay, age 23, Vancouver) Other young people continued to use services, however, in a more strategic manner (such as meeting staff when residents were sleeping or out of the establishment) or more focused manner (linking with services that maintain one?s stability and distance from street culture). Participants said that it was as difficult to leave street culture and street friends as it was to enter mainstream society and build new rela- tionships. Despite the emotional strains of leaving relationships with people who had helped support them on the street, building new rela- tionships outside street culture was highlighted as essential for a healthy transition. New friends and communities tended to be seen by partici- pants as ?good influences? in their day-to-day living: I think it?s having a network of people outside of street life. Because I mean, when you?re on the street, your whole world, your whole family, everyone you spend time with, everyone that you see is pretty much out and about here. But once you?re off the street, your friends have places to live, you know? (Barb, age 22, Ottawa) Youth expressed that the transition period between leaving street friends and developing new relationships was difficult. They often spoke of feelings of loneliness and uncertainty: I think it?s really hard because I?m, like, in between right now because a lot of my friends still live street lives. They?re all about partying and pan- ning and I?m just not. So I guess it?s kind of a lonely time because you?re figuring out yourself and what you want to do. (Heidi, age 19, Halifax) Changing Routine Layer 5 involves restructuring of one?s routine in terms of employment, education, and housing; a shift in thinking about future aspirations; and acquiring some form of social assistance to support one?s transition. Dur- ing this stage, young people highlighted a renewed sense of health and wellness, self-confidence, and personal motivation. A sense of changing routine emerged for participants as they made the transition from living on the streets to mainstream society. Partici- pants described both physical and psychological shifts occurring in their lives, such as sleeping better, feeling healthier, and experiencing in- creased self-esteem and self-confidence. Such changes tended to be linked to young people having more stability and consistency in their lives. Shifts in routine were commonly seen as interwoven with the no- tion of building new communities and tended to focus on replacing street activities with formal employment and returning to school. How- ever, subtle day-to-day shifts in routine (such as waking up and making some coffee or coming home and watching television) were as celebrated as more tangible elements (such as living in one?s own apartment or going to work each day). According to young people, the most consequential change came with employment. More than simply providing for basic needs, work translated into a gradual shifting in general lifestyle. Such changes gen- erally involved the way participants managed time (work and free time) and perceived their future: I can just compare my old lifestyle, where I would wake up in the morn- ing, if I found someone?s house I could crash at, definitely take a shower if that was available, usually didn?t have any clean clothes to put on, so I?d maybe try to rinse the ones I?d worn the day before out. Do my best to find something to get stoned on and go out into the world and bum change from people. Well, now? I work nights, so I don?t wake up in the morning but I wake up, I have my shower, I get something to eat. I?m tak- en care of, I?m happy, I?m fed, and I go to work. It makes me feel mean- ingful about what I do with my day and so, I go out and I?m able to give to the world instead of just trying to take for myself, which is an amazing- ly positive feeling. And I can pursue the things that make me mentally healthy. The depression that goes with the street life isn?t there, the feel- ing that I?m less than? My old idea of intellectual pursuit was dropping acid and talking about this and that. Now, I can read a novel and write a poem? I have all these options to me that I can go and take the time to do these things. Some of the differences I don?t even notice because they?re so... it seems so normal now. (Ahmed, age 23, Vancouver) Reintegrating into mainstream culture introduced young people to a new way to live their lives, and much of their new structure came from work or school. Participants experienced routine changes in most aspects of their day-to-day lives, from sleep habits to eating arrangements and free-time pursuits. These transitions allowed many youth to reflect on their past experiences, and for the majority of participants, this meant perceiving the street as an unhealthy and destructive environment. Along with a healthier sense of self, young people were more ready to develop longer-term plans and envision some control in their futures: Now, I wake up and I have something to live for ? before, I didn?t have anything to live for, really. . . But now, it?s like, okay, I have a son to take care of and I have myself to take care of. My mornings are amazing be- cause it?s just like getting stuff together and going somewhere. Before, I didn?t have anywhere to go, it was just like bouncing from mall to mall or shelter to shelter. Now it?s just like, I get up, I go to baby-and-mom pro- grams or we go to the library where they have the mom-and-baby read- ing sessions and it?s just like a wonderful, wonderful thing for me now. (Cynthia, age 20, Toronto) ?Successful? Exiting The final stage has been termed ?successful? exiting, which embodies young people?s emotional and spiritual sense of identity. Successful exit- ing was exemplified by a sense of ?being in control? and ?having direc- tion? in one?s life. The majority of participants spoke of feeling proud of their movements out of street life; being able to finally enjoy life on their own terms; healthy self-esteem and self-confidence; being able to take care of themselves; and feeling stable in terms of housing security and wellness. Youth described a variety of concepts when discussing what it meant to successfully make transition away from street culture. Getting off the street translated into more than simply finding an apartment and physically removing oneself from a street lifestyle. Truly becoming an ex?street youth entailed emotional and spiritual shifts within the indi- vidual. Many young people described success as involving stability and being comfortable in their living environment. Youth spoke of feeling ?self-sufficient,? ?stable,? ?being able to take care of themselves,? and ?being in control? of their lives. Often, success was equated with feelings of self-sufficiency. This translated into not having a need for street youth services or relying on social assistance benefits for support. For youth currently living on the street, they pondered what success would look like for them in the fu- ture, and many concurred that it would entail reducing perceived de- pendency on services: Well, to be self-sustaining, you know, to at least be able to come up with my own food money, spend it on food and, you know, pay rent. (Danny, age 22, Calgary) It is not surprising that youth often described obtaining housing, employment, and education as successful exiting: Successfully getting off the streets is getting your own apartment, having a very successful job, avoiding street life like not panning, not having to fly a sign or go squeegeeing or anything like that. (Roger, age 21, Halifax) Other young people expanded on these dimensions and suggested that rather than simply being housed and fed, they desired a sense of ?home? and ?stability?: I have a home. I don?t have to worry about weather. I don?t have to worry about ? I mean, I?m a woman ? so I don?t have to worry about being as- saulted or stuff like that. Like just things that people don?t even think of. Like I don?t have to worry about where my next meal is coming from or how I?m going to get heat or hot water or the embarrassment of going somewhere. (Patricia, age 21, Halifax) Leaving dangerous street activities (such as drug abuse and sex trade work) was also noted as a measure of success and stability: I?m not out doing drugs downtown. I?m not hanging with the street kids. I?m not stealing, keeping myself out of jail, not partying or pimping. (Chris, age 21, Calgary) Participants also cited positive feelings, emotions, and relationships when discussing the concept of success. For some youth, success was defined as a spiritual state of being ? an emotion or feeling that provided a renewed sense of self: I think success is a peace of mind. It?s being able to sit down at the end of the day and feel satisfied with what I?ve done, with who I am and to live life to its fullest. Every minute is a success. That?s where I want to be. I?m getting there. (Dana, age 18, Vancouver) In all, successfully exiting street life incorporated various dimen- sions made up of both tangible and intangible constructs. For almost all participants, becoming an ex?street youth required stable housing, a re- turn to employment and/or school, and a move away from street culture and activity. Other young people, especially those who had left the street, supplement these comments with notions of spiritual and emo- tional growth and stability. Conclusion Street youth exist within excluded realms. They are a traumatized popu- lation located outside the formal market economy. They describe expe- riences of marginalization and stigmatization within civil society, are continually kept under surveillance and harassed by social control agents and members of civil society. In their situation of ?being home- less,? they are poor and isolated, have little social capital and social mar- gin, appear ?different? in looks and attire, have difficulty locating em- ployment and shelter, and spend much of their existence in the public arena, concerned with basic survival needs such as shelter, food, cloth- ing, and social support. As one young person suggested, Like you don?t feel right in your skin yet, like you?re not really a success- ful member of society quite yet, but you?re not panhandling on the corner, right? It would look bad if I went out and panhandled now, right? But on the other hand, you know, you don?t have any money and what are you supposed to do? (Heather, age 23, Calgary) Everything about being young and homeless inspires critical and of- ten demeaning responses from others in mainstream society. Within each of the stages of exiting, young people spoke about so- cial exclusion. For example, attempting to secure housing options and employment opportunities proved extremely difficult and often demean- ing. As one young person noted: Who wants to give me a job? I look like a homeless kid. I am a homeless kid. (John, age 20, Vancouver). Each stage of exiting intersected with numerous challenges and ob- stacles, making successful exiting difficult and often including numerous trials. Re-entering mainstream culture proved the most difficult dimen- sion, as young people were required to make the transition from ?identi- ties of exclusion? (i.e., being different, feeling stigmatized and margina- lized) to one of ?fitting in? to mainstream lifestyles. But there are also signs of inclusionary dimensions within the street youth populations. The majority of street youth spoke of street life as a safer space than their previous environments, suggesting the traumatic or horrific experiences that led young people to the street. There is also evidence that street life can provide feelings of community and family for many inhabitants, a space where some do feel cared for, accepted, and even protected. Moreover, findings suggest that for the most part, street youth services act as surrogate families for homeless youth, pro- viding needed basic amenities and safe and caring environments. It is precisely these characteristics of inclusion that make it difficult for most young people to move away from street culture. When asked about their plans for the future, about dreams and hopes, most young people indicate a great desire to belong, have a fami- ly, find a loving partner, seek meaningful employment, accrue a safe place to live, and be part of civil society. And although their current lives are chaotic, unhealthy, and distressed, they hoped for a brighter future. This finding provides direction for how we as a society should construct meaningful approaches to build a culture of hope and inclusion. There are important avenues that we should embark on immedi- ately; the first two recommendations come directly from participants, and the others from reflections from the data in general: 1. Invest in existing frontline (?in the trenches?) support ? shelters, drop-ins, health clinics, and outreach services. They are the first supportive and healthy adult contacts that most young people expe- rience when living on the street and offer creative and compassio- nate responses to basic needs. The majority of youth participants spoke eloquently and passionately about the significance of such re- sources throughout the street exiting process. 2. Forge thoughtful long-term structural development initiatives, in- cluding supportive and independent housing and meaningful em- ployment opportunities. There are many examples throughout North America of innovative linkages between government, busi- ness, and non-profit sectors to build such initiatives (e.g., Montreal?s Dans La Rue, Toronto?s Covenant House and Eva?s Place, and Cal- gary?s Open Door). Young people in the sample were unequivocal about the need for safe and sustainable housing in order to seek out employment opportunities. 3. Enhance social action campaigns that speak to and on behalf of this marginalized group. Examples include fighting against legislation that targets young people as criminals (such as Safe Streets legisla- tion); youth groups that provide consciousness-raising alternatives; attempts to increase per diem rates for service provision operations; and opposing police harassment and abuse (Karabanow, 2004a). 4. Initiate preventive structures that tap into the true reasons for youth homelessness: child welfare failures, poverty, family distress, abuse, neglect, and violence. We need thoughtful educational strategies (such as runaway prevention programs carried out by street youth organizations) to disentangle myths and stereotypes about why these young people enter street life, remain on the street, and suffer. 5. Build national and regional coalitions of street youth, policy makers, service providers, housing specialists, and academics that can share best practice approaches on service delivery, policy development, education, advocacy, and voice. Such distinct yet interwoven dimensions will provide our young people with the proper support and a fighting chance to climb out of homelessness and, equally significant, provide opportunities for them to become citizens rather than clients, victims, criminals, or worse ? invisi- ble and insignificant bodies. Jeff Karabanow, Ph .D., is a full professor at the School of Social Work and cross-appointed with International Development Studies and Health and Hu- man Performance at Dalhousie University. His teaching interests involve com- munity development, social policy, research methodology, organizational theory, and international social work. He has worked with street youth in Toronto, Montreal, Halifax, and Guatemala and published articles about street youth culture. References Alleva, F. (1988). Youth at risk, systems in crisis: A dialogue with youth who needed shelter. Unpublished doctoral dissertation, Boston University. Charmaz, K. (2004). Grounded theory: Objectivist and constructivist methods. In N. K. Denzin, & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp. 249- 291). Thousand Oaks, CA: Sage. Edelbrock, C. (1980). Running away from home: Incidence and correlates among children and youth referred for mental health services. Journal of Family Is- sues, 1(2), 210-228. Ensign, J. (1998). Health issues of homeless youth. Journal of Social Distress and the Homeless, 7(3), 159-171. Green, D. (1998). Hidden lives: Voices of children in Latin America and the Caribbean. Toronto: Between the Lines Press. Karabanow, J. (2003). Creating a culture of hope: Lessons from street children agencies in Canada and Guatemala. International Social Work, 46(3), 369-386. Karabanow, J. (2004a). Being young and homeless: Understanding how youth enter and exit street life. New York: Peter Lang. Karabanow, J. (2004b). Making organizations work: Exploring characteristics of anti-oppressive organizational structures in street youth shelters. Journal of Social Work, 4(1), 47-60. Kufeldt, K., & Nimmo, M. (1987). Youth on the street. Child Abuse and Neglect, 11(4), 531-543. Kurtz, P. D., Jarvis, S. V., & Kurtz, G. L. (1991). Problems of homeless youth. So- cial Work, 36(4), 309-314. McCarthy, W. (1990). Life on the streets. Unpublished doctoral dissertation, Uni- versity of Toronto. Michaud, M. (1989). Dead end: Homeless teenagers, a multi-service approach. Calgary: Detselig Enterprise. Morrissette, P., & McIntyre, S. (1989). Homeless youth in residential care. Social Casework, 20, 165-188. Strauss, A., & Corbin, J. (1990). The basics of qualitative research: Grounded theory procedures and techniques. Newbury Park, CA: Sage. Wiseman, J. (1970). Stations of the lost: The treatment of skid row alcoholics. Chicago: University of Chicago Press. Chapter 3.7 The Peel Youth Village: Designing Transitional Housing for Suburban Homeless Youth RAE BRIDGMAN This case study presents findings from research about the preliminary design and development of Peel Youth Village, a transitional housing project designed for suburban homeless youth in the Region of Peel, a suburb of Toronto. Of particular interest are how planning and design processes and decisions come to be negotiated and re-negotiated on the part of all players, according to a complex mix of user needs, service providers? needs, funding mandates, budgetary constraints, site condi- tions, building code regulations, social values, and political will, among many other issues. Projects evolve within a complex of all these com- plementary and at times potentially conflicting elements. In this article, I concentrate on insights arising from several focus groups with homeless youth, as well as the impact of political processes, building codes and planning approval processes on the design of the facility. The article closes with recommendations for design interventions to address homeless young people?s needs, and highlights the value of youth participation in any project designed for their needs. Also high- lighted is the value of project documentation and dissemination to offer concrete guidance about the strengths and weaknesses inherent in vari- ous design approaches. Background While youth homelessness in the United States has become the subject of a growing body of literature, especially in the fields of social work, psy- chology, and medical health, and homeless young people have also been the focus of a great deal of work in Britain, particularly by Susan Hutson and Suzanne Fitzpatrick (see, for example, Fitzpatrick, 2000; Liddiard and Hutson, 1991), the literature on youth homelessness in Canada is comparatively recent. (See Kraus et al., 2001 for a national overview and annotated bibliography.) Emphasized in the Canadian literature on best practices for alleviating homelessness is the degree to which homeless persons, as well as front-line workers, are involved in developing solu- tions, together with the empowerment of homeless persons to access sta- ble housing and services, develop skills and actively pursue the goal of independence (Canada Mortgage and Housing Corporation, 1999). My ethnographic research has revolved around documenting inno- vative projects for alleviating chronic homelessness among single women and men in Toronto (e.g., see Bridgman, 1998, 2003, 2006). This work involves an extended commitment to follow a project through its development over several years. The researcher in effect becomes ?an- chored? to innovative housing projects to document their development, to expose the common objectives yet different perspectives involved in bringing a project to fruition, and to consider how what has been learned ?here? may potentially be applied ?there? (Bridgman, 1998: 12). Such research contributes to knowledge-building around what has been called ?utopian pragmatics.? ?Utopian pragmatics? refers to the study of how initiatives meant to disrupt and redress existing (oppres- sive) social conditions ? in this instance, youth homelessness ? actually get implemented (Bridgman, 1998). How are such alternative visions made real? My research has also involved documenting a housing and em- ployment training program for homeless youth in downtown Toronto. One of the articles arising from this work explores several chal- lenges to an organization?s capacity to develop an innovative project for homeless youth, and the degree to which homeless youth are able to be involved in decision-making (Bridgman, 2001). A second article (Bridgman, 2004) theorizes public-private partner- ship processes and offers a theoretical framework for charting the com- plexities of coordinating responsibilities, decision-making, and account- ability in developing housing for homeless young people. This article explores the degrees of youth participation during early development processes for Peel Youth Village, based on a participatory model devel- oped by Roger Hart (1997). Research Methods My research involved extended participant observation from January 2000 to June 2002 to document discussions among architects, develop- ment consultants, local social service providers, municipal officials and homeless youth. Meetings of the Peel Region Homeless Youth Task Force (subsequently renamed the Homeless Youth Network for the Region of Peel) involved representatives from approximately 45 youth-serving agencies ? including mental health services, advocates, boards of educa- tion, drop-in centres, sexual health clinics, drug abuse counselling, criminal justice counselling, police services and faith-based organiza- tions. The Homeless Youth Network Working Group (with approxi- mately 12 members) also took place each month. While drawing in part on discussions from these large group meet- ings, this article concentrates on insights arising from a housing design charrette (November 2000), three focus groups held with homeless youth (March/April 2001), and a series of design meetings held between Janu- ary and June 2002, as well as an in-depth interview with one of the pro- ject?s architects. Excerpts from the field notes and minutes of meetings have been cited. Care has been taken to change names and identifying details in order to maintain confidentiality for those involved in develop- ing the Peel Youth Village. Peel Youth Village The report of the Peel Regional Task Force on Homelessness to Regional Council in May 1999 identified a need for a task force on homeless youth. The Task Force on Youth Homelessness started meeting in Janu- ary 2000, and plans for a transitional housing project for homeless youth began to coalesce. The Region of Peel, just northwest of Toronto, is the second-largest municipality in Ontario, and in 2000, its population was about 1 million. At the time of writing this article, the Region had 18 shelter beds for youth. Of these, six were designated long-term, and young people could stay for up to a year and develop their life skills. The remaining 12 pro- vided emergency shelter, with a three-week limit. (In contrast, the City of Toronto with a then population of 2.5 million had more than 400 hostel beds for youth, with a stay limit of three months.) With funding of approx. $4 million (Canadian) from mult iple levels of government, Peel Youth Village was attempting to address the needs of the suburban homeless, who had only recently been recognized in the literature (see Crane & Takahashi, 1998). Key components of Peel Youth Village were to include housing to accommodate approximately 64 young people, assistance with employ- ment opportunities involving a range of private and public partnerships, and involvement of homeless youth in the development, construction, and management of the project. At the time of my research, design draw- ings had almost been completed, working drawings were to start soon, and construction was slated for the summer of 2003. The metaphor of a ?village? suggests not just a housing project physically bounded within its own community; rather it suggests a con- cept beyond the mere walls of one building, one inclusive of the sur- rounding neighbourhood community. The concept of ?villaging? is meant to create ?an environment that allows and encourages a health community to evolve,? according to discussions during a retreat in Janu- ary 2002 with representatives from approximately 20 youth-serving agencies in the Peel region. The Peel Youth Village model was developed in parallel with con- sultations with youth, and after network agencies participated in a self- reflection process to discuss with other agencies the areas in which they were able to help at-risk youth, and the areas in which they were not as effective. All the agency reports highlighted the lack of resources to offer badly needed services. Overall, service providers in the Peel Region were becoming increasingly concerned that young people were migrating to the City of Toronto because they were unable to access supportive ser- vices in their own home communities. Not only did the Peel Region have an acute shortage of beds for homeless youth, but youth with histories of criminal behaviour, addictions, and mental illness were often excluded from local services. And those youth who secured one of the few beds available were expected to move on quickly. Once youth left the Peel Region, ?going back? became more difficult due to distance and transportation issues, and youth could lose their so- cial networks of support (family and friends). Design Decisions Insights from Youth During March and April 2001, three youth discussion forums were held at local drop-ins and youth centres in malls and elsewhere, with pizza and pop served. Approximately fifteen attended the first group, and six to eight youth attended the other two. The youth talked about physical design issues as well as program design (e.g., expectations and rules). The richest insights were gained from the third discussion group, in which the youth had all experienced or were presently experiencing homelessness. Youth in the other two groups held at different agencies had not experienced homelessness to the same degree. Many seemed to hang out at the youth centres, but had a home to go to when the centres closed. Following are just a few of the many ideas gathered, particularly from the third discussion group. While these excerpts from field notes are ordered under topic headings, the words of the youth themselves offer powerful testimonies. Lack of Shelters One young woman named Moira spoke up right away at the beginning of the meeting: ?There are not enough shelters in Brampton. Most of the time you have to travel to Toronto or Mississauga. And if you are not old enough to get to Toronto on your own then you have nowhere to go. There is OPP [Our Place Peel] but the waiting list there is forever just to get in.? Greg added, ?I have a whole list of shelters in the Toronto area right here [he pulled out three or four pages] and they are either full, or they want to know if you have a criminal record, or you can?t stay there be- cause of your age or something.? Shawn said, ?I?m not allowed to stay at the shelter with my Dad be- cause it?s a men?s shelter, and I?m still under 16, so I had to sneak in so that I could stay with him.? Greg added, ?Or there will only be two beds for five people, and you have to do rock-paper-scissors to see who gets the bed and who sleeps on the floor.? Time Frames Almost all those consulted agree that people staying at Peel Youth Vil- lage should move on. Peel Youth Village should be a place to help youth get going; it should not be a place where people would be living forever, or even for a great length of time. At the same time, however, the youth emphasized the stress created by having strict time limits for moving on. The length of stay should be determined on an individual basis. It was important not to have strict time limits because as one person noted, ?That can really freak people out, if they are always counting the days until they will have to move again.? The Drop-In Community Centre and Residences The architects asked the youth about how the drop-in community centre and residential area should be connected ? whether they should be physically separate or in the same building. Two of youth spoke simultaneously: ?They should be separate so that people from the community centre can?t get upstairs to the bed- rooms. Have separate entrances.? As Greg said, ?You might want to have friends in the recreation room who you don?t necessarily want to know that you are living there. It?s a privacy thing.? The need to keep homeless youth safe from others in the commu- nity turns on its head the usual concerns of the public-at-large to keep the community safe from homeless people. Accommodating Couples Moira suggested, ?You should have special rooms for couples so that they can have their own room. I think this is a big problem right now, because most couples are not allowed in shelters. They are either all males or all females. And even if they are, both of you are not allowed to share a room.? Sue chimed in: ?It can be really bad. Like you are not even allowed to show affection. There is absolutely no touching.? You could have separate units maybe with five other people or so. But no more than five because they might not get along, or someone might not take care of themselves in terms of hygiene, and that would create some problems. Plus if you have more than five people waiting to use the bathroom, that could be a problem.? Later she explained: ?Me and Greg are a couple and it is hard because I am pregnant and I get really tired and moody and stuff, and at times like that I don?t want to be alone. I want him to be there for me, you know. You want to be with your partner.? Some of the insights from the youth consultation, as one of the ar- chitects suggested during an interview in July 2002, actually changed the entire project. ?It really would have been a straight housing project if it hadn?t been for [those meetings in the malls] . We sort of threw out the whole idea we started from and said, ? Okay, so what has everyone told us?? And everyone was saying it wasn?t about housing. They need play space, places to be and to do things in.? It is a community project ? the whole community centre was really what everybody wanted and it still is. The housing is great for street youth and all the homeless kids at risk ? all of that is fantastic, but I think it is the combination of the two pro- jects that is the most interesting.? C C Impact of Political Processes, Building Codes, and Planning Regulations There are many interconnected levels of negotiation involved in design and development processes. The original plans for Peel Youth Village included several compo- nents ? a clubhouse/drop-in, an emergency shelter (15-30 youth), an en- try-level shelter (25-30 youth), and transitional shelter (25 youth) tar- geted to youth. Permanent housing (25 youth) was also part of original plans. Federal funding mandates precluded supporting the category ?permanent housing,? however. As is often the case, funding mandates can have a profound impact on the nature of what is considered ?fund- able.? A Network member put it bluntly: ?We cannot call this piece of the project ?permanent housing.? We have to call it ?transitional housing? to qualify for HRDC money, but ideally we would want it to be a place where kids can stay. Our private aspiration is to have the people living here making it a different thing. But for expediency, permanency will not be part of our language.? As plans developed, the project became a four-storey building with a fully developed basement below grade. The basement was given over to a youth development centre, or community centre (with gymnasium and recreational games). The second, third, and fourth floors were for residential use (total of 12 four-bedroom units) ? with the second floor functioning in a short-stay capacity much like a dormitory (16 rooms, with two people per room), and the top two floors for longer-term resi- dents (16 rooms on each floor). The ground floor was to function as a lounge or town hall to the youth development centre, and would act as a mediating space between the basement and the upper residential floors. Each floor would be seen and heard from any other floor as a result of the common stairs. When it came time to seek municipal approvals from the planning department, the project had to be presented, however, not as a shelter, and not as transitional or supportive housing. Supportive housing, as one Network member clarified at a design meeting, carries connotations in the Peel Region of mental health services and institutionalization. Dis- cussion at design meetings centred on the apparent contradiction that as far as the municipal planners were concerned, the project did not require any staff, and was for housing only. Yet requests for funding from Re- gional Council would be made for operational staff funding. One Network member tried to clarify her concerns during one of the design meetings: ?On the one hand we are saying this housing ? but not supportive housing. What will happen when after it opens the Mayor drops in to see what is going on, and sees all these programs in place, and learns that, in fact, it is supportive housing?? This member continued with a quip about the ?honesty angel that was sitting on her right shoulder.? Another member responded: ?Honesty is important, but so is serving the needs of these young people. So what?s more impor- tant?? Discussion continued at length with some suggesting that there were different stories for different people, and others insisting that they were the same stories just with different highlights. In a subsequent de- sign meeting, one of the members present addressed the same subject: ?Architects have different sets of drawings for different purposes. Maybe this is audiences, and we use different words for different audiences.? Another member emphasized: ?We must avoid terminology like ?support? and ?care? because it implies mental health in too many quar- ters. We can?t call it a gymnasium, but rather a recreation space. And we have to avoid the word ?counselling? or any terminology that smacks of business. This has to come off as a housing project. Anything that smacks of a ?care? facility mea ns they will go back to zoning and argue that this is not, in fact, housing.? With regional concerns over rising taxes and fiscal constraints, the Mayor of Mississauga had stated publicly early in the spring 2002 that Mississauga would not be participating in any provincial or federal housing initiatives. As a result, the Peel Youth Village presentation to City Council was an important site of negotiation. The strategy devel- oped for presenting the Peel Youth Village proposal was to first present less contentious local housing projects, such as a project for seniors, to smooth the way for Peel Youth Village?s approval. The Regional housing department, having agreed to subsidize op- erating costs of $700,000, had a great deal of input into the design as a result. The department was particularly concerned that costs be kept as low as possible, so there would be no air conditioning, and no extra ex- penses that could be perceived as extravagant. According to fire codes, individual bedrooms could not be locked. The suites themselves could have a lockable door, but the individual bedrooms could not. People could put a latch on their door, and lock it when they were in the rooms, but they were not to lock the door when they left the room. Consultations with youth, however, had pointed to the importance of being able to lock up possessions. Lockable cupboards were proposed at design meetings as one way of addressing the issue. One of the project architects reflected: ?Whenever we have talked to people in the past, being able to lock up stuff is pretty important. So now what are we saying? You can lock some of their stuff but not all of their stuff ? do you lock your sleeping bag, or do you lock your clothes, or your food? What don?t you lock up? I don?t know.? Regional fire code officials also were concerned about all the open- concept common spaces on each floor and the main stairs running straight through the middle of the building (for ease of sight lines and auditory cues). They had not reviewed a housing model designed with so much open space before, and were more comfortable with an apart- ment- or condominium-styled double-loaded corridor (units on either side of a corridor). The officials wanted these spaces to be treated like corridors, or alternatively they wanted the open spaces enclosed (e.g., in glass). These strictures resulted in some of the common areas being di- vided off into rooms. The architect expressed frustration with the approvals process: ?It is forcing us to make the decision about where the rooms go. ? We always thought that if one day they need to close off one whole floor for pro- gram space or if they wanted it for more residential, there would be in- credible flexibility. What they?re making us do is to make those decisions now and we don?t know how to make them. Because we don?t know what those rooms should be like. So the area that we were leaving for future development, we?re having to pre-determine that, and I think that is a mistake.? At the time of writing the article, it was not yet clear whether the Region or one agency (or group of agencies) would take over the actual operation of Peel Youth Village once it opened. This uncertainty also had its impact on design-related questions, as different agencies with differ- ent mandates and interests attempted to come to consensus. The site for the Peel Youth Village sat in the midst of a cluster of four high-rise non-profit apartment buildings (528 units, 1,500 residents). The majority of the families were single-parent families led by women. Approximately 700 children and youth lived in the apartments. At the time, there were no green spaces or parks nearby for the children and youth to play, and it was anticipated that Peel Youth Village would pro- vide badly needed recreational space. Within this broader neighbour- hood context, Peel Youth Village would offer much more than housing. It had a much larger mandate to extend its presence to the community- at-large through the proposed public recreation space. Importantly, the Peel Youth Village model was premised upon a continuum of housing and support options that youth would move through at their own pace. In other words, the stages proposed by the model were needs-based and not time-based. Conclusion Directions for Future Research Systematic project documentation, evaluation, and dissemination are of tremendous value for offering concrete guidance about the strengths and weaknesses inherent in various design development approaches. Analy- sis of housing development processes over extended ?real-time? leads to an appreciation of the ?shifting sands? nature of participation ? depend- ent upon funding pressures, political will, mandates of partners partici- pating in the development, municipal plans inspection and review proc- esses, and other such externalities. More systematic documentation of project development processes is badly needed. This kind of research is not promoted enough. Lessons learned during implementation may have application elsewhere. Docu- mentation and dissemination can offer concrete guidance about the strengths and weaknesses inherent in various approaches. Unfortunately, non-profit organizations can rarely commission such research. As Novac et al. (1999, 3) point out, ?widespread funding con- straints, which necessitate the development of small local projects and low-cost solutions, diminish opportunities for documentation and in- formation exchange regarding the strengths and weaknesses of new pro- jects, service innovations, and integration of multiple services.? Further- more, most process-oriented or evaluative housing research occurs at post-occupancy stages, and therefore misses key insights about decision- making processes or barriers experienced from the very beginning of a project. Researchers themselves have to take responsibility for securing multi-year funding for this kind of research. Often funding mandates or research timelines do not dovetail neatly with design and development processes. Some of the recommendations for design interventions to address homeless young people?s needs arising-from this research include the following. Recommendations for Design Interventions Consultation with youth is essential : Findings from this research high- light the obvious ? the importance of youth participation in any project designed for their needs. Without youth consultations, Peel Youth Vil- lage would have embarked on very different design directions. Design must be flexible : Potential flexibility in design for ease of retrofitting or renovating will accommodate changing needs on the part of youth over the long-term, as any project once opened grows and ma- tures. Shifts in programming may require spaces to be re-configured. Housing must be linked to community -building: Shared so-called amenity space is intended to enhance safety and security (visual sight lines) for both residents and staff, and to facilitate a sense of collective well-being. Conventional models of apartment living with double- loaded corridors and private units do not offer the same potential for community-building. The relation to the larger community must be considered : Within the larger community and site context, Peel Youth Village offers not only housing for homeless youth, but also recreational space to youth living in the broader Peel community. A range of needs must be accommodated: Homeless youth have a range of social service and housing needs, including emergency shelter, transitional housing, long-term housing, educational training, employ- ment training, among others. Youth-serving agencies and organizations often address these needs in an uncoordinated fashion. Peel Youth Vil- lage acted as a catalyst to encourage discussion about gaps in existing services, and to point to ways of cooperating among sometimes compet- ing interests. A time-based model has limits: Peel Youth Village was envisioned as offering a continuum of housing options and supportive pro- gramming that youth would move through at their own pace, and not according to dictated time limitations. Emphases shift during the approvals processes: The way in which a room is described can affect the way those reviewing plans, political decision-makers, or even the general public judge the feasibility or worth of a project. Creativity in presenting design ideas and an acute aware- ness of who the audience is can smooth approvals processes. In the case of Peel Youth Village, during municipal approvals processes, for in- stance, the supportive or community-building aspects of the project were downplayed in favour of housing provision. Demonstration projects are important: Demonstration or pilot pro- jects may inspire other constituencies to build upon lessons learned and develop new projects addressing similar or related social conditions. The designation ?demonstration project? provides an exploratory framework for developing any project so that lessons, whether positive or negative, can inform future initiatives elsewhere. As one Network member ex- plained during a design meeting, ?By referring to this as a demonstra- tion project or pilot project or experimental project, we move from the exact to something with a little more latitude.? Every demonstration project surely arises from that primordial utopian impulse to disrupt, as James Holston proposes, ?the imagery of what...society [ understands] as the real and the nat ural...[and] defami l- iarize[s] the normative, moral, aesthetic, and familiar categories of social life.? Holston asks: ?Without a utopian factor are not plans likely to re- produce the oppressive status quo?? He answers his own question: ?Without a utopian factor, plans remain locked in the prison-house of existing conditions? (Holston, 1989:315-317). Epilogue The Peel Youth Village community centre opened its doors to the com- munity in the autumn of 2005, and its housing became operational the following summer. For more information about Peel Youth Village, visit http://www.peelregion.ca/ow/ourservices/community- program/housing/pyv/. See also the Eva?s Initiatives Tool Kit, which profiles Eva?s Phoenix in Toronto, Peel Youth Village, and other innovative projects for home- less youth across Canada: http://www.evasinitiatives.com/EVAsToolKit/. Acknowledgements The author gratefully acknowledges the support of the Social Sciences and Humanities Research Council of Canada through a Standard Re- search Grant (2000?2003). This article represents a revised version of an article originally published in Open House International 29(2): 62-70, 2004. Rae Bridgman is Professor, Department of City Planning, Faculty of Architec- ture, University of Manitoba, and a founding member of the architectural de- sign and planning firm BridgmanCollaborative Architecture. References Bridgman, R. (2006). StreetCities: Rehousing the Homeless. Peterborough, ON: Broadview Press. Bridgman, R. (2004). Bridging Public-Private Partnerships in a Case Study of Housing and Employment Training for Homeless Youth. Canadian Journal of Urban Research 12(2): 205-230. Bridgman, R. (2003). Safe Haven: The Story of a Shelter for Homeless Women. To- ronto: University of Toronto Press. Bridgman, R. (2001). I Helped Build That: A Demonstration Housing and Em- ployment Training Programme for Street Youth in Toronto. American An- thropologist 103(3): 779-795. Bridgman, R. (1998a). A ?city? within the city: a Canadian housing model for the homeless. Open House International 23(1):12-2. Bridgman, R. (1998b). The architecture of homelessness and utopian pragmatics. Utopian Studies 9(1): 50-67, 1998. Canada Mortgage and Housing Corporation. (1999). Documentation of Best Prac- tices Addressing Homelessness. Research Report. Ottawa: Canada Mortgage and Housing Corporation. Crane, R., and Takahashi, L. M. (1998). Who are the suburban homeless and what do they want? An empirical study of the demand for public services. Journal of Planning Education and Research 18: 35-48. Fitzpatrick, S. (2000). Young Homeless People. London: Macmillan Press Limited. Hart, R. (1997). Children?s Participation : The Theory and Practice of Involving Young Citizens in Community in Development and Environmental Care. London: Earthscan Publications Ltd. Holston, J. (1989). The Modernist City: An Anthropological Critique of Brasilia. Chi- cago: University of Chicago Press. Kraus, D., Eberle, M., & Serge, L. (2001). Environmental Scan on Youth Homeles s- ness: Final Report. Ottawa: Canada Mortgage and Housing Corporation. Liddiard, M. & Hutson, S. (1991). Homeless young people and runaways: agency definitions and processes. Journal of Social Policy 20: 365-388. Novac, S., Brown, J., & Gallant, G. (1999). Women on the Rough Edge: A Decade of Change for Long-Term Homeless Women. (Research Report.) Ottawa: Canada Mortgage and Housing Corporation (CMHC). Chapter 3.8 The ?hand - to- mouth? existence of homeless youths in Toronto VALERIE TARASUK, NA OMI DACHNER, BLAKE P OLAND , AND STEPHEN GAET Z With homelessness recognized as a growing problem in many developed countries, ?the homeless? have become an increasing focus of nutrition research and intervention. Problems of insufficient food access (Anto- niades & Tarasuk, 1998; Burt et al., 1999; Children?s Sentinel Nutrition Assessment Program, 2005; Gunderson et al., 2003; Khandor & Mason 2007; Whitbeck et al., 2006) and nutritional vulnerability (Cohen et al., 1992; Darmon et al., 2001; Evans & Dowler, 1999; Gelberg et al., 1995; Johnson & McCool, 2003; Langnase & Mullis, 2001; Silliman et al., 1998; Tarasuk et al., 2005; Wolgemuth et al., 1992) have been documented among homeless groups in many affluent Western nations. Ethnographic research findings suggest that homeless individuals live a ?hand-to- mouth? existence, locked in a daily struggle to meet their immediate needs for food and shelter (Dachner & Tarasuk, 2002; Hagan & McCar- thy, 1997; Wingert et al., 2005). Their nutritional vulnerability has been linked to the inadequacy of meals served in soup kitchens or shelters (Burt et al., 1999; Cohen et al., 1992; Darmon et al., 2001; Johnson & McCool 2003; Silliman & Wood, 2000 In 2003, we undertook a study of 261 homeless youths in Toronto to characterize the extent and nature of their nutritional vulnerability ( ), but there has been little examina- tion of the role of other food acquisition strategies. Tarasuk et al., 2005; Gaetz et al., 2006; Li et al., 2008). Most youths we interviewed existed outside the ?social safety net,? obtaining money through the informal (and often illegal) economy and living in public spaces. Dietary assessments (results of which have been reported else- where) indicated that most had inadequate intakes of folate, vitamin A, vitamin C, zinc and magnesium; additionally, more than half of the young women in the sample had inadequate intakes of iron and vitamin B12 (Tarasuk et al., 2005 Talking to homeless youth ). Here we examine the relationship between chronic food deprivation and food acquisition practices among this sam- ple to gain a fuller understanding of their vulnerability. Data collection occurred between April and October 2003. Youths were eligible to participate if they were: (a) 16? 24 years of age; (b) not preg- nant; and (c) without stable, secure housing arrangements, defined as having spent 10 or more of the previous 30 nights sleeping in a tempo- rary shelter, indoor or outdoor public space, or friend?s place, because they had no place of their own. Six drop-in centres and 28 outdoor locations where homeless youths ?hung out? (e.g., under bridges or in abandoned buildings, parks, or ga- rages) in downtown Toronto were identified as recruitment sites. Drop- in centre workers were contacted to obtain estimates of the number of eligible youths using their facilities, and field observations were con- ducted to estimate the number of homeless youths in each outdoor area. Quotas proportional to these estimates were developed for each site, as- suming a target sample of 240 youths (120 male, 120 female). Because the number of homeless youths in any location at any time was relatively small, random sampling was not feasible. Instead, inter- viewers systematically approached each youth they encountered at each site. Of the 483 youths approached, 170 were deemed ineligible (68 per- cent because they failed to meet the criteria for unstable housing, 24 per- cent because they were over 24 years of age, 4 percent because they were pregnant, 4 percent for other reasons), 40 declined to participate, and 12 were subsequently dropped from the study (11 because they were found to be duplicates and one because of data quality concerns). A final sam- ple of 261 youths was achieved, reflecting an 83 percent participation rate. Of the final sample, 70 percent were recruited from outdoor loca- tions. From a review of earlier studies of homeless youths in Canada (Ba- ron, 1989; Dachner & Tarasuk, 2002; Gaetz, 2004; Gaetz & O?Grady, 2002; Hagan & McCarthy, 1997; McCarthy & Hagan, 1992), we identified five means of food acquisition common among this group: (a) purchasing food with money obtained through activities like panhandling; (b) ob- taining food from other people (passersby or those with whom they had some relationship); (c) obtaining food free of charge or at nominal cost from charitable meal programs; (d) stealing food; and (e) retrieving food that had been discarded by others. Participants were interviewed when recruited and a time and loca- tion for a second interview was arranged. Out of 261 youths in the sam- ple, 195 (75 percent) completed second interviews, and 91 percent of these occurred within 14 days of the first interview. Both interviews in- cluded a multi-pass 24-hour dietary intake recall, but the first interview also included an interviewer-administered questionnaire designed to capture sociodemographic characteristics, living circumstances over the previous 30 days, frequency of alcohol and drug use over the previous 30 days, food security, food acquisition strategies, and strategies used to obtain drinking water. Food security was assessed using the 30-day Food Security Module and a 6-month measure adapted from the Household Food Security Survey Module (Bickel et al., 2000). To characterize participants? use of charitable meal programs, we asked how often in the previous seven days they had obtained meals from a soup kitchen, drop-in centre, shelter, or mobile van (the primary routes through which food assistance is dispensed to homeless individu- als in Toronto). To determine their use of other strategies, we developed a series of closed-ended questions to ask how often over the previous 30 days they had engaged in specific activities to get food when they had no food or money for food; frequency was recorded as ?never,? ?some- times,? or ?often.? The questionnaire was pilot-tested on a sample of 25 homeless youths to ensure the acceptability and comprehensibility of all items. Data analysis To classify food security status over six months, we applied the thre- sholds used to classify adult food insecurity in U.S. population surveys (Bickel et al., 2000; Nord et al., 2006). Food security over the previous 30 days was assessed in terms of chronic food deprivation, defined as re- porting three or more of five conditions (i.e. skipped meals, ate less than you felt you should, felt hungry but did not eat, cut the size of meals, went a whole day without eating) for 10 days or more during this period. We used logistic regression to compare food security prevalence rates by gender and identify personal characteristics associated with chronic food deprivation over the previous 30 days, considering age (under 19 years, over 19 years), duration of homelessness (less than one 1 year, more than 1 year), education (completion of grade 12 or not), fre- quent drug use (defined as using crack, cocaine, speed/crystal, opiates, glue, gasoline, tranquilizers, hallucinogens or ecstasy every day or sev- eral times per week) over the previous 30 days, and consumption of al- cohol every day or several times per week over the previous 30 days. E xperiences of food deprivation and food insecurity We also used logistic regression to examine the association between chronic food deprivation and reported problems obtaining water to drink, the frequency of program use, considering rare (0-2 days) and fre- quent (6-7 days) use over the previous seven days, and the frequent use of other specific food acquisition strategies (defined as ?often? using the strategy in the previous 30 days). Because youths? food acquisition pat- terns and experiences of chronic food deprivation differed by gender, all analyses were stratified. Sample characteristics are summarized in Table 1. Almost all youths had been food-insecure over the previous six months and most experienced severe food insecurity (Table 2). Table 1 : Sociodemographic characteristics and present circumstances ( % ): homeless youths, Toronto, Canada, 2003 Males (n = 149) Females (n = 112) Age 20 16?18 years 38 80 19?24 years 62 Ethno-racial identity 84 White 77 6 Black (African/Caribbean) 5 7 First Nations, Inuit, Metis, Other Aboriginal 13 3 Other (Asian, Latin American, etc.) 5 Education1 14 Grade 8 13 63 Grade 9?11 71 24 Grade 12 or more 16 Time since leaving home 2 months 12 18 11 3?6 months 18 8 7?12 months 8 68 >12 months 56 Place where previous night had been spent 56 Outdoors 67 21 Friend?s place 22 Squat 8 2 4 8 Shelter 4 Other 7 3 3 Main source of income in previous 30 days4 Panhandling or squeegeeing 56 5 54 22 Theft or drug trade work 13 6 Sex trade work 19 7 Government transfers 3 4 Selling items (handicrafts, etc.) 4 3 Money from family or friends 4 1 Paid employment 4 Frequent heavy drug use in previous 30 days 34 Yes 45 66 No 55 Consumption of alcohol every day or several times per week in previous 30 days 48 Yes 43 52 No 57 1. Generally, youths in Canada complete Grade 12 at the age of 18 years. 2. Squats are makeshift shelters in abandoned buildings. 3. Included jail, Internet cafe, bathhouse, hotel, ?with client? and ?own place.? 4. Two males reported no source of income. 5. The practice of washing the windows of vehicles stopped at intersections and then ask- ing motorists for money. Over the previous 30 days, 43 percent of females and 28 percent of males experienced chronic food deprivation. Severe food insecurity and chron- ic food deprivation were more prevalent among females. Thirty-two per cent of females and 48 percent of males reported problems obtaining drinking water. For males, this was positively asso- ciated with chronic food deprivation, but no significant association was observed for females. The most commonly reported sources of drinking water were fast-food restaurants and washrooms (Table 3). Chronic food deprivation appeared unrelated to youths? age or education level. There was also no relationship between the duration of homelessness and chronic food deprivation among males, but the odds of chronic food de- privation among females who had been homeless for a year or more was higher than for those who had become homeless more recently. Females who reported consuming alcohol daily or almost daily had higher odds of chronic food deprivation, but no similar association was observed for males. Frequent heavy drug use was not associated with chronic food deprivation. Relationship between chronic food deprivation and food acquisition strategies In the previous 7 days, 87 percent of males and 89 percent of females had made at least some use of charitable meal programs, with drop-in centres the most common source of meals (Table 4). The frequency with which youths used meal programs was unrelated to their experiences of chron- ic food deprivation. When they needed food over the previous 30 days, almost three- quarters of youths panhandled and about half stole food, but neither strategy was associated with chronic food deprivation for males or fe- males (Table 5). Over the previous 30 days, 44 percent of males and 47 percent of females had borrowed money from someone to buy food; the median amount of money borrowed was $ 15. The behaviour was asso- ciated with chronic food deprivation for females, but not males. Further indication of the pervasive vulnerability associated with indebtedness came from youths? reports of putting off paying for other things as a way to free up money for food. Almost half reported such behaviours in the previous 30 days, and youths who often postponed payments had signif- icantly higher odds of chronic food deprivation (Table 5). Approximately half of the youths surveyed had eaten food dis- carded by others, and almost half reported getting free day-old food from fast-food establishments at some point in the last 30 days. The latter strategy was not linked to chronic food deprivation, but for both males and females, the odds of reporting often eating discarded food increased if they had experienced chronic food deprivation over this same period (Table 5). Table 2 : Food security status over previous 6 months and previous 30 days ( % ): homeless youths, Toronto, Canada, 2003 Source Table 3 : Reported sources of drinking water*: homeless youths, Toronto, Canada, 200 3 Proportion reporting use (%) Males (n = 149) Females (n = 111)** Fast-food restaurants, coffee/doughnut shops 36 46 Washrooms (in food outlets and public places) 32 43 Public drinking fountains 36 24 Other people (friends or acquaintances) 28 30 Social service agencies (e.g. drop-in centres, outreach vans) 17 36 Purchased bottled water 22 27 Outdoor locations (e.g. lawn-watering devices, outdoor taps at gas stations, private residences) 13 10 *Because respondents could report more than one source, values do not add to 100%. **Data missing for one female. Males (n = 149) Females (n = 112) Food security over previous 6 months 9 Food secure 6 18 Moderately food insecure 9 73 Severely food insecure 85 Chronic food deprivation over previous 30 days 28 Yes 43 72 No 57 Table 4 : Frequency of meal acquisition from charitable food assistance programs over previous 7 d ays : homeless youths, Toronto, Canada, 2003 Number of days in last 7 days when 1 meal was obtained Proportion reporting use (%) Shelters Soup kitchens Outreach vans Drop-in centres Any program Males (n=149) Never 83 62 51 32 13 1?2 days 8 19 28 29 27 3?5 days 3 10 13 32 38 6?7days 7 9 9 7 21 Females (n=111)* Never 86 73 39 25 11 1?2 days 6 18 33 24 22 3?5 days 4 8 24 46 57 6?7 days 3 1 4 5 11 *Missing responses for one female. At times when they needed food, it was not uncommon for youths to seek out the company of others who could provide it. Approximately half of the youths reported going to a friend or relative?s place to eat, and one-quarter of youths reporting ?hanging out? with people just be- cause they had food (Table 5). The frequent use of these strategies was associated with chronic food deprivation for males, but not females. Ele- ven per cent of males and 23 percent of females had exchanged sex for food or money for food in the previous 30 days, but the frequent use of this strategy was associated with chronic food deprivation only among females (Table 6). Table 5: Frequency of u se of food acquisition strategies over the previous 30 days (%) : homeless youths, Toronto, Canada, 200 3 Males (n = 147) Females (n = 112) Often Sometimes Never Often Sometimes Never Panhandled to get money for food 46 25 29 57 16 27 Stole food 15 32 53 16 41 43 Postponed payment of rent, debts, etc. 15 37 47 10 36 54 Got free day-old food from restaurants 9 32 59 7 39 54 Ate food discarded by others 21 28 50 18 25 57 Went to a friend?s or relative?s place for food 11 52 36 10 49 41 ?Hung out? with people because they had food 5 19 75 7 20 73 Traded sex for food or for money for food 5 6 89 11 12 78 Table 6 : Strategies employed to acquire food routinely or in times of desperation: homeless youths, Toronto, Canada, 200 3 Routine strategies Going to charitable meal programs . Panhandling to get money for food Getting free day-old food from restaurants Stealing food Desperate strategies* Postponing payments of debts, rent, etc. Eating food discarded by others Males only: Going to a friend or relative?s place for food ?Hanging out? with someone just because they have food Females only: Trading sex for food Borrowing money for food *Food acquisition strategies associated with significantly increased odds of chronic food deprivation over the previous 30 days. Nutritional vulnerability and poverty This study of homeless youths was undertaken to characterize the extent and nature of their nutritional vulnerability. The portrayal of food inse- curity that emerges from our research differs markedly from the pheno- menon commonly assessed among domiciled groups. The youths re- ported much higher levels of food deprivation than are typically observed in general population surveys (Health Canada 2007), highlight- ing the extreme vulnerability that comes with homelessness and the ab- ject poverty that underscores this condition. In addition to problems of food deprivation, many youths reported problems getting sufficient drinking water. Similar findings emerged from a recent study of street-based sex workers in Miami (Kurtz et al., 2005 The ways in which homeless youths endeavoured to manage their food needs reflect a ?hand-to-mouth? existence, characterized by the use of a wide diversity of strategies to obtain small amounts of food for immediate consumption. Many of these strategies were stigmatizing and unsafe; some were illegal. Our examination of the relationship between youths? use of specific food acquisition strategies and their level of food deprivation suggests that some strategies such as eating food discarded by others are acts of extreme desperation, whereas other behaviours like panhandling and stealing food are routine (Table 6). Although other re- searchers have not differentiated homeless youths? food acquisition be- haviours in this way, the fact that similar behaviours have been re- ported by others (Hagan & McCarthy, 1997; Wingert et al., 2005; McCarthy & Hagan, 1992) suggests that our findings are characteristic of homeless youths in this country. ). Without housing and with insufficient funds to purchase bottled water, homeless people are forced to rely on public sources of water or negotiate access to private supplies. In urban settings such as Toronto, access to public washrooms and drinking fountains has become increa- singly limited because of concerns about cost and liability. Thus inade- quate and insecure access to drinking water is an added dimension of food insecurity among homeless populations. Although many youths in this study routinely used charitable meal programs, this practice did not protect them from chronic food depriva- tion, nor did it obviate the need for them to acquire food in other ways as well. These findings highlight the need for a better ?safety net? to help youths meet their basic needs. In our qualitative research with homeless youths, they complained about the infrequent service, limited meal hours, and need to travel considerable distances to attend different cha- ritable meal programs at different times of the day or week (Dachner & Tarasuk, 2002; Gaetz et al., 2006). Our subsequent inventory of local cha- ritable food provisioning efforts confirmed that meal services for those outside the shelter system are, for the most part, intermittent and un- coordinated (Dachner et al., 2009), and the food served is generally of limited quantity and nutritional quality (Tse & Tarasuk, 2008 Our examination of youths? food acquisition strategies highlights the gendered nature of homelessness, a phenomenon documented else- where as well (Ensign & Bell, 2005; Gaetz, 2004; Khandor & Mason, 2007; McCarthy & Hagan, 1992; Roy et al., 1999). Other research with street youths has found that males generally earn more than females and are more likely to operate independently, whereas females tend to engage collectively, both in money-making and in living arrangements (Gaetz, 2004). Consistent with this research, we found that using social relation- ships as a means to acquire food was routine for females, but such beha- viour indicated desperation for males. Nonetheless, female youths may engage more in high-risk, exploitive relationships, trading sex for food when they are desperate. ). While the establishment of ad hoc, charitable food programs for homeless individ- uals is a testament to community concern and resourcefulness, our re- search results argue strongly for a more coherent response. In conclusion, the pervasiveness and severity of food insecurity expe- rienced by homeless youths in the present study and their desperate means of food acquisition highlight the urgent need for more effective responses to food insecurity among this group. While more work could be undertaken to improve youths? food access through charita- ble meal programs in the community, we worry that this would amount to ?treating the symptom? rather than the problem. Homeless youths? food acquisition behaviours reflect the extreme desperation of their situ- ations, providing a moral and public health imperative to find solu- tions to the problem of youth homelessness in Canadian cities. Valerie Tarasuk is a Professor in the Department of Nutritional Sciences at the University of Toronto. Naomi Dachner is a study coordinator and researcher in the Department of Nutritional Sciences at the University of Toronto. Blake Poland is a Professor in the Dalla Lana School of Public Health at the Univer- sity of Toronto. St ephen Gaetz is the Associate Dean of the Faculty of Educa- tion at York University. This article was adapted from Tarasuk V, Dachner N, Poland B & Gaetz S (2009) ?Food deprivation is integral to the ?hand to mouth? existence of homeless youth in Toronto.? Public Health Nutr ition 12(9), 1437- 1442. Acknowledgements The research was funded by an operating grant from the Canadian Insti- tutes of Health Research (CIHR). The authors are indebted to Mark Nord for assistance with the food security measure and to Tony Jinguang Li for preliminary data analysis. References A Baron, S. W. (1989). Resistance and its consequences: The street culture of punks. Youth & Society , 21(2), 207? 237. ntoniades, M., & Tarasuk, V. (1998). A survey of food problems experienced by Toronto street youth. Canadian Journal of Public Health, 89(6), 371? 375. Bickel, G., Nord, M., Price, C., Hamilton, W. L., & Cook, J. (2000). Guide to measur- ing household food security. Washington, DC: U.S. Department of Agriculture, Food and Nutrition Service, Office of Analysis, Nutrition and Evaluation. Burt, M. R., Aron, L. Y., Douglas, T., Valente, J., Lee, E., & Iwen, B. (1999). Home- lessness: Programs and the people they serve. Washington, DC: The Urban Insti- tute. Cohen, B. E., Chapman, N., & Burt, M. R. (1992). Food sources and intake of homeless persons. Journal of Nutritional Education, 24(1), 45S? 51S. Children?s Sentinel Nutrition Assessment Program (2005). Homelessness: Conse- quences for young children?s health and nutrition (C-SNAP Brief). Retrieved December 2005, from http://www.c-snap.org. Dachner, N., & Tarasuk, V. (2002). Homeless ?squeegee kids?: F ood insecurity and daily survival. Social Science & Medicine, 54(2), 1039? 1049. Ensign, J., & Bell, M. (2005). Illness experiences of homeless youth. Qualitative Health Research, 14(9), 1239? 1254. Darmon, N., Coupel, J., Deheeger, M., & Briend, A. (2001). Dietary inadequacies observed in homeless men visiting an emergency night shelter in Paris. Public Health Nutrition, 4(2), 155? 161. Evans, N. S., & Dowler, E. A. (1999). Food, health and eating among single home- less and marginalized people in London. Journal of Human Nutrition and Di- etetics, 12(3), 179? 199. Gaetz, S. (2004). Safe streets for whom? Homeless youth, social exclusion, and criminal victimization. Canadian Journal of Criminology and Criminal Justice, 46(4), 423? 455. Gaetz, S., & O?Grady, B. (2002.) Making money: Exploring the economy of young homeless workers. Work, Employment & Society , 16(3), 433? 456. Gaetz, S., Tarasuk, V., Dachner, N., & Kirkpatrick, S. (2006). ?Managing? homeless youth in Toronto. Canadian Review of Social Policy, 58(43), 43? 61. Gelberg, L., Stein, J. A., & Neumann, C. G. (1995). Determinants of undernutri- tion among homeless adults. Public Health Reports, 110(4), 448? 458. Gunderson, C., Weinreb, L., Wehler, C., & Hosmer, D. (2003). Homelessness and food insecurity. Journal of Housing Economics, 12(3), 250? 272. Hagan, J., & McCarthy, B. (1997). Mean streets: Youth crime and homelessness. Cam- bridge: Cambridge University Press. Health Canada (2007). Canadian community health survey, Cycle 2.2, Nutrition (200 4 ) ? Income-related household food security in Canada. Ottawa: Office of Nutrition Policy and Promotion, Health Products and Food Branch, Health Canada. Khandor, E., & Mason, K. (2007). The Street Health Report. Toronto: Street Health. Johnson, L. J., & McCool, A. C. (2003). Dietary intake and nutritional status of older adult homeless women: A pilot study. Journal of Nutrition for the Elder- ly, 23(1), 1? 21. Kurtz, S., Surratt, H., Kiley, M., & Inciardi, J. A. (2005). Barriers to health and social services for street-based sex workers. Journal of Health Care for the Poor and Underserved, 16(2), 345? 361. Langnase, K., & Mullis, J. M. (2001). Nutrition and health in an adult urban homeless population in Germany. Public Health Nutrition, 4(3), 805? 811. McCarthy, B., & Hagan, J. (1992). Surviving on the street: The experiences of homeless youth. Journal of Adolescent Research, 7(4), 412? 430. Li, A., Dachner, N., & Tarasuk, V. (in press). Food intake patterns of homeless youth in Toronto. Canadian Journal of Public Health. Roy, E., Haley, N., Lemire, N., Boivin, J., Leclerc, P., & Vincelette, J. (1999). Hepa- titis B virus infection among street youths in Montreal. Canadian Medical As- sociation Journal, 161(6), 689? 693. Nord, M., Andrews, M., & Carlson, S. (2006). Household food security in the United States, 2005 . Washington, DC: US Department of Agriculture, Economic Re- search Service. Silliman, K., & Wood, S. (2000). Evidence of nutritional inadequacy of meals served to homeless populations in rural northern California. Ecology of Food and Nutrition, 40(3), 285? 297. Silliman, K., Yamanoha, M. M., & Morrissey, A. E. (1998). Evidence of nutritional risk in a population of homeless adults in rural northern California. Journal of the American Dietetic Association, 98(8), 908? 910. Tarasuk, V., Dachner, N., & Li, J. (2005). Homeless youth in Toronto are nutri- tionally vulnerable. Journal of Nutrition, 135(8), 1926? 1933. Tse, C., & Tarasuk, V. (2008). Nutritional assessment of charitable meal pro- grammes serving homeless people in Toronto. Public Health Nutrition, 11(12), 1296? 1305. Whitbeck, L. B., Chen, X., & Johnson, K. D. (2006). Food insecurity among home- less and runaway adolescents. Public Health Nutrition, 9(1), 47? 52. Wingert, S., Higgitt, N., & Ristock, J. (2005). Voices from the margins: Under- standing street youth in Winnipeg. Canadian Journal of Urban Research, 14(1), 54? 80. Wolgemuth, J. C., Myers-Williams, C., Johnson, P., & Henseler, C. (1992). Wast- ing malnutrition and inadequate nutrient intakes identified in a multiethnic homeless population. Journal of the American Dietetic Association, 92(7), 835? 839. Chapter 4 .1 Supporting Young Homeless Mothers Who Have Lost Child Custody SYLVIA N OVAC , EMILY PARADIS, J OYCE BR OWN , AND HEATHER MORT ON Homeless and pregnant Each year throughout the 1990s, about 2,000 young women 15 to 24 used shelters in Toronto. Homeless young women are often searching for someone to love and protect them and may become sexually active as a way to maintain relationships and avoid being alone, even if the rela- tionship is a bad one. Many of them ? some studies have found that as many as half of the homeless young women using health services or shelters ? are or have been pregnant. Homeless pregnant adolescents are a vulnerable group. Both home- lessness and pregnancy are risk factors for poor health among youth. Pregnancy among homeless young women is associated with earlier and more severe abuse during childhood, earlier onset of drug use, and poor mental health. Several Toronto social service agencies have organized a network called Young Parents No Fixed Address (YPNFA) to address the needs of homeless young pregnant women and parents. The YPNFA Commit- tee commissioned research to explore service interventions for young homeless mothers who lose custody of their child, with a focus on help- ing them cope with their loss and move on with their lives. Longing for a family A study by Sarah Gorton in London, England, found that many home- less young women have never had a family life. When they become pregnant, they often imagine changing their lives and creating a family. Some hope that their parents will accept them again, on behalf of the baby. Birth fathers may demonstrate a short-lived interest in the preg- nancy, but the fathers are rarely involved when their baby is born. Some young mothers become inspired and motivated to attend to the baby?s needs, find child care, return to school, and change their lives. Those who succeed draw on inner resources, which probably include a sense of competence as a parent. Other young women are too burdened with their own problems to change their lives. When homeless young pregnant women fantasize a new family life for themselves, that fantasy and the hope it represents are destroyed when their baby is taken into care. Young parents are not necessarily bad parents. An analysis of more than 8,500 Illinois child welfare case files showed that adolescents were not overrepresented among the cases of parents who mistreated their children. Substantiated harms to children are associated not with paren- tal age, but with lack of prenatal care, low maternal education, single- parent status, unshared responsibility for child care, and poverty. Repeating patterns Studies have found an intergenerational pattern between state care of children and homelessness. Homeless adults who themselves were in foster care are more likely to have their own children in foster care. There is also a link between homelessness and the child welfare sys- tem. At least in part, this is the result of welfare reform that reduced funding for income support and other programs. To some extent, home- lessness may be, in itself, a trigger for child protection referrals. More- over, without adequate support from other agencies, young mothers who experience homelessness may be unable to maintain independent living with their children, even after they are re-housed. Homeless parents face a dilemma. For families who receive or re- quire social assistance, social benefits or a housing subsidy may be re- duced or terminated when a child is placed in temporary care; this re- duction in income may threaten the family?s ability to maintain its housing. The parents may be forced to move to a smaller home or end up in a shelter, which makes them unable to provide a suitable home for their children to return to. Even if there is an opportunity to reunite with the children, parents are not eligible for social assistance or subsidized housing until the child resides with them, yet they cannot afford the housing they require without that assistance. Separating mother and child Long-term homelessness, drug use, and mental illness increase the odds of mother-child separation. A U.S. study that compared homeless moth- ers with and without their children found that almost half of those who were separated from their children had a severe mental illness and one- third suffered from alcoholism. When mothers are forced to give up a baby, most experience normal initial grief reactions (anger, guilt, and depression), but for homeless young mothers, these emotions may persist and lead to chronic, unre- solved grief. Some studies found that mothers who had relinquished a child had more grief symptoms than women who had lost a child to death, including more denial, despair, atypical responses, and distur- bances in sleep, appetite, and vigour. Although society expects the relinquishing mother to resume her former role as if the experience had not occurred, these women are at risk for long-term physical, psychological, and social problems. Lack of social acceptance of the grief of relinquishing mothers contributed to chronic, pathological grief. Family and marital problems were common, as were fantasies and searching behaviour. If the decision to relinquish was forced on the woman, searching may be a way of finding a lost part of the self, more than an attempt to reclaim the lost child. Women who had been drug users before becoming pregnant some- times sought refuge in drugs after having a child removed from their care. From their perspective, they had not only lost a child, but the op- portunity to take on a positive social role. Another common maladaptive response is rapid subsequent pregnancy in an effort to replace the lost baby. Various interventions have been proposed (including counselling, dream analysis, role playing, interviews with adoptive parents, and con- tinuing support from a health care professional), but there has been no long-term follow-up to evaluate the effectiveness of such interventions. There appear to be no studies that have asked the birth mother what she thought would have been helpful in her situation. Many birth mothers have indicated that they did not receive acknowledgement of their loss from health professionals involved in their care. The role of child welfare agencies Child welfare legislation is directed by the ?best interests of the child? and, in practice, child welfare agencies are increasingly focused on pro- tection efforts. Destitute young mothers must therefore provide adequate parenting with minimal support from a badly eroded social welfare sys- tem. If a young woman fails, the only available solution is to apprehend her child. Once a child has been removed from the mother?s custody, the odds of family reunification are low. When a child is born, the most common reasons for taking that in- fant into care are confirmed drug use by the mother during her preg- nancy, lack of prenatal care, lack of an appropriate place for the mother and child to stay, and a history of apprehension of other children of the same mother. These problems are usually coupled with a general lack of formal supports. The likelihood of an apprehension at birth is also associated with administrative factors, such as the length of stay in hospital after the child?s birth, the services available at the time the child is born, and the admissions criteria for supportive residential programs. The usual length of stay in hospital for a vaginal birth is two days, which is not long enough for the child protection agency to make a thor- ough assessment of an unknown mother?s parenting prospects and available supports. As a result, a newborn might be apprehended unless the mother can quickly demonstrate that she has a strong network of supports in place and a supervised environment to return to with her baby. Risk of apprehension increases further when a baby is born late in the evening or on a weekend or holiday, when the hospital social worker is not available to set up supports. Even when the worker is available, there are usually waiting lists or time-consuming procedures for admis- sion to residential programs that offer safe environments for a woman and her baby while child protection conducts a more thorough investiga- tion. If these exceed the length of stay at the hospital, the child may be apprehended as a precaution. Child protection agencies generally inform hospitals and residential programs of their intention to apprehend a child, either when a child protection worker is on the way to apprehend the child or before. They sometimes do not inform the women themselves, due to concerns that the woman will leave with her child to avoid the apprehension. Mid- wives or outreach workers who were present at the birth may or may not be notified. Some workers, specifically those whose focus is counselling and advocating for the mother, are not informed in advance because it would present an ethical conflict for them to keep this information from the mother. The apprehension process The following information on the apprehension process was drawn from interviews with Toronto social service workers. When newborns are apprehended at a hospital, the child protection agency representatives usually arrive within hours of the birth, although some social workers ask that the process take place the following day. A midwife or outreach worker may be present to support the mother. The hospital social worker or an outreach/agency worker usually mediates between the mother and the child protection agency representative. If time allows, the representative might gather information first, and return once the decision to apprehend has been made. Sometimes, how- ever, the child protection agency representative arrives with a letter of apprehension. The latter situation is hard on the mother. If there is con- cern that the mother will flee or become violent, the baby will be taken from the room before the mother is informed of the apprehension. When children are apprehended at shelters or residential programs, the apprehension takes place in an office. Workers are present to support the mother, and will accompany her to her room to pack the baby?s things. Occasionally, an apprehension may take place when the mother is not present, for example, when a mother has left the child with an in- adequate alternative caregiver, or has failed to return at an expected time. In some apprehensions, the mother hands the baby or child directly to a child protection agency representative; in others, the mother gives the baby to a trusted social worker or advocate to be given to the child protection agency representative. One child protection agency representative removes the baby to a waiting vehicle while the other stays to provide information to the mother, including written information about child protection and her rights. The mother also receives a copy of the warrant of apprehension, which describes the grounds on which the apprehension has been made. The child protection agency representative also explains the arrange- ments for visitation and information about the first court date. The mothers? reactions vary from woman to woman. Some yell, scream, or cry; some become enraged or even violent; others appear ex- pressionless, numb, or dissociated. Whatever their emotional expression, the immediate trauma usually makes it impossible for mothers to re- member the sequence of events and absorb the information that they are being given. Their understanding of what has taken place, what will happen next, and what is expected of them will often be inaccurate or incomplete. A Family Court hearing must be held within five days of the appre- hension. If the mother has not retained her own lawyer, duty counsel is available at the court to ensure that she understands her rights and that there is a plan for access to the child. Though there is no legal directive stipulating when a mother is entitled to visit her child after an apprehen- sion, such visits generally happen as soon as possible. The first court appearance usually results in an adjournment and a Temporary Order for Care and Custody until the next court date. The duty counsel provides the mother with information on how to obtain a lawyer for subsequent court dates. For young homeless mothers, this means applying for legal aid. The court system is alienating for a young homeless woman who does not understand the proceedings and who may feel that the child protection agency and the court are conspiring against her. Young women with low levels of literacy are at a particular disadvantage. For the children, state wardship may mean adoption into a new, permanent family, but for many it means a childhood spent in foster families or group care. Some young people run away from state care and adoption. Others, lacking the life skills and socio-economic support that are usually provided by a family of origin, end up on the streets in late adolescence. If these young women themselves become pregnant, the cycle may begin anew. Left alone When they lose custody of their children, women lose access to many services and resources. Distraught and mistrustful of staff, they may leave the hospital against medical advice. This may place their postpar- tum recovery at risk, especially if they are returning to the streets or to a shelter where they will not be permitted bed rest. Women who have been staying in a shelter or residential program for pregnant women or mothers may leave immediately after the apprehension because it is too painful to stay with other mothers and their children. Even if she chooses to stay, the mother will be required to move once a plan for her relocation is in place. Service providers who work with women during the perinatal period, such as the Toronto Public Health program, are often reluctant or unable to stay involved with a woman whose child is not in her care. After an apprehension, Child Health public health nurses link women to appropriate services in the community, but do not have the mandate to follow up. A mother who loses her child also loses National Child Benefit and social assistance payments associated with the child, causing a sharp re- duction in her income. Later in the process, a woman who is living in subsidized housing may be required to move to a smaller unit if she permanently loses custody of her child. Young women with no home to go to must move to a shelter in the youth system or one for women unaccompanied by children, where there are usually few or no supports for women coping with the child protection system. This withdrawal of services and resources affects women?s ability to meet the conditions for reunification with their chil- dren. The mother?s social, emotional, and relational responses to her loss, and her way of coping, may limit her capacity to work through the loss or seek reunification with her child. Women often terminate relation- ships with service providers or agencies who were involved with the decision to apprehend. Although these women may return later, the re- sponse of ?taking a break? from services leaves them without access to the support of workers who know them well during the initial crisis pe- riod of coping with the apprehension. The loss of a child can also result in the loss of a mother?s motiva- tion to get off drugs, turn her life around, or even stay alive. For women whose babies are apprehended at birth, there is a profound sense of de- spair because they were denied any chance to be a mother. For women whose children are apprehended later, the implication that they are ?un- fit mothers? is a blow to their self-esteem. A later loss may be even more painful, because it ruptures the bond that has formed between mother and child, and signals a woman?s fail- ure in spite of her best efforts. The immediate grief may be complicated by postpartum depression that is often unacknowledged by service pro- viders. The demands made of young homeless mothers fail to take into ac- count their physical and emotional situation. The day-five court appear- ance and all the activities that lead up to and follow it are required, even when a woman is in the immediate postpartum recovery period. In other words, during a time when middle-class mothers are being advised by medical professionals to do nothing but eat, sleep, and breastfeed, young homeless mothers (who are more likely than middle-class mothers to be undernourished, to have had a Caesarean delivery, and to be at risk of complications) are walking or taking the TTC to legal clinics, legal aid offices, appointments, and visits. In spite of postpartum hormonal fluctuations and the trauma and grief of the loss, women must present themselves in court and at ap- pointments as reasonable, trustworthy, and cooperative individuals. Even then, if a woman is too successful at holding herself together, she risks being accused of not having a healthy attachment to her baby. What might help This study?s findings and the interviews with agency workers suggest ways in which the system could better respond to the needs of young homeless mothers. Bereavement support programs Specialized bereavement support for young homeless women who lose custody of their children would help these women cope with their loss and grief. These services could take the form of group or individual counselling, but whatever the form of support, the process needs to hon- our the emotional impact of the loss and recognize the woman?s experi- ence, including the impact of wider societal issues ? such as poverty, abuse, and racism ? on their lives. A woman?s own experience in care and the factors that contributed to her own apprehension also need to be considered. Programs for women with addictions and programs that address the effects of post-traumatic stress disorders should also address the im- pact of child apprehension. Peer-support models allow women to share strategies and skills for coping with custody loss, and may be less threat- ening than intensive individual counselling. Young women may not be ready for counselling until years after an apprehension, so these services must remain available for a woman when she is ready. Bereavement support and related services for mothers involved in child protection should be offered outside child protection agencies, so that mothers feel safe revealing their thoughts and feelings without fear that this will affect their chances of reunification with their children. And while there is a need for specific programs in bereavement support, agencies such as shelters should also incorporate training in bereave- ment support for their staff. Service Coordination and Protocols YPNFA?s efforts to ensure service coordination among hospitals, treat- ment programs, and shelters have improved the situation for young homeless mothers. These efforts should be expanded to include shelters for youth and single women, perhaps by designating beds and services in these shelters for women leaving family shelters after the apprehen- sion of their children. This would enable the ongoing provision of sup- port and a linkage with the family shelter in the event of reunification. Protocols are needed to guide service providers in the event of child apprehension. Hospitals, shelters, and other settings that deal with young homeless mothers need such protocols to ensure consistent staff response and appropriate, timely interventions. Workers also need support as they accompany mothers through their grief and loss. The consequences of lack of support include burnout and a decline in workers? ability to remain empathic and non- judgmental. Organizational policy should provide for such support, in- cluding immediate critical incident debriefing and regular clinical super- vision that is not tied to performance evaluation. Education and information Educating nurses, doctors, and social service workers would ensure more respectful, appropriate and sensitive services for young homeless women that will assist them in coping with the trauma. With its empha- sis on reunification within a community and supportive groups for women, the holistic approach of Aboriginal services may provide a model for other service providers. Residential services and housing Giving women an appropriate place to stay is a crucial factor in main- taining or regaining custody of a child, as well as in working through the grief of custody loss. Supportive residential services provide vital stabil- ity for mothers?and reassurance for child protection agencies?in the perinatal period. At present, there is no environment that offers the high-intensity supervision of a hospital and the possibility of immediate referral for young homeless or underhoused women whose first contact with ser- vices is their arrival at the hospital to give birth. The lack of such an en- vironment means that some apprehensions happen by default ? not be- cause the child protection agency has assessed that the child is unsafe in the mother?s care, but because the service workers do not have enough time to make that assessment during the brief postnatal hospital stay. A residential service to fill this gap would prevent unnecessary postnatal separation, and enable women to make the most of the optimism and motivation that motherhood brings. There are also no residential substance abuse treatment programs where women can stay with their children. Pregnant women should also be allowed to move from single women?s or youth shelters into family shelters earlier in their pregnancies, in order to benefit sooner from the enhanced services and more stable environment that these shelters offer. Housing for young homeless mothers must be affordable, with op- tions for transitional, supervised housing as well as independent living. It should be located near appropriate services. On-site supports and training should be available, including parenting classes and home- based mentorship in life skills. Finally, such housing must accommodate both women who are parenting alone and women with partners. Policy Changes Some studies of child protection suggest that conditions identified as ?neglect? in fact are the effects of poverty, therefore policy changes in income security are required to enable young homeless mothers to fulfil their potential as good-enough parents. In Ontario, a single mother with one child receives $860 a month from wel fare. Meanwhile, the cost of maintaining a child in state care is estimated to be $1,950 a mo nth. Clearly, the state should redirect financial support from maintaining children in care to improving conditions in the parents? homes. Child protection agencies require more resources to function effec- tively. Escalating caseloads and declining numbers of workers have forced the adoption of an increasingly bureaucratized and risk-oriented approach that tends to focus on poor, racialized, and immigrant families. Moreover, current child protection budgets no longer let workers em- ploy creative measures to relieve stressors related to neglect or abuse. For example, in the past, budgets included discretionary funds that workers could use to replace a family?s broken washing machine or to provide training in homemaking skills. Some child protection policies demand reassessment. Problems with foster care and group care must be addressed to prevent the per- petuation of the cycle of abuse, trauma, and homelessness that compli- cate many of these young mothers? attempts at parenting. The poten- tially discriminatory bias of risk assessment instruments may contribute to the overrepresentation of racialized, poor, young, mother-headed, and Aboriginal families in child protection. Finally, policy changes currently under consideration?including a shift away from apprehension and towards family-centred support?would enable more young mothers to maintain custody of their children. Changing the pattern During our research, we heard about one success story: A homeless 16- year-old who had lost custody of her baby received support throughout the apprehension process. A social worker accompanied her on every access visit, treated her as a parent, and acknowledged her pain and loss. Together, they discussed the young woman?s future. The worker assisted woman with practical matters, such as birth control and access to transit fares, taxi chits, and menstrual napkins. She acknowledged the links be- tween the birth mother?s family history and current situation. The CAS workers involved were also supportive. They recognized the mother?s childhood trauma and perceived her as a victim of bad cir- cumstances rather than as a bad mother. They were kind, took the time to get to know her, and made a long-term commitment to the relation- ship. The social worker and the mother maintained contact over several years. The young woman did not become pregnant again until she was 19 years old and better prepared for motherhood. At that time, she re- ceived good, supportive supervision. All these factors contributed to a successful outcome in the second pregnancy. Change is possible. Sylvia Novac is a research associate at the Centre for Urban and Community Studies, University of Toronto. E mily P aradis is a doctoral candidate at the Ontario Institute for Studies in Education, University of Toronto. Joyce Brown is Executive Director of Ontario Council of Alternative Businesses, an organization that provides employment for psychiatric consumer/survivors. Heather Morton is a Mental Health Nurse with Toronto Public Health and a member of the bereavement sub-committee with Young Parents No Fixed A d- dress. Chapter 4.2 Better Off in a Shelter? A Year of Homelessness & Housing among Status Immigrant, Non - Status Migrant, & Canadian- Born Fam ilies EMILY PARADIS, SYLVIA N OVAC , MONICA SARTY, & J. DAVID HULCHANSKI One significant segment of Canada?s unhoused population is families with children. Within this group are many immigrant and refugee fami- lies. Homelessness and shelter life impose great stress on parents and their children. For immigrants who are also undergoing the stress of adapting to a new environment and a new culture, which may include learning English, the stress is compounded. The problem is particularly acute in Toronto, where almost half of all immigrants settle after their arrival in Canada. Toronto is also one of the highest-cost housing markets in Canada and the city where newco- mers face the greatest affordability problems, and therefore the greatest risk of homelessness. A better understanding of the way in which discrimination contri- butes to homelessness among immigrant and refugee families with children can improve public policy and programs for immigrant fami- lies, thereby reducing family homelessness. We therefore conducted a study of women and their families who were living in shelters, to com- pare the experiences of Canadian-born women, and women who had come as immigrants, refugees or other migrants to Canada. Talking to homeless women The study involved a panel study that followed 91 women staying in family homelessness shelters with their children. Participants were di- vided into two groups: (1) women who had been in Canada between one and ten years, and (2) women who were Canadian-born. Each woman was interviewed three times. The first interview, which took place at one of six homeless shelters in Toronto, was retrospective and focused on the women?s housing pathways and life experiences up to that time. The questionnaire also included measures of discrimination, perceived dis- crimination, and symptoms of stress. The second was shorter and inves- tigated changes in the women?s circumstances since the previous inter- view. The third, completed about a year after the first interview, was an in-depth discussion of their lives and housing situation since the first interview, to identify changes and the reasons for those changes. In par- ticular, we asked about perceived discrimination and other sources of individual and family stress. Those born outside Canada came from 22 different countries of ori- gin. Half were from countries in the Caribbean, nine from Africa, six from Asia, three from Europe, and two from Latin America. Three- quarters had been in Canada five years or less, with the average being 4.7 years. The overall study retention rate was 63 percent. Immigrant women were much more likely than Canadian-born women to complete a third interview. Also, women who had previously been homeless were less likely to complete the study than those for whom this was their first experience of homelessness. Most women in the sample had completed secondary school. Immi- grant women, especially non-status migrant women, reported more edu- cation and credentials than Canadian-born women. The sample was self-selected, and interviews were conducted in English. As a result, the study did not reflect the additional barriers and stresses facing immigrant and non-status migrant women who do not speak English. Table 1: Overview of study participants Time 1 Time 3 Number (N=91) % Number (N=57) % Immigrant 41 45 32 56 Canadian- born 50 55 25 44 Racialized 52 57 35 61 Aboriginal 10 11 overall, 20 Canadian- born 5 9 overall, 20 Canadian- born Age range 19 ? 48 21 - 48 Median age 26.5 27 Over 30 Canadian- born: 21 Immigrant: 13 42 32 Canadian- born: 12 Immigrant: 11 48 34 E xperiences of housing, homelessness, and life in a shelter Many of the women (43 percent) had been homeless in Toronto before, while this was the first experience of homelessness for 57 percent of res- pondents. The Canadian-born women were more likely than the immi- grants to have been homeless before: 65 percent of Canadian-born wom- en had been homeless, compared to 44 percent of immigrants. Immigrant women with permanent resident status and migrant women without status were equally likely to have previously experienced homelessness in Toronto. Respondents reported considerable housing instability in the two years before the first interview, having lived in an average of four places, including the shelter in which they were interviewed. Some respondents had moved as many as eight times in two years. Women?s most common reason for leaving their last stable place of residence was abuse: 30 percent had left because of abuse. Other com- mon reasons for leaving included bad housing conditions and affordabil- ity problems. A few respondents had been evicted by landlords, but many were told to leave by roommates and other cohabitants. Over- crowding, crime and violence, bad physical conditions, and family con- flict were also frequently cited as reasons for leaving. Some women?s reasons for leaving suggest the particular difficulties of women who are pregnant or caring for children in finding and keep- ing housing. Some respondents who could not afford their rent reported that this was due to having lost a job because of their pregnancy, while others were asked to leave by cohabitants or landlords because they were pregnant or because their children were noisy. Two women who had worked as live-in caregivers were evicted by their landlord- employers because they were pregnant or had children. Several women were told by child protection authorities that they had to move from their homes into shelters in order to maintain or regain custody of their children. Others decided to leave conditions they considered unfit for their children or the babies they were expecting, and then were unable to find suitable housing. Most respondents were referred to the shelter in which they were interviewed by a social service agency. Almost all respondents had de- pendent children staying with them at the shelter at the time of the first interview; two were due to give birth within days of the first interview, and had other children who were not at the shelter with them. More than half of the mothers were accompanied by one child at the shelter, one- quarter were with two children, 13 percent had three, and 6 percent had four, for a total of 150 children. Homelessness is connected with family separation in complex ways. At the time of the first interview, 19 percent of respondents were sepa- rated from one or more of their children, and about one in four families were separated at some point during the study. Status immigrant wom- en were less likely to be separated from their children than were Cana- dian-born or non-status migrant mothers. While homelessness and hous- ing problems sometimes cause family separation, some women became homeless in order to regain custody of their children. Some women born outside Canada had children still living in their countries of origin. Some women with older children did not have their children with them ? in some cases, because shelter rules restricted the entry of older children, or because their children had left the family shelter to stay with friends or in a youth shelter. Their comments hinted at the effects of family home- lessness on some homeless youth. Just as many seemingly ?single? homeless women are in fact mothers separated from their children, some youth in homeless shelters are separated from their homeless families. At the time of the first interview, more than three out of four re- spondents said they were satisfied or very satisfied with the shelter they were in: they commented on the helpfulness of the staff, the cleanliness of the shelter, and the safety, comfort, privacy, and independence they experienced. Many appreciated the services that were available, particu- larly childcare. Those who were less satisfied cited shared rooms, crowded conditions, and the noise of shelter life; several were unhappy with the quality of the food. The difference a year makes At the time of the final interview, only two respondents were still in the same shelter. The vast majority of the others had stayed in the shelter for less than one year, with respondents about evenly divided among those who stayed from one to three months, four to six months, and seven months to less than a year. More than 30 percent of the women had received help from shelter staff in finding their current place and about 25 percent had been as- sisted by a housing help centre or subsidized housing provider. About 10 percent had been helped by another agency, and another 10 percent by a friend, partner, or family member. Twenty percent had found hous- ing on their own. Most of the women were housed in above-grade apartments, while some were in basement apartments or houses. A few were in other forms of housing, including transitional housing. Three women were staying in the homes of family members. Nearly all had their children with them in their new housing and about a quarter of the women were living with a partner. This represents a decrease compared with 41 percent who had done so in their last stable place, but an increase compared with 12 percent (of those respondents still in the study at Time 3) who had lived with a partner in the shelter. Of the 16 women who had left home because of abuse, only one was liv- ing with a partner at the time of the third interview, which suggests that women did not return to situations of abuse. Women expressed more satisfaction with their current housing at the time of the third interview than they had with the last stable place they had occupied before entering the shelter: only 44 percent overall were satisfied or very satisfied with their last stable place, while 68 per- cent found their current place satisfactory. Most who said they were sa- tisfied, however, also expressed concerns about their homes. Positive comments reflected the things respondents value most in housing: good conditions, cleanliness, safe neighbourhoods, proximity to amenities, and affordability. The many comments about privacy, quiet, and space suggest that women were comparing their new homes to the shelter, and in some cases women explicitly stated that the best attribute of their housing is that it belongs to them. Most negative comments from all groups reflected women?s concerns about poor maintenance, disre- pair, infestations, and other bad conditions in their current housing. Differences by group: Canadian- born, status immigrant, non-status migrant Canadian-born women Most Canadian-born women had been homeless in the past, and most had lived in at least four places in the preceding two years. Their last stable places were often in such poor condition that it was their reason for leaving. About half were very dissatisfied with their last homes, the lowest satisfaction rating of all groups. Many were forced to leave home due to family conflict with parents, or to secure a safer, drug-free envi- ronment for themselves and their children. Once homeless, Canadian-born women moved around more than the other groups. Most stayed in more than one shelter during the cur- rent period of homelessness. Nevertheless, they spent less time homeless than immigrant women. Almost half stayed in the interview shelter less than four months, and most had moved into or secured a place of their own by the time of the second interview, although they were more likely than the other groups to have moved again by Time 3. This may explain the finding that they were more likely than the other groups to have found their current place with the assistance of an agency, and less likely to have been helped by a shelter. Of all groups, they were least satisfied with the interview shelter, although almost three-quarters were satisfied. At the time of the third interview, women born in Canada lived in the greatest variety of housing forms, mostly above-grade apartments, but also basement apartments or houses. Most lived in units that were two bedrooms or larger, and their households were the largest of the three groups; many lived with a partner, half of them lived with at least one other adult, and most had more than one child. A large majority were satisfied or very satisfied with their current places, although many cited concerns with bad conditions and poor maintenance. They were more likely than the other groups to have problems with their current landlords, and less likely to be unhappy with their neighbourhood. Status immigrant women Immigrant women with permanent resident status tended to have a his- tory of more stable housing, with fewer moves in the preceding two years. Although most had lived in places that were overcrowded, they were more likely than Canadian-born or non-status migrant respondents to live in a house, and less likely to live in a basement. About half had lived with partners before entering the shelter. Many had left their homes because of partner abuse or crises such as job loss or fire. Almost all had moved directly from their last stable home into the current shel- ter, without periods of hidden homelessness or other shelter stays. Once in the shelter, status immigrant women were somewhat less likely to be with a partner than women in the other groups. Their shelter stays were somewhat longer than those of Canadian-born women, but most spent less than six months in shelters. They had the highest rate of satisfaction with the shelter of all three groups. They were less likely than the other groups to have found their new housing on their own, and more likely to have received housing search help from family, friends, or partners. Almost all were satisfied with the help they had re- ceived. At the time of the third interview, most were living with more than one dependent child, in above-grade apartments with two or more bed- rooms. The information available suggests that they were more likely than other groups to have moved out of the shelter into subsidized hous- ing. Like Canadian-born women, a strong majority were satisfied or very satisfied with their new homes, although many had concerns about physical conditions. Non-status migrant women Immigrant women without status had the most unstable pre-shelter housing of the three groups: two-thirds had moved four times or more in the preceding two years. Their last relatively stable homes were often short-term, informal arrangements with acquaintances or extended fami- ly members, in which they lacked security of tenure, and they were vul- nerable to sudden eviction, exploitation, and invasion of privacy. They were less likely than the other groups to have lived with a dependent child at their last stable place, and considerably more likely to have been forced from their precarious homes due to pregnancy. Non-status women had the fewest, and youngest, children of all groups: three-quarters had only one child with them in the shelter, none had more than two children, and many had a baby under one year old. They stayed in the shelter much longer than the other groups, with one- third staying in the interview shelter more than one year. Non-status women?s households and unit sizes were also smaller than those of any other group at the time of the third interview. Com- pared with their situation before moving to the shelter, they were least likely to live with friends or family members, and most were the only adult in their households. The majority lived in one-bedroom apart- ments, and none lived in large places with three or more bedrooms. At first glance, it appears that non-status women?s housing after leaving the shelter was more stable than that of other groups: none had stayed in any other places between the interview shelter and their cur- rent place at the time of the third interview, and they were more likely to have been in their new place for at least seven months. Their satisfaction ratings, however, suggest that lack of options might be a more accurate explanation than stability: unlike the other groups, most were unsatisfied with their current place. Discrimination The questionnaire included three measures of perceived discrimination. 1. Dealings with other people: respondents rated the frequency with which they believe they have been treated in negative ways by oth- ers in the past year. 2. How society views me: respondents were asked to agree or disagree with five statements about their sense of belonging in Canadian so- ciety and general societal attitudes towards them. 3. Life events: respondents identified specific events of discrimination experienced in the past year in housing, employment, and social ser- vices, and the grounds for that discrimination. At the time of the first interview, negative dealings with other peo- ple were a regular occurrence for most respondents. Three-quarters said they had been treated with less courtesy than others, more than once a month for the past year. More than two-thirds said they had been regu- larly treated with less respect than others, and that others acted as if they were better than themselves. More than half said that other people regu- larly acted towards them as if they were not smart, and that they had been threatened or harassed. One year later, overall ratings for regular occurrences of all forms of negative treatment by others had declined. Ratings for daily occurrences of such treatment decreased especially sharply, with very few women reporting daily negative interactions. Most notably, none reported daily threats and harassment, though almost one in four had reported this at the first interview. However, rates for non-racialized women declined more in most categories than did rates for racialized women. The majority of respondents felt excluded from or judged by society in some way. There was little change across the three interviews in over- all rates of agreement with statements such as ?I feel that Canadian soci- ety discriminates against me,? and, ?I feel that I am not given opportuni- ties that are generally available to others.? And one in five agreed with the statement, ?I feel that I don?t belong in Canadian society.? This sug- gests that whether homeless or housed, women?s sense of social exclu- sion remained consistent. Women also reported their experiences of discrimination in trying to secure housing, or find a job, or in dealings with social service agen- cies, educational establishments, or the police. Discrimination particular- ly affected women?s access to housing: almost one in three women (31 percent) said that a landlord had discriminated against them in the year preceding the study, and almost one in four (23 percent) had been dis- criminated against by a landlord during the time they were homeless and searching for housing. In general, events of discrimination decreased over the course of the year, with the exception of non-status migrant women, many of whom were caught up in the process of trying to achieve immigrant status, a process involving intrusive questioning and difficult dealings with authorities. What got better; wha t got worse At the end of the third interview, respondents were asked to comment on significant changes in their lives over the course of the study: What had changed? What had improved? What had worsened? Almost three out of four respondents said that their housing situa- tion had improved in the previous year. Many women also reported im- provements in family (35 percent) and community (27 percent). One in five respondents reported improvements in income, and the same num- ber said they were pursuing their education. About one in seven said their employment situation had improved. Most women reporting other improvements also said that their well-being had improved in the past year. Women described feeling happier, more hopeful, less stressed, and more confident and indepen- dent, compared to how they had felt when homeless. While almost all respondents reported improvements in their lives, nearly everyone said that some things had gotten worse. One of the most common concerns among women in the study was threats and stalking by ex-partners. Many respondents had left home due to family violence. Now that they were no longer in a shelter with a confidential address and round-the-clock staff and security, many were dealing with in- creased unwanted and threatening contact by their ex-partners. Some women were unhappy with their new housing. Common con- cerns included poor maintenance and dangerous neighbourhoods. Some women encountered barriers trying to re-enter the labour market. Some respondents felt they were financially better off in the shelter, since their apparent increase of income from personal needs allowance to welfare benefits was completely consumed by housing and other necessities, making it more difficult to provide for their children than it had been in the shelter. Several women also noted that since moving into their own homes, they had lost access to services they still required. What we learned This study confirmed conclusions reached by other researchers into homelessness, particularly family homelessness. The fundamental causes of homelessness are low incomes, high housing costs in the private rental market, and insufficient subsidized housing to meet the needs of people living in poverty. Moreover, our findings confirm that women?s incomes from employment and social assistance are often too low to provide for adequate, safe, stable housing for themselves and their children, and that lone mothers? low incomes and lack of access to housing are exacerbated by discrimination in housing and job markets. Violence, and especially partner abuse, is a significant precipitator of homelessness among wom- en, and inadequate housing and employment prospects also expose women to the risk of further abuse and sexual exploitation. However, this study reached some new conclusions that should in- form both further research and the actions of those who work with homeless families and with immigrants. 1. The intersection of homelessness and lack of permanent resident status This study systematically analyzed for the first time the causes and ef- fects of homelessness for women living in Canada without permanent resident status. Toronto is not only Canada?s largest immigrant reception centre, it is also an important destination for refugee claimants, many of whom spend years here before their claims are decided, and some of whom will never be granted permanent status. Likewise, Toronto and its surrounding areas are home to a large percentage of people admitted to Canada as temporary workers. Temporary workers account for about half of all people admitted to Canada each year, but most never become permanent residents, and many are subject to severe limitations in their employment options, housing, and access to social benefits. Finally, as a large metropolitan area, Toronto is also home to many people who live and work in Canada with no legal status, though their numbers cannot be known. This study showed that women without status ? whether they are temporary workers, awaiting resolution of a refugee claim, or living ?underground? ? are extremely vulnerable, often living in conditions of deep poverty, housing instability, danger, and exploitation. Because they have limited access to social assistance, health care, and other social ben- efits, non-status women must rely on under-the-table employment or the compassion of others to secure housing. Pregnancy and childbirth represent a crisis for non-status migrant women, making employment impossible, incurring health care costs, and disrupting precarious housing arrangements. Having nowhere else to turn until they are able to return to work, these women are usually forced to go to family shelters with their babies and young children, who may be Canadian-born. Once there, they are required to try to regularize their status, although many will not qualify as refugees, and their cases for Humanitarian and Compassionate status are generally considered weak. Some are deported, while others wait years and spend substantial sums in legal and administrative fees before they and their families can enjoy a life of stability. The status regularization process is so protracted and complex that few mothers in this study had reached the end of it by the time of the final interview. Some of the findings, though, hint at the benefits and costs of this process for many non-status women. On the one hand, women and their children gained access to social assistance, health care, work permits, and other social benefits that improved their incomes and stability. On the other hand, the study revealed that migrant women ex- perienced decreased levels of employment, increased perceptions of dis- crimination, and reduced mobility while attempting to gain status in Canada. 2. Family shelters as transitional and supportive housing A second important finding of this study is that family shelters ? in- tended as a crisis resource of last resort ? are instead functioning as tran- sitional and supportive housing for certain types of families for whom dedicated housing programs are needed. This critique often has been made about shelters for women and families fleeing violence. In this study, it appeared that homeless shelters were functioning in a similar way for other groups. Non-status migrant women may demonstrate a strong ability to maintain housing and employment, sometimes for years, without access to services of any kind; but when pregnancy, childbirth, violence, and other crises disrupt their precarious jobs and housing arrangements, they have nowhere to turn but shelters. Their long shelter stays suggest that they would be better served by a housing program in which they could live with their children while undergoing the status regularization process. Such a program should be more home-like than a shelter, incor- porating separate living quarters, food preparation space, and less regi- mentation, so that families may maintain autonomy. This program need not be as resource-intensive and costly as a shelter, which requires round-the-clock staffing. Another group in need of a more appropriate residential program is that of mothers involved with child protection services. Some mothers in this study were told by child protection authorities that to maintain or regain custody of their children, they had to leave housing that was con- sidered unsafe. While for non-status migrant women, the intensive staff- ing and regimentation of the shelter are intrusive and unnecessary, these qualities of the shelter are what child protection agencies want for these mothers. Again, the use of crisis shelters to fill a specific need for high- support, intensively supervised housing for mothers and children at risk suggests the need for dedicated services. 3. Comparing shelters and independent housing: a series of trade -offs Our third conclusion is that, in some respects, women were often better off in the shelter than they were in their own homes. Women?s pre- shelter housing was often unaffordable, unsafe, inadequate, isolating, and in poor condition; women lacked adequate incomes to provide for a better home or other necessities; and many women did not have access to needed services such as childcare, advocacy, and housing search assis- tance. Unfortunately, for most women, their post-shelter housing, in- comes, and service access represented only a partial improvement. In both pre- and post-shelter housing, women were faced with trade-offs and compromises: dangerous locations in exchange for affordability, poor physical conditions in exchange for lack of discrimination from neighbours and landlords. The shelter itself sometimes represented a trade-off: overcrowding in exchange for food security; regimentation in exchange for safety; lack of autonomy in exchange for access to services. Shelters represent an invaluable service, offering an environment of relative safety and stability in which women and children may recover from crises and violence, gain access to services, and search for new homes. At the same time, what does it mean for women and children to be, at times, ?better off? in a shelter than in their own homes? What are the costs, both financial and human, of using shelters as a catch-all for families with widely varying needs and capacities? What are the ramifi- cations of forcing vulnerable women in poverty to choose between au- tonomy and access to services? How are mothers and children affected, psychologically and socially, when they are forced to reside for extended periods in situations of overcrowding, scrutiny, and the stigma of the label ?homeless?? And what are the long-term prospects for stability for families who leave the shelter, but continue to face the same barriers of poverty, inadequate and unaffordable housing, discrimination, violence, and lack of access to childcare and other services, which caused them to become homeless in the first place? Homelessness is neither inevitable nor natural. Each time a family becomes homeless represents a failure of services and supports to keep them housed, and suggests a gap which must be filled. The following directions for policy suggest some initiatives which could begin to fill some of the gaps that led to homelessness for families in this study. Directions for public policy Income support Most women in the study received no child support from their children?s fathers. Those who did receive it sometimes did not receive the full amount on a regular basis, even though the full amount was clawed back from their social assistance payments. Also, most families in the study had multiple income sources. Finally, some mothers in the study became homeless because their student loan entitlements were insufficient, but rendered them ineligible for welfare. Families? complicated incomes from multiple sources underline the need for a guaranteed income bene- fit that tops up all other income sources to a level that is adequate for sustaining stable housing, food security, childcare, and other necessities. Housing All levels of government must act quickly to increase the supply of sub- sidized housing. Crisis shelters have become de facto transitional housing for specific groups who require long-term residential and other supports. Instead of being forced to live in overcrowded conditions in crisis shel- ters, these families need access to appropriate housing to meet their needs, especially non-status migrant women and women who are in- volved in the child protection system. Human Rights Housing is recognized as a human right in international treaties to which Canada is a signatory. Recent extensive consultations by the Ontario Human Rights Commission found that discrimination on the basis of income, family status, race, and other prohibited grounds interferes with access to rental housing. The findings of our study support and extend the recommendations of that report. In particular, we learned that women often knew that they had experienced discrimination in housing, but were not aware that they could pursue remedies. In a few cases, re- spondents were not even aware that some common landlord practices ? such as applying rent-to-income ratios ? constitute discrimination. At the same time, some landlords may not be aware that refusal to rent on the basis of grounds such as family status or receipt of social assistance is prohibited by law. Education of both tenants and landlords is needed. Services Although most of the women we spoke to were satisfied with their shel- ter stay, many respondents stated that they had been treated unfairly in shelters, welfare offices, and other services. Women need to be informed about all social assistance and housing benefits available to them and the means to request them. At the same time, front-line service providers should have sensitivity training, based upon input of service users. Res- pondents also stated that they did not have access to services they needed, or lost access to services once they were housed. Sometimes they were led to expect follow-up services after they were housed, but these services never materialized. Existing shelters need funding to offer direct housing search and accompaniment services and provide follow-up for at least one year while families re-establish themselves in the communi- ty. Immigration Women living without permanent resident status encountered barriers as they attempted to make a stable home for themselves and their child- ren. Increasingly, advocates for nonstatus and temporary workers rec- ommend that the federal government regularize status and ensure access to services for all persons living and working in Canada. They also rec- ommend that labour protections and benefits extend equally to all work- ers, including temporary workers; and that non-status persons have access to all health, crisis, and other services without fear of being re- ported to immigration authorities. Our study identified a need for a cen- tralized source for information and advocacy for persons seeking to re- gularize their status. Finally, family planning, prenatal, labour and delivery, and postnatal care should be available free to all mothers, whatever their status. Childcare Access to childcare is vital to ensure that women can take up opportuni- ties for employment and education. Childcare as it is currently struc- tured does not meet the needs of low-income women whose jobs are of- ten temporary, part-time, casual, shift work, or home-based. Childcare subsidies that are flexible to allow for varying schedules and varying forms of childcare (part-day, part-week, before- and after- hours, drop- in) would allow parents to use childcare as needed and improve their employment situations. E mily Paradis is research manager at the Cities Centre, University of Toronto, and a Ph.D. candidate in Adult Education at the Ontario Institute for Studies in Education. She has worked in research, advocacy, activism, and service provi- sion with women who are homeless for more than 15 years. Sylvia Novac is research associate at the Cities Centre, University of Toronto. Monica Sarty is research assistant at the Cities Centre. She holds an M.A. in Sociology from the University of Toronto. Her focus is on methodology and material inequality. David Hulchanski is associate director of the Cities Centre. He teaches gradu- ate courses and supervises dissertation research on housing policy, social wel- fare, community development and human rights in the Faculty of Social Work. Based on Better Off in a Shelter? A Year of Homelessness & Housing among Status Immigrant, Non-Status Migrant, & Canadian-Born Fami- lies, Research Paper 213, published by Cities Centre, July 2008 Chapter 4.3 Making the invisible visible: Canadian women, homelessness, and health outside the ?big city? CAR OLYN WHIT ZMAN Introduction This article is an exercise in ?making the invisible visible? (Sandercock, 1998). ?Invisibility? has not received the comprehensive analytical atten- tion it deserves in spatial literature. However, invisibility is a common reference point, especially in feminist and postmodern literature on space and place. For instance, Sandercock (1998) refers to the historical and contemporary invisibility of racialized groups, women, and other subject peoples as well as the invisibility of radical ideas in her introduc- tion to ?insurgent planning historiographies.? Sibley (1995) also refers to exclusionary practices that have resulted in the invisibility of ideas and people, in both the past and present. Power is expressed by the monopo- lization of space, both in the literal sense and in the world of policy dis- course (Sibley, 1995). The notion that the lives of poor women are invisible to policy mak- ers has been expressed in the feminist literature since the early 1980s. Hays, who writes about ?invisibility and inclusion? in her book on low- income U.S. women, emphasizes that the women she spoke to ?told their stories ? with the hope that they would be recognized not simply as a composite of clich?s, but as whole persons ? citizens and social me m- bers? (2003, p. 139). What these references to invisibility have in common is a set of multiple meanings embodied in the term. Individual people?s interior lives, their reasons for being at a particular place in their lives can be made invisible, misrepresented, or simply not cared about (for example, the ?welfare queen? stereotype promulgated in the United States in the 1990s). Groups of people can be made invisible in both the spatial sense (?visibly? homeless people excluded from using public space) and in the policy sense (?hidden? homelessness not counted in statistics). Although feminists have been providing a gendered perspective on housing and homelessness in English-speaking developed nations since the mid-1980s (McClain & Doyle, 1984; Watson & Austerberry, 1986), popular conceptions of homelessness in these nations still focus on the ?visibly homeless? ? those who ?stay in emergency hostels and shelters and those who sleep rough in places considered unfit for human habita- tion? (Novac, 2001, p. 3). Less visible to public scrutiny and policy initia- tives are women and men who face ?hidden homelessness,? those ?who are temporarily staying with friends or family? and those living in households where they are subject to family conflict or violence,? or who are at risk of homelessness because they are ?paying so much of their income for housing that they cannot afford the other necessities of life such as food; those who are at risk of eviction; and those living in illegal or physically unsafe buildings, or overcrowded households? (Novac, 2001, p. 3). While men are more likely to be visibly homeless, sleeping rough or in shelters, women?s homelessness is largely hidden from pub- lic view (Novac, 2001). A second dimension to the ?invisibility? of women and homeless- ness is the focus on visible homelessness in the centres of large cities. As Cloke, Milbourne and Widdowfield have pointed out in the British con- text, ?the spatiality of homeless people is entirely encompassed by city limits? (2000a, p. 716). Soup kitchens, drop-ins, shelters for homeless people and other visible symbols of homelessness are concentrated in downtown areas of larger cities. Suburbs, small towns and rural areas share an idealized image of ?old-fashioned? communities, ?purified space? free from the lack of social cohesion and control that is assumed by some conservative commentators to breed homelessness (Cloke, Mil- bourne, & Widdowfield, 2000b, p. 715; see also Sibley, 1995). In Canada, a growing research focus on housing affordability, on those who spend more than 30 percent of pre-tax income (or in a more extreme measure, spending more than 50 percent of pre-tax income) as an ?indicator of changes in the risk of homelessness? (Moore & Skaburskis, 2004, p. 397, emphasis in original) has only recently shifted attention to ?stressed households? living in suburban and ex-urban locations (Bunt- ing, Walks, & Filion, 2004). While there is a higher number of single- person households facing affordability stress in inner cities, single-parent households are spread more evenly across metropolitan areas. Local and senior government policy responses have not yet caught up with this demographic reality, leading to a mismatch between where housing af- fordability stress is ?produced? and where services for the homeless and those at risk of homelessness are ?consumed? (Bunting et al., 2004). A third dimension, and a focus of this chapter, is the invisibility of the ?landscapes of despair? (Dear & Wolch, 1987) within women who are homeless and at risk of homelessness. A recent literature review by the U.S. National Organization for Women concludes that the primary cause of homelessness among women in developed nations continues to be inadequate affordable housing and insufficient income, a situation which is often set into motion by physical abuse by a male partner (NOW Legal Defense and Education Fund, 2002). The World Health Organization?s landmark report on violence and health (Krug, Dahlberg, Lozano, Mer- cy, & Anthony, 2002) cites national and cross-cultural studies that show partner violence has huge impacts on physical, reproductive, and mental health, as well as economic impacts such as lessened ability to stay in paid employment and lower personal incomes. Given the links between domestic and sexual violence, poverty, and homelessness, it is hardly surprising that impoverished women often speak of extreme physical and mental stresses associated with keeping their lives and their fami- lies? lives together. Recent Canadian and international policy approach- es use WHO?s integrated, holistic and values-based definition of health, going beyond the absence of illness to describe a state of physical, men- tal, spiritual, and social wellbeing (Shah & Hodge, 1997; Thurston & O?Connor, 2002). While more affordable housing and better income and social supports are central to solutions of homelessness, the health care system plays an important role in preventing homelessness and in assist- ing women who are episodically homeless. As bell hooks says: ?you can?t effectively resist domination when you are all messed up? (1990, p. 218). Methods This article aims to combine the policy-based and empirical work under- taken for a study in Ontario, with insights gained from post-structural theoretical literature, particularly from feminist geography, in order to illuminate the relationships between gender, health, homelessness, ser- vices, and space. It draws upon research conducted for an Ontario Wom- en?s Health Council (OWHC) report on integrated health services for women who are homeless or at risk of homelessness (Inner City Health Research Unit, St Michael?s Hospital and Oriole Research and Design, 2004). The client organization was established in 1998 to ?advise the Mi- nister of Health and Long Term Care; ? advocate for improv ements to women?s health care in Ontario; promote, influence, and disseminate research into women?s health issues; and to reach out and empower women across the province to make informed decisions that will contri- bute to improvements in their health.? In consultations with health-care providers across the province in 2000-2001, the OWHC heard that poverty was a key determinant of women?s health. The OWHC commissioned a research study on the health issues of women who are homeless and at risk of homelessness (OWHC, 2002a) and also held a one-day ?think tank? with health-care workers in November 2001 (OWHC, 2002b). At the think tank, the im- portance of providing integrated health care for women who are home- less or at risk of homelessness was a major theme. While the initial terms of reference requested only a North Ameri- can literature review and survey of health services across the province, the research team (of which I was a member) decided to organize four focus groups with female clients of health services, to elicit their opinions on quality of current health care and ideas for improvement. The loca- tions for the focus groups were selected with the aim of speaking with a range of women across the province, and were organized in partnership with agencies serving low income women. One focus group was held at the Parkdale Activity and Recreation Centre, a drop-in centre for homeless people in west downtown Toronto with an on-site health clinic. A second focus group was held in a com- munity health centre in Oshawa, an industrial city 50 kilometres east of Toronto with a population of 150,000. The third focus group was held at a street health clinic specializing in services for substance abusers and/or people with mental health issues in Kingston, a city of a little under 100,000 people, 250 km east of Toronto, known for its prisons and uni- versities. The fourth focus group was at a family services centre in Hali- burton, a small town of 5,000 people 250 km due north of Toronto. Hali- burton, adjacent to Canada?s most popular outdoor recreation area, Algonquin Park, is heavily dependent on ?cottage country? tourism. The 40 participants in the OWHC focus groups ranged in age from teenagers to women in their late sixties. The women were also varied in their marital and family status, ethnic backgrounds, and physical and mental health histories. Many of the women in the Haliburton, Kingston, and Oshawa groups had custody of dependent children, although sever- al had lost custody at some point due to housing crises. Several of the women in the Kingston and Parkdale groups spoke of not having custo- dy of dependent children. All the women had direct experience of home- lessness or being at risk of homelessness. Their experiences ranged from living on the street, to extended periods without secure private accom- modation, to one woman who still owned the marital home, but was at imminent risk of losing it. Because of the small sample size, I was unable to break down responses by self-identified ethnicity and family status, but multiple burdens (e.g., low income combined with care of a mentally ill dependant) were referred to in the focus groups. In the sections below, I focus on the words of women in the three focus groups outside central Toronto to describe their own experiences of invisibility, and strategies for obtaining health services within these interlocking structures of invi- sibility. Homelessness Inside and Outside the ?Big City? Homelessness is a growing crisis around the world, although the absence of systemic and comparable data makes quantification difficult. Demand for emergency shelter beds and other stop-gap services such as food banks is growing, even as the number of these services increases. How- ever, a state of denial over the existence of homelessness and absolute poverty outside the centres of large cities continues to be fed by politi- cians and developers (for U.K. and U.S. examples, see Cloke, 1997). In 1991, the head of the London (Ontario) Development Institute, told the Commission on Planning Reform in Ontario to forgo ?Toronto solu- tions,? such as a requirement to make 25 percent of all new housing de- velopments affordable to low- and moderate-income households, as ?af- fordability is not an issue in London.? (New Planning News, 1991) Ten years later, the City of London, Ontario, with a total population of 300,000, was serving 4,000 people per year in its shelters and had de- mand sufficient for over 2,000 more beds. In 1997, the successful candi- date for mayor of the newly amalgamated City of Toronto said, ?There are no homeless people in North York,? the suburb he had governed for 20 years. That same night, the body of Linda Houston was found in the washroom of a North York gas station, where she had sought shelter from the cold (?A Loving Mom,? 1997; ?Remembering the Homeless,? 1997) Homeless women are of every age and socio-economic background, but there are certain risk factors within the broad gender category of ?women.? Gender, radicalization, ethnicity, physical ability, life cycle stage, sexual orientation, and access to income form interlocking ?struc- tures of constraint? that limit access to social goods and choices for indi- vidual women (Young, 2002; see also Jacobs & Fincher, 1998). For in- stance, the areas where housing affordability stress is concentrated in the Toronto region correlate with high levels of new immigrants and refugee claimants, above-average rates of unemployment, and a high proportion of single-parent households. There are two wedge-like arcs of social de- privation, which start in the east part of downtown and the northwestern inner suburbs and radiate outwards (Bourne, 2000). In Canada, census areas with a high proportion of Aboriginal people also tend to have high levels of housing affordability stress (Moore & Skaburkis, 2004). Almost three-quarters of Aboriginal single mothers live below the poverty line in Canada (OWHC, 2002a), housing conditions on many reserves are overcrowded, unsanitary and unheal- thy (CERA, 2002), and Aboriginal women are more likely than other women to be sleeping rough in the centres of big cities (Novac, Bour- bonnais, & Brown, 1999). Newcomers to Canada, particularly refugee claimants, are at risk of homelessness, and lack of credit and employ- ment histories in Canada leads to difficulties in accessing private rental accommodation (Access Alliance Community Health, 2003; CERA, 2002). Visible minority women are nearly twice as likely as non-visible minority women to live below the poverty line, and racial discrimination restricts access to housing (CERA, 2002; see also Informal Housing Re- search Network, 2003). Over two-thirds of women with disabilities or chronic health problems live below the poverty line (Chouinard, 1999). Young women form an increasing cohort of the visibly homeless, with almost a quarter of shelter admissions in Toronto comprising young people between the ages of 15 and 24 (CERA, 2002). As in the case of re- cent immigrants, the lack of credit and employment histories is a barrier to accessing rental housing (CERA, 2002). Lesbians are over-represented among homeless young women in Toronto, but there are no comparable data for Ontario or Canada (Novac, 2001). The poverty rate for all single women, including mothers with young children, almost triples after divorce or relationship break-up, and many women who leave their spouses (including victims of wife assault) move in with relatives or friends immediately after separation, which may be the beginning of a spiral into long-term homelessness (Novac, 2001). Women over 60 years old are less likely than their male counter- parts to own homes, and approximately half live below the poverty line (CERA, 2002). Chronic or long-term homeless women, those who have lived for more than a year in shelters or other short-term accommoda- tions tend to be older women, many of whom have severe mental and physical health problems, often compounded by addictions (Novac et al., 1999). Women, homelessness, and health: The internal lan dscape of despair Homeless women in Canada are subject to nutritional deficiencies, expo- sure to pollutants and extreme temperatures, lack of access to basic ser- vices such as a telephone, lack of money for basic hygiene products (toothbrushes, soap, menstrual supplies, and so on), insufficient sleep and other by-products of extreme poverty and lack of stable housing. Because of these poor living conditions, homeless women are subject to higher rates of almost every disease and poor health condition, as com- pared to the general female population (Ambrosio, Baker, Crowe, & Hardill, 1992; Craft-Rosenberg, Culp, & Powell, 2000; Hwang, 2001; No- vac, 2001). A recent literature review on the risk of death among home- less women in Toronto found that the mortality rate for homeless wom- en under 45 years was about five- to thirtyfold higher than in the general population of housed younger women (Cheung & Hwang, 2004). Along with health issues common to both genders, homeless wom- en are canaries in the coal mine of gendered health concerns. Physical and sexual violence is a common experience in the histories of homeless women, both as a precipitator to homelessness and as a result. One To- ronto study found three-quarters of a sample of 84 single homeless women in Toronto had been physically or sexually abused, usually by a male family member, prior to becoming homeless (Novac, 2001). Almost half the women in a 1992 Toronto street health survey had been physical- ly assaulted in a 1-year period, with 21 percent reporting sexual assault (Ambrosio et al., 1992). In addition, North American studies have found that homeless women are at greater risk of abnormal pap smears, STDs, HIV/AIDS and unwanted pregnancy than the general female population (Ensign & Panke, 2002; Weinreb, Goldberg, & Lessard, 2002; Wenzel, Andersen, & Gelberg, 2001). Women who are assaulted by their spouses often report the abuse beginning during pregnancy, which suggests that the onset of pregnancy may be an instigator of abuse and subsequent homelessness among women (Weinreb, Browne, & Berson, 1995). Unsurprisingly, given their greater risk for violence and poor phys- ical health, homeless women are at much greater risk than homeless men or women in the general Canadian population for serious depression and other mental health disorders (Mental Health Policy Research Group, 1997). In all four focus groups, the difficulties of meeting basic needs for shelter, food, clothing and transportation led to a great deal of stress, depression, and anger, which in turn often resulted in physical illness: Getting any decent sleep is a big problem. You are always tired and cold. You can?t think and have no energy. (Kingston) The injustice of it all makes you angry?results in ulcers, indigestion, makes you crazy. You get panic attacks, headaches. You turn into an emo- tional basket case. (Haliburton) The worst is seeing things fall apart, getting to homelessness, and think- ing about being on the street. Before I became homeless, and I could see it coming. It is like a rollercoaster of stress. (Oshawa) Several women with dependent children were particularly affected by stress, compounded by their insecure living situations and their de- pendence on others for housing: It is stressful to be a mom with kids in a shelter. No privacy and a lot of unrealistic and unwanted interference on how to handle kids. Stressed out moms and stressed out kids cause problems. (Kingston) People you are staying with may want to take over how you are raising your kids. (Haliburton) These quotes above suggest an internalized ?reprivatization dis- course? (Fraser, 1989). While some women were extremely aware of what external supports might help them in their present untenable situa- tion (particularly housing and income supports), their day-to-day stresses were exacerbated by feelings of self-blaming. This was especially true among two women who described wealthier and more secure pasts, and who could not help wondering what they might have done to change their current situation: I didn?t think I would lose what I had, what I had built up, and have to start again with almost nothing. (Oshawa) I feel it is my fault. I have to blame someone, who else is there to blame? (Oshawa) As Hays (2003) points out, the message that women?s poverty is an outcome of personal ?bad choices,? a common trope in welfare rhetoric, has triumphed, not only within the policy realm, but also within wom- en?s perceptions of their own lives. Invisibility Within the Health Care System Several recent North American studies have asked homeless women about barriers and enablers in relation to health care access (Acosta & Toro, 2000; Craft-Rosenberg et al., 2000; Ensign & Panke, 2002; Hatton, 2001; Kappel Ramji Consulting Group, 2002). One theme that emerges from these studies is that there are significant differences between home- less women and men in relation to comfort with emergency housing provision. Homeless women, whether single or with children, prefer strongly to avoid shelters, including shelters for abused women. This is because of legitimate concerns for their safety, and also because they wish to avoid the stigma and disruption caused by leaving their imme- diate environs, especially if they have children in school (Hatton, 2001). Several women in the Haliburton, Kingston, and Oshawa groups spoke of hiding their homeless status, partly because ?coming out? as living in insecure accommodations would be harmful to their children: You want to protect your kids from being stigmatized or teased at school. (Haliburton) If you have kids, there is the danger of [Children?s Aid Society] involv e- ment as soon as you get services from an agency. You can lose your kids. When the crisis is over, you are dropped from the CAS caseload, but the record stays with you and can be used against you later in custody issues. (Kingston) This dilemma echoes the findings of other North American re- searchers, who have found that women often stay in an abusive relation- ship or double up with family and friends, rather than seeking refuge in a shelter, even if the shelter has access to health, housing, and legal ser- vices that are otherwise unavailable. The recognition that women?s homelessness is largely hidden has important implications for access to health care services, and may provide one of the reasons why low- income women with dependent children and fewer social supports are so much less likely to utilize health care options than low-income men or higher-income people (Acosta & Toro, 2001). Given the extremely limited options available, women may ?choose? to avoid giving control to po- tentially intrusive and insensitive health, judicial, welfare, and housing systems. For instance, several women in the focus groups described the measures that they took (credit card debt, appeals to welfare authorities) so that they could avoid moving from their original home, even through that home was no longer affordable to them. As the Ontario Women?s Health Centre had already discovered, health services are often scattered in agencies throughout rural Ontario (with attendant transportation problems), and lengthy waiting lists for many services, such as free dental care, alcohol and drug treatment. There is a shortage of doctors in many rural and isolated areas, leading to closed patient lists. Simply not knowing about a service is a common barrier in all settings (OWHC, 2002a). Another Canadian study has found that physical access to appropriate services is especially acute among small town, rural, and isolated women, many of whom lack access to transportation, do not have family or social networks to rely on, and lack access to a telephone (Craft-Rosenberg et al., 2000). Several women in the focus groups complained of how local medical clinics pro- vide inconsistent, incomplete and insensitive service: The local clinic is staffed by residents, so it is never the same doctor. No consistency. You have to repeat your health history each time you visit. I?d have to be pretty ill before I go. (Haliburton) Not enough good family doctors. Where I used to live, my doctor was three or four towns away. I used the walk-in clinics, where their whole at- titude is ?NEXT!.? They don?t have time to ask you any questions. (Oshawa) I was concerned about my underweight baby, and visited the local clinic. They didn?t show concern, but when I went to a Bobcaygeon doctor [50 km south], he immediately referred me to a pediatrician in Peterb o- rough [another 50 km south]. (Haliburton) There is never any help beyond the immediate crisis at [a local hospital]. (Oshawa) Yet transportation costs to the ?big city? are onerous: I see a doctor in Toronto. Sometimes gas money comes out of food mon- ey, housing budget, kids? costs, or you beg from people you are staying with. (Haliburton) My son has diabetes, and I need to go to Oshawa [150 km south] for a doctor who knows what he is doing. I went to the local clinic when my son was having an episode, and they weren?t listening. Having money for gas has been a problem. (Haliburton) When my baby was born, she had to be in Sick Kids [Hospital in Toronto] for about five months. We couldn?t afford the parking there. We got a pass for a couple of weeks from a nurse and that was it. We brought food, but it wasn?t enough, and the food around there is too expensive. (Osha- wa) In small towns and suburbs, overcrowded emergency rooms (a primary source of health services for many homeless women) and lack of money for public transit fares (assuming that there is public transit) are common concerns for women seeking access to health services (Ambro- sio et al., 1992; Zabos & Trinh, 2001). One older woman in our focus group spoke about how: I got a lift to physio, but it took me two hours to walk home. (Oshawa) Four low-income women in the focus groups who were not on gov- ernment benefits said they could not afford medications or preventive services, the ?benefits trap? described by a number of feminist research- ers (Little, 2002, p. 110): I can?t afford money for cold medicine, so I miss work and then lose more money. (Oshawa) I don?t have money to pay for prescriptions and food, rent and electricity, so I?ve stopped my medication. (Oshawa) No dental care and my teeth are rotting. (Oshawa) I get no help with nutrition as a diabetic. (Oshawa) As Canadian research has shown, some women refuse to see physi- cians because of a past negative experience (Ontario Medical Associa- tion, 1996). Others report that medical staff are judgmental and refuse to treat symptoms and health issues adequately (Ambrosio et al., 1992; En- sign & Panke, 2002; Kappel Ramji Consulting Group, 2002; Zabos & Trinh, 2001). This was echoed by two women in the focus groups: When the public health nurse does her Healthy Baby visits, you are just a number. They don?t want to get to know you, there is no compassion, no real help. (Haliburton) The emergency room humiliated me when I brought in my son without a health card. (Oshawa) North American studies have shown that homeless women access- ing health services, particularly new migrants and visible minority women, report linguistic barriers and cultural biases (Attala & Warming- ton, 1996; Kappel Ramji Consulting Group, 2002), while some people with histories of disruptive behaviour are barred from services (Ambro- sio et al., 1992; Hatton, 2001; Novac, 2001). A few women in the focus groups described turning to alcohol or illegal drugs as a coping mechan- ism, which in turn, increased health and housing stresses: If I go to a shelter, I?m worried that exposure to others will start me using again. (Kingston) If you even mention Street Health, agencies assume you are on drugs. Hospitals think you are a junkie, and dismiss you. (Kingston) Follow-up is difficult for homeless women, who may not be able to afford or access tests, pay for prescriptions, follow special diets or store medication at the correct temperatures. Medical follow-up is very diffi- cult for women who live in temporary accommodations and do not have access to a telephone or mailbox. Frequent moves exacerbate the scatter- ing of medical records, and the constant necessity to repeat symptoms and keep track of previous treatment suggestions (Hatton, 2001; Hwang, 2001; Novac, 2001; OWHC, 2002a; Wenzel et al., 2001). Homeless wom- en?s need for non-acute medical care is often balanced against more im- mediate needs, such as food and shelter, leading to chronic problems being neglected until they become emergencies (Luck et al., 2002). Al- most every woman in the focus groups had a story of how she ignored ill health or medical instructions, simply because there were other financial or time priorities. Finding a Way Out of Invisibility: Homeles s women?s experiences and ideas on integrated services One U.S. summary of barriers to homeless women accessing services describes three common themes: ?not knowing,? ?runarounds? and ?constantly starting over again? (Hatton, 2001; see also Jezewski, 1995). Given the frequent recurrence of these themes in the literature, almost every report on health services for homeless people stresses the impor- tance of integrating health services in order to reduce barriers to access and provide continuous and coordinated prevention-orientated health services. However, there is no agreement about the meaning of the term ?integrated.? In my review of the literature, integration was defined in at least four ways: (1) interagency coordination, (2) co-location of services, (3) case management approaches, and (4) holistic health. Interagency coordination involves integrating service delivery among agencies serving homeless people. Methods range from coalitions that exchange information and undertake advocacy and/or needs as- sessments on issues of common concern, to service delivery teams that coordinate services and may undertake cross-training or develop intera- gency protocols and shared funding mechanisms, to management infor- mation systems that may track clients through shared record keeping, from intake assessment to patient records (Randolph et al., 1997). Intera- gency coordination can occur in any geographic setting, from the centres of large cities to rural areas. Although interagency coordination has been a common recommen- dation of recent policy reports, the most comprehensive evaluation of integrating health services for homeless people suggests that it is not a ?magic bullet.? From 1994 to 1998, the ACCESS project funded 18 centres in 15 U.S. cities serving hard-to-house men and women with severe men- tal illness to determine the effectiveness of strengthening linkages be- tween agencies providing psychiatric care, and those that provided med- ical, substance abuse, housing and income support and employment assistance to their clients (Randolph et al., 2002). There was no significant impact on the housing or health outcomes of the clients in intervention sites, although the evaluation did find that cities with more community social capital, as measured by citizen involvement in organizations, projects, volunteer work and interaction with neighbours, also had stronger network strength (i.e., more effective coordination between agencies) and better housing outcomes for homeless individuals. The city?s housing affordability, as measured by the proportion of house- holds paying less than 30 percent of their income on housing, was also significantly correlated with positive housing outcomes (Rosenheck et al., 2001). The ACCESS evaluation suggests that a focus on agencies? formal linkages and policies may be less effective than a broader and possibly more diffuse community development approach. None of the women in the focus group specifically addressed interagency collabora- tion. Co-location, also known as the ?service hub concept,? concentrates on the geographic co-location of services for homeless people (Dear, Wolch, & Wilton, 1994). Community economic development workshops and other employment-generating activities can be located next to or as part of shelter for abused women or an agency serving homeless people. Hospitals can have social services nearby, while mobile health units can visit shelters. Service hubs can be provided in suburban and small town locations as well as the centres of larger cities. In rural areas, the aim might be virtual co-location through a telephone network of organiza- tions. There was certainly some unprompted support for the service hub concept by women in the focus groups, either in one location or in the form of mobile health units: Provide one-stop shopping: a combination of health and social services all in one building. OW [Ontario Works, i.e. government income support and benefits], food, clothing, health care all together. This would avoid the stigma of having to access the services individually. You could walk in the door and no one seeing you would know what service you were there for. It would cut down on gas costs and would be easier with kids. This type of facility would need to be available in different towns. (Hali- burton) A mobile health and outreach service for street youth and others who are homeless, with primary health care, needle exchange, hygiene basics, ad- vice, info and referrals. (Kingston) The need to integrate health services for children with help for par- ents was also brought up by most of the focus group participants who were single mothers: I wish there was counselling for kids at the same time as counselling for parents. (Oshawa) There needs to be more suicide prevention for teens. My daughter needed to spend three months in a psychiatric hospital before she was assigned a therapist. (Oshawa) The CMHA [Canadian Mental Health Association] is going to help find housing for my son with schizophrenia, and then I can find housing for myself. I had to leave a lot of messages over a couple of weeks, though, before they called back. (Oshawa) At-home help for single moms; help them out with their kids, help them to keep their kids. Help keep them from getting evicted. Provide a care- giving service, to allow moms some time to take care of themselves. (Kingston) In the case management approach, the emphasis is on the individu- al homeless person, where integrated services are facilitated by an indi- vidual case worker, such as a social worker, a primary care provider (a doctor or nurse who provides regular health care to the person) or a team. Shelters for battered women routinely use a case management ap- proach, as do mental health services and drug and alcohol addiction re- covery centres (Attala & Warmington, 1996; Rosenheck et al., 2002). For instance, a pregnant woman who has been battered by a spouse might require assessment and treatment of physical and emotional injuries, continuous prenatal care (including screening for injuries which might have occurred to the fetus), referral to housing and legal services and income support information (Weinreb et al., 1995). Case managers may help by providing referrals to specific services or people, transit fares and detailed directions, advice on behavioural risk reduction and infor- mal counselling (Hatton, 2001). There may be confidentiality and privacy issues related to the case management approach, especially in smaller communities. Two women in the focus groups specifically mentioned being assisted by a case management approach: I?ve been here since the health centre started seven years ago. My family doctor moved here, and I followed her. My husband had just left, my ba- by had severe chicken pox, and I was alone with my three kids. The doc- tor came to my house. I was really depressed, but I didn?t know about depression. I got help and a prescription from the therapist here. The the- rapist and the doctor consult together, with my permission. (Oshawa) My child was taken in care by the CAS, and I had to attend a parenting course here to get her back. Now we are attending couples counselling. (Oshawa) A fourth element to integration has been identified by homeless women themselves: the importance of integrating women?s physical, mental, emotional, and spiritual needs in a holistic health model (Kappel Ramji Consulting Group, 2002). Health services are seen as one strand of a comprehensive web of services that can seek to reverse the vicious cir- cle of homelessness. For instance, women may become homeless because of low income exacerbated by a marital breakdown, which in turn may be related to violence in their past or present lives. Yet lack of decent, stable, safe, and affordable housing may lead to poor physical and emo- tional health, which in turn may prove to be a barrier to actions that might improve income prospects, such as employment or education. Ob- taining housing may be the first step to better health prospects. North American research consistently shows that the further along a woman is in the good housing continuum (sheltered as opposed to unsheltered), the more likely she is to obtain adequate and appropriate health care (Lim, Andersen, Leake, Cunningham, & Gelberg, 2000; Nyamathi, Leake, & Gelberg, 2000). Alternatively, adequate and appropriate health and other social supports provided on the streets or in shelters can help women find more long-term housing options. In contrast to the three other approaches, which stress organizational structures, procedures, and locations, the notion of holistic health is based on respectful listening and choices provided to individuals. Women in the focus groups stressed the importance of better infor- mational and geographic links between services. However, their empha- sis was on the quality of care provided. While staff in several walk-in clinics, including the one in Haliburton, were described as insensitive to women?s needs and inconsistent in their services, other health care ser- vices, including the Oshawa Community Health Centre, were praised for their knowledgeable and empathetic staff: I found out about [other services] through the health centre. They are good at referring to other services. Even when I don?t know what to ask for, the health centre tells me about stuff. (Oshawa) You can always call the health centre. When I was in ?spin-cycle stage,? I knew someone would call me back. (Oshawa) In other words, the women found predictability, security, a sense of caring and useful information at the Oshawa Community Health Centre, a finding that echoes other studies of the community health care centre model in Canada and Australia (Warin, Baum, Kalucy, Murray, & Veale, 2000). The discussion on integrated health service tended to reinforce this holistic and woman-centred model. A common theme in the focus groups was that services should respond to women?s stated needs. Women?s Wellness, a weekly program where the theme is chosen by the participants, was praised by several Oshawa focus group members: Women?s Wellness made a huge difference to me. I felt it was okay to take care of myself. I got another message from Welfare! Flexibility, empathy, and the need for a ?harm reduction? approach were stressed: Show consideration, show empathy, give us choices, make it so we can get ahead and out of this situation. (Haliburton) Talk to the clients. Find out what they want and need. Don?t assume to know better or judge them if they don?t want what you think is best. Have respect for the clients. (Kingston) Protect confidentiality. Recognize a client?s right to get help anonymous- ly. I don?t always want to give my name. Sometimes I just use my street name. (Kingston) Women told stories of how their lives had been transformed through appropriate health care, which they defined as responsive, em- pathic, continuous, and holistic: When my mother died in March, I was really depressed. I stayed at ? a residence for the mentally challenged, but they let me stay for a while. They had 24-hour counselling there, and it really helped. (Oshawa) I was living on the street, and I was sick, but I was afraid of doctors and hospitals. I got sick and went to a shelter. They brought me to the hospital where I was diagnosed with thyroid and diabetes. I ended up staying in the [name deleted] for more than a year. Now I?m living in a shared house with a Christian family. They are very kind. (Oshawa) It is worth noting that in both these cases, ?the rules? were bent: in one case, a woman was not eligible to stay in a residence, and in the oth- er case the length of tenure exceeded the regulations. Conclusion: Locating invisibility and devising responses to it As Kearns points out, home is commonly a site where people feel ?in place? (Kearns, 1991). If a secure home does not exist, or there is violence in the home, then where are the sites that people can begin to feel in place? And what is invisibility if not a condition of being ?not in place?? Women in the focus groups described struggling to survive within these interstices of spatial and policy invisibility. Physical and mental health problems resulted from the enormous stress of not having enough money to cover housing and other costs. Focus on survival meant little time to take care of the self. Fear of being stigmatized and of possibly losing child custody kept several focus group women from revealing the extent of their housing problems. Weak public transit infrastructure, so- cial isolation and low-quality social services combined to create further barriers to seeking help. Walk-in clinics and hospitals, often the sole op- tions outside the big city, were seen as emergency-focused and not pro- viding any continuity of care. Some health care professionals were de- scribed as acting as barriers to women finding the help they need, and there was little choice as to health services in the suburban, small town and rural sites. The possibilities of a space for healing are not limited to locations in the big city. When speaking about integrated health services, a focus on whole family and the whole individual woman was a common theme in the three focus groups, along with a client-centred approach that stresses choices and respect, and continuity of care. In other words, integration was sought within the individual bodies of women and within house- holds, rather than at the policy level of inter-agency collaboration or ser- vice hubs. Community-based services, including health services, can be a space of at least temporary safety for otherwise invisible women: women who are marginalized, excluded, and isolated within the dominant dis- course of neoliberal abandonment of social supports (Staeheli, 2003). The Oshawa Women?s Health Centre served as a model for this kind of inte- grated health services, where women?s needs were heard and responded to, instead of being met with standardized services and a call of ?Next! ? It is difficult to quantitatively measure the positive economic and social impacts of these good services, how they can save lives and improve health outcomes. However, the potential of integrated health services combating the individual and societal invisibility of homeless women should not be underestimated. Homelessness is an issue which cannot be contained within the boundaries of the ?big city,? just as health cannot be contained within the boundaries of traditional medical services. At the edge of what is still a barely visible policy issue, the experts reside in shelters, motel rooms, and trailers, on couches and in sleeping bags. The question is how to tap into their considerable powers of observation and survival, in order to provide the best services in all settings. This chapter is based on ?At the Intersection of Invisibilities,? Gender, Place & Culture , http://www.informaworld.com.proxy.lib.uwaterloo.ca/smpp/title~d b=all~content=t713423101~tab=issueslist~branches=13 - v1313 (4) August 2006, 383 ? 399. Carolyn Whitz man is in the Department of Urban Planning, Faculty of Ar- chitecture, Building, and Planning, University of Melbourne, Australia. Acknowledgements The Ontario Women?s Health Council funded the research on which this article is based. I would like to thank my research associates, Deborah Hierlihy, Alison Hamilton and Stephen Hwang, for their help in formu- lating several ideas in this article. I would also like to thank Liz Bondi, Linda Peake, Ruth Fincher, Fran Klodawsky and three anonymous re- viewers for helpful comments on preliminary drafts. References A Loving Mom. . . Lost to the streets. (1997, October 17). Toronto Star, p. A1. Access Alliance Community Health (2003). Best practices for working with homeless immigrants and refugees. Toronto: Author. Acosta, O., & Toro, P. (2000). Let?s ask the homeless people themselves: A needs assessment based on a probability sample of adults. American Journal of Community Psychology, 28(3), 343-366. Ambrosio, E., Baker, D., Crowe, C., & Hardill, K. (1992). The Street Health Report: A study of the health status and barriers to health care of homeless women and men in the City of Toronto. Toronto: Street Health. Attala, J., & Warmington, M. (1996). Client?s evaluation of health care services in a battered women?s shelter. Public Health Nursing, 13(4), 269-275. Bourne, L. (2000). People and places: A portrait of the evolving social character of the Greater Toronto Area. Toronto: University of Toronto Department of Geo- graphy. Bunting, T. R., Walks, A., & Filion, P. (2004). The uneven geography of housing affordability stress in Canadian metropolitan areas. Housing Studies, 19(3), 361-393. CERA [Centre for Equality Rights in Accomm odation]. (2002). Women and Housing in Canada: Barriers to Equality. Available at http://www.equalityrights.org/cera/docs/barriers.htm. Cheung, A., & Hwang, S. (2004). Risk of death among homeless women: A cohort study and review of the literature. Canadian Medical Association Journal, 170(8), 1243-1247. Chouinard, V. (1999). Body politics: Disabled women?s activism in Canada and beyond. In R. Butler & H. Parr (Eds.), Mind and body spaces: Geographies of illness, impairment, and disability (pp. 269-293). London: Routledge. Cloke, P. (1997). Poor country: Marginalisation, poverty, and rurality. In P. Cloke & J. Little (Eds.) Contested countryside cultures: Otherness, marginalisation, and rurality (pp. 252-271). London: Routledge. Cloke, P., Milbourne, P., & Widdowfield, R. (2000a). Homelessness and rurality: ?Out of place? in purified space? Environment and Planning D, 18(6), 715- 735. Cloke, P., Milbourne, P., & Widdowfield, R. (2000b). Partnerships and policy networks in rural local governance: Homelessness in Taunton. Public Ad- ministration, 78(1), 111-113. Craft-Rosenberg, M., Culp, K., & Powell, S. R. (2000). Health status and resources of rural homeless women and children. Western Journal of Nursing Research, 22(8), 863-878. Dear, M., & Wolch, J. (1987). Landscapes of despair: From deinstitutionalization to homelessness. Princeton: Princeton University Press. Dear, M., Wolch, J., & Wilton, R. (1994). The service hub concept in human ser- vices planning. Progress in Planning, 42, 173-271. Doyal, L. (1999). Sex, gender, and health: A new approach. In S. Watson & L. Doyal (Eds.), Engendering Social Policy (pp. 30-51). Milton Keynes: Open University Press. Ensign, J., & Panke, E. (2002). Barriers and bridges to care: Voices of homeless female adolescent youth in Seattle, Washington, USA. Journal of Advanced Nursing, 37(2), 166-172. Fraser, N. (1989). Unruly practices: Power, discourse, and gender in contemporary social theory. Minneapolis: University of Minnesota Press. Hatton, D. C. (2001). Homeless women?s access to health services: A study of social networks and managed care in the US. Women and Health, 33(3-4), 149-162. Hays, S. (2003). Flat broke with children: Women in the age of welfare reform. New York: Oxford University Press. hooks, b. (1990). Yearning: Race, gender, and cultural politics. Boston: South End Press. Hwang, S. (2001). Homelessness and health. Canadian Medical Association Journal, 164(2), 229-233. Informal Housing Research Network (2003). Living on the ragged edges: Absolute and hidden homelessness among Latin Americans and Muslims in West Central Toronto. Toronto: Metropolis Joint Centre of Excellence for Research on Immigration and Settlement. Inner City Health Research Unit, St Michael?s Hospital, & Oriole Research De- sign (2004). Models and practices in service integration and coordination for wom- en who are homeless or at risk of homelessness: An inventory of initiatives. Toron- to: Ontario Women?s Health Council. Jacobs, J. M., & Fincher, R. (1998). Introduction. In R. Fincher & J. M. Jacobs (Eds.), Cities of Difference (pp. 1-25). New York: Guilford Press. Jezewski, M. A. (1995). Staying connected: The core of facilitating health care for homeless persons. Public Health Nursing, 12(3), 203-210. Kappel Ramji Consulting Group (2002). Common occurrence: The impact of home- lessness on women?s health. Toronto: Brown Books. Kearns, R. (1991). The place of health in the health of place: The case of the Ho- kiaga Special Medical Area. Social Science and Medicine, 33(4), 519-530. Krug, E., Dahlberg, L., Lozano, R., Mercy, J., & Anthony, Z. (2002). World Report on Violence and Health. Geneva: World Health Organisation. Lim, Y.- W., Andersen, R., Leake, B., Cunningham, W., & Gelberg, L. (2002). How Accessible is Medical Care for Homeless Women? Medical Care, 40(6), 510-520. Little, J. (2002). Gender and rural geography: Identity, sexuality, and power in the coun- tryside. Harlow: Prentice Hall. Luck, J., Andersen, R., Wenzel, S., Arangua, L., Wood, D., & Gelberg, L. (2002). Providers of primary care to homeless women in Los Angeles County. Jour- nal of Ambulatory Care Management, 25(2), 53-67. McClain, J., & Doyle, C. (1984). Women and housing: Changing needs and the failure of policy. Ottawa: Canadian Council on Social Development. Mental Health Policy Research Group (1997). Mental illness and pathways into homelessness: Proceedings and recommendations. Toronto: Centre for Addiction and Mental Health. Moore, E., & Skaburskis, A. (2004). Canada?s increasing housing affordability burdens. Housing Studies, 19(3), 395-413. New Planning News. (1991). Newsletter of the Commission on Planning and De- velopment in Ontario. Vol. 1, no. 1, November/December. Novac, S. (2001). Women?s health and homelessness: A review of the literature. In Kappel Ramji Consulting Group (Ed.), Common occurrence: The impact of homelessness on women?s health (Appendix 2). Toronto: Brown Books. Novac, S., Bourbonnais, C., & Brown, J. (1999). Women on the rough edge: A decade of change for long-term homeless women. Ottawa: Canada Mortgage and Hous- ing Corporation. NOW Legal Defense and Education Fund (2002). Surviving violence and poverty: A focus on the link between domestic and sexual violence, women?s poverty and wel- fare. New York: Author. Nymathi, A., Leake, B., & Gelberg, L. (2000). Sheltered versus nonsheltered homeless women: Differences in health, behaviour, victimization, and utili- zation of care. Journal of General Internal Medicine, 15(8), 565-572. Ontario Medical Association (1996). Exploring the health impact of homeless- ness. Ontario Medical Review, May. Ontario Women?s Health Council [OWHC] (2002a). Health status of homeless wom- en: An inventory of issues. Retrieved June 28, 2006, from http://www.womenshealthcouncil.on.ca/English/Homeless-Women.html. Ontario Women?s Health Council [OWHC] (2002b). Strategies to address the health status of homeless women. Retrieved June 28, 2006, from http://www.womenshealthcouncil.on.ca/English/Homeless-Women.html. Randolph, F., Blasinsky, M., Morrissey, J. P., Rosenheck, R. A., Cocozza, J., Goldman, H. H., & ACCESS National Evaluation Team Randolph, F., Morrissey, J., Lam, J., Calloway, M., Stolar, M., Johnsen, M., Ran- dolph, F., Blasinsky, M., & Goldman, H. (1997). Creating integrated service systems for homeless persons with mental illness: The ACCESS program. Psychiatric Services, 48(3), 369-373. (2002). Overview of the ACCESS program. Psychiatric Services, 53(8), 945-948. Remembering the Homeless. (1997, December 7). Toronto Star, p. A1. Rosenheck, R., Morrissey, J., Lam, J., Calloway, M., Stolar, M., Johnsen, M., Randolph, F., Blasinsky, M., & Goldman, H. (2001). Service delivery and community: So- cial capital, service systems integration, and outcomes among homeless persons with severe mental illness. Health Services Research, 36(4), 691-710. Sandercock, L. (Ed.). (1998). Making the invisible visible: A multicultural plan- ning history. Berkeley: University of California Press. Shah, C., & Hodge, M. (1997). Position paper on homelessness and health. Ottawa: Canadian Public Health Association. Sibley, D. (1995). Geographies of exclusion: Society and difference in the West. London: Routledge. Staeheli, L. (2003). Women and the work of community. Environment and Plan- ning, 35(5), 815-831. Thurston, W., & O?Connor, M. (2002). Health promotion for women: A Canadian perspective. Ottawa: Health Canada. Retrieved June 28, 2006, from http://www.hc- sc.gc.ca.proxy.lib.uwaterloo.ca/canusa/papers/canada/ english/promotab.htm. Warin, M., Baum, F., Kalucy, E., Murray, C., & Veale, B. (2000). The power of place: Space and time in women?s and community health centres in South Australia. Social Science and Medicine, 50(12), 1863-1875. Watson, S., & Austerberry, H. (1986). Housing and homelessness: A feminist perspec- tive. London: Routledge. Weinreb, L., Browne, A., & Berson, J. D. (1995). Services for homeless pregnant wom- en: Lessons from the field. American Journal of Orthopsychiatry, 65(4), 492-501. Weinreb, L., Goldberg, R., & Lessard, D. (2002). Pap smear testing among home- less and very low-income housed mothers. Journal of Health Care for the Poor and Underserved, 13(2), 141-150. Wenzel, S., Andersen, R., & Gelberg, L. (2001). Homeless women?s gynaecologi- cal symptoms and use of medical care. Journal of Health Care for the Poor and Underserved, 12(3), 323-341. Young, I. M. (2002). Lived body vs. gender: Reflections on social structure and subjectivity. Ratio, 15(4), 410-428. Zabos, G., & Trinh, C. (2001). Bringing the mountain to Mohammed: A mobile dental team serves a community-based program for people with HIV/AIDS. American Journal of Public Health, 91(8), 1187-1189. Chapter 4.4 Rights to the City: Thinking Social Justice for Chronically Homeless Women Is New York? s homeless strategy the answer for Toronto? F RAN KLODAWSKY ?Mr. Abdella is one of nearly 800 homeless people Toronto says it has helped over the past year and a half by borrowing a new, more aggressive... approach to street outreach pioneered in New York City... which sees outreach workers target homeless people on the street and move them directly into housing instead of shelters. (Gray, 2006, M1) The ?new aggressive approach? described here is a strategy that has come to be known as ?Housing First.? Initiatives similar to this have re- cently captured the attention of businesspeople and bureaucrats, politi- cians and academics, front-line workers and homeless people them- selves, in cities across Canada including Victoria, Calgary, Edmonton, and Ottawa (Cyderman, 2006; Hume, 2006; Lavoie, 2006; O?Leary, 2006; Yedlin, 2007). The basic premise of Housing First is that chronically homeless people1 1 The definition of this term used by the Homelessness Community Ca- pacity Building Steering Committee in Ottawa is ?A chronically homeless need and have the right to regular housing first and 3 foremost, even if they might also require individualized support from social services to stay housed. In these discussions, Housing First is pre- sented as a superior alternative to Continuum of Care. Closely aligned concepts are supported versus supportive housing (see Table 1). Table 1: Concepts and Definitions: Supported (Housing First) and Supportive (Continuum of Care) Housing Supported housing: ? Non-segregated housing managed by a not-for-profit agency. ? Residents have control over where they live and who their living companions are. ? Participating in psychiatric treatment is not a requirement and any support services are provided by an outside agency ? An example is individual or independent apartments. Supportive Housing: ? Offers a continuum of residential facilities managed by not-for-profit agencies. ? Facilities offer varying levels of supervision and social support and ? Residents are often required to be in outpatient treatment. ? Individuals stay in each setting for a limited time and are expected to move up the continuum to independent housing (Parkinson et al. 1999). ? Examples are group homes, halfway house, community integrated living apartments, and supervised apartments. Source: Adapted from Kyle and Dunn, 2008, 8. The standard approach to addressing chronic homelessness, against which Housing First is juxtaposed, begins with the premise that individ- uals with severe mental health and addiction problems are incapable of living in regular housing until these problems are addressed. As a result, a continuum of specialized housing facilities and support is person is one who has spent 60 or more cumulative nights in the past year in an emergency shelter and has reached the point where he or she lacks the physical or mental health, skills and/or income to access and/or maintain housing? (2008, 3). required, in 4 37 lockstep with helping individuals to gain mastery over their mental health and substance use challenges. Access to housing beyond the emergency stage is structured as a reward for movement along a path to sobriety (Tserberis et al., 2004). Housing First, as originally conceived, is a reaction to these Conti- nuum of Care ideas, particularly when they are interpreted in a manner that is rigid and patronizing. There is much to commend Housing First in its contestation of outdated interpretations of Continuum of Care. Ac- cording to some homeless individuals and their advocates, however, the rush to jump on a Housing First bandwagon has been accompanied by a worrying disregard of the considerable benefits that some individuals reap from living in supportive, congregate, or group settings, especially in environments shaped by a flexible harm reduction approach.2 Current Canadian discussions about the relative merits of Housing First and Continuum of Care are not innocent differences of opinion about what works best in helping chronically homeless people to im- prove their lives. Rather, these debates raise both theoretical and subs- tantive questions about neoliberalization as a governmental orientation that promotes certain types of social rule, such as what should be legiti- mate social policy goals for initiatives directed at marginalized popula- tions. In this chapter, I raise the possibility that a wholesale shift to Housing First might well become a vehicle for further excluding margi- nalized peoples, not only in terms of their rights to public space, but also in terms of their visible presence 2 Harm reduction rejects the approach that sobriety is required in order for drug- or alcohol-addicted individuals to leave the streets and become stably housed. Rather it posits that there are effective strategies to reduce the potential harm of substance use problems and that these strategies are compatible with a variety of housing arrangements (Centre for Addic- tions and Mental Health, 2003). in spaces in the city, including the spe- cialized congregate spaces of emergency, transitional, and supportive housing associated with Continuum of Care. Writing to date suggests that neoliberalization tends to promote individualization while it sub- sumes other kinds of goals, such as collective ?rights to the city? (Purcell, 5 2008). Moreover, these examinations have not been sensitive to gender- and race-sensitive differences (Kyle and Dunn, 2008). The shortcomings of roll-back neoliberalism have been ac- knowledged in numerous places, but the question of what comes next in relation to neoliberalization remains to be answered. This chapter is one response to Peck?s observation that: It is no longer really enough to say that... neoliberal states are ?differently interventionist? and that the attendant processes of institutional change are qualitatively rather than quantitatively distinctive; there is a growing need to add content to these assertions, to track actual patterns and prac- tices of neoliberal restructuring, and to make meaningful part-whole con- nections between localized and institutionally specific instances of reform and the wider discourses and ideologies of neoliberalism (Peck 2004, 396). Emerging debates about Housing First as an alternative to Conti- nuum of Care provide fertile ground for such a response, particularly given the rapidity with which Housing First arguments have gained traction among politically powerful actors in the United States and Can- ada, in a manner reminiscent of fast policy transfers in other realms (Peck, 2002, 2005). This chapter has four sections. It begins with an overview of the theoretical scope of this primarily conceptual discussion, highlighting the contributions of Nancy Fraser (1997, 2003), Peter Graefe (2006) and Mark Purcell (2008). Also described are the research and experiential context within which the arguments presented here were developed. The second section explores the context within which governmental interest in Housing First appears to be growing, including a brief overview of Canada?s history of supportive housing. This discussion is followed in the third section with a closer look at the implications of these debates when gender and race-sensitive analysis is foregrounded, and when the particular situations of and challenges faced by chronically homeless women are considered. Conclusions are presented in the final section. 6 Scope and Context Theoretical Frameworks This article is primarily a theoretical and conceptual effort to foster an understanding of how neoliberalization is unfolding in a particular so- ciospatial arena, with insights drawn from feminist and progressive scholars who are preoccupied with what it means to develop, evaluate, and/or implement inclusive and socially just public policies (Fraser, 1997; Flyvbjerg, 2001; Fraser, 2003; Larner, 2003; Bondi and Laurie, 2005; Graefe 2006; Purcell 2008). These scholars have an interest in closely ex- amining and raising questions about inequalities in ?actually existing? places, while focusing on the multi-scalar, socio-economic and political interactions though which such inequalities are exacerbated or reduced (Purcell 2008). In this paper, I draw upon three complementary frame- works: the first assesses the impacts of a social policy vis-? -vis neolibera- lization (Graefe 2006); the second considers group experiences of the im- pacts of social policy, particularly in relation to questions of social justice (Fraser 2003); and the third focuses on the meaning of ?rights to the city,? also from a social justice perspective (Purcell 2008). Graefe (2006) distinguishes between neoliberalism as an ideal type of social rule and actually existing processes of ?neoliberalization,? that produce ?neoliberal state forms...in an uneven fashion? (200). His subs- tantive interest is in investigating the extra-regional influences on Que- bec?s understanding of the social economy and the extent to which three distinct knowledge networks in that province have taken up ideas asso- ciated with the ?American model? of neoliberalism. Focusing on the women?s movement, ?left-centre? intellectuals associated with commu- nity economic development initiatives, and civil servants within the Finance Department, Graefe identifies how each interpreted the social economy differently vis-? -vis furthering neoliberalism as a ?new form of social rule?: ?three different versions of the social economy figure in pol- icy debates, ranging from a feminist one that seeks to break with neolibe- ralization, to one that seeks to flank it, to a third that seeks to roll out market relations? (2006, 198). 7 ?recognize [ing] that the character of activities being undertaken (e.g. , serving markets with low effective demand; employing people with ?low productivity?) require that state financial and technical support be made available over the medium-to-long term (208). He labelled the feminist version as ?countervailing? and observed that certain elements of the women?s movement were interested in pro- moting governmental economic priorities in a manner quite distinct from neoliberalism: ?centred on the idea of meeting needs rather than on prof- it-making? and recognizing ?that social production reaches beyond the formal market economy to include economies of care that require greater recognition and support? (205). The second (which he calls ?flanking?), he links to community-based organizations that helped develop ratio- nales for institutional ?exceptions? and by so doing, contribute to main- taining the integrity of the neoliberal project more generally: Fraser?s (1997, 2003) orientation is quite distinct from Graefe?s, but equally relevant to this discussion. Her concerns are about the impacts of public policies on the lived experiences of particular groups. She asserts that justice is unlikely to be served without simultaneous consideration of three dimensions: redistribution, or targeting material inequalities; recognition, or targeting injustices that stem from cultural differences, and representation, which involves addressing political power imbal- ances (2003). In identifying what she regards as the core values of mod- ern liberal societies (?the equal autonomy and moral worth of human ?Rollout? is a term he uses to describe a third perspective that Graefe associates with Quebec?s ?mainline state ministries? (209). This option refers to Peck and Tickell?s (2002) characterization of the manner in which an initial period of ?rollback? neoliberalism in the United States and Britain was revised over time to respond to new challenges, without significantly modifying its market-driven orientation. In Quebec?s social economy, it ?had a surface resemblance to the flanking approach...but with an eye more firmly on the goal of creating social enterprises capable of meeting new social needs in a productive and cost-effective manner? (Graefe 2006, 209). As will be illustrated below, Graefe?s threefold framework is useful in helping unpack the diverse impulses shaping Canada?s housing and homelessness policies. 8 beings?), she interprets their social justice implications in a manner that Purcell (2008) characterizes as a radically pluralist form of democracy: ?to respect the equal autonomy and moral worth of others one must a c- cord them the status of full partners in social interaction. That? means as- suring that all have access to the institutional prerequisites of participato- ry parity ? above all, to the economic resources and social standing needed to participate on a par with others? In the end, as such matters are highly contentious, the parity standard can only be properly applied dialogically, through democratic processes of public debate (Fraser, 2003, 228). In this manner, Fraser provides a series of strategies to en- courage debate about what would constitute more socially just policies for particular groups. Drawing from these insights, this paper?s argument is that these strategies are as relevant to chroni- cally homeless women as they are to other groups, such as child- ren, despite the commonplace assumption that children are more deserving of governmental resources, given the potential human capital benefits to be gained by directing resources to them over the medium and long term.3 Purcell (2008) articulates the socio-spatial implications of the radi- cally pluralist democratic ?attitudes? that he favours,4 ...reject[ing] the notion that all democratic politics must aim at the common good. Rather they embrace an antagonistic model in which adversaries with unavoidably divergent interests struggle with each other to win a temporary hegemony that favours their agenda. It is a social-movement vision of democracy, one that im- agines distinct movements that act together in networks of equiva- lence (2008, 104). which he de- scribes as: 3 See McKeen (2006) for a critical examination of this assumption. Purcell sees these attitudes as the best hope for challenging the re- gressive aspects of neoliberalism, and he examines the notion of ?rights to the city? with this motivation and perspective in mind. He distin- 4 Purcell?s (2008) version of radical pluralism is distinct from, though not incompatible with, that of Fraser (2003). 9 guishes between a vision of the ?neoliberal city-as-property? and an al- ternative ?city-as-inhabited? (105) and uses the metaphor of ?linchpin? to describe how social movements might work in ?networks of equiva- lence? towards diverse rights to the city. Through this framework, Pur- cell identifies both commonalities and differences in how distinct social movements would begin to work towards ?rights to the city?: commo- nalities insofar as the preeminence of city-as-property is rejected, but differences in the choices made to emphasize particular claims for inha- bitance. He sees these sorts of differences as productive in the sense of helping to translate abstract ideas about social justice into context- specific social, cultural, economic, and architectural claims for what cities should accommodate. Research Grounding These three scholars frame my theoretical approach, which has also been inspired by ?concrete cases in particular contexts? (Flyvbjerg 2001, 143). More precisely, the arguments here build upon more than six years of participant and non-participant observation of deliberations on how best to address and end homelessness in Ottawa and in Canada. As one of two primary investigators for the Panel Study on Homelessness in Ottawa, I am known and trusted within the ?homeless sector.? The Panel Study, which began in 2001, was the first large-scale longitudinal study of its kind in Canada, conducted through a collaborative university- community-local government approach (Klodawsky 2007; Klodawsky et al., 2007). The success of this study was in part dependent on the re- searchers being able to garner a high level of confidence from communi- ty actors as well as from local government staff and politicians. This con- fidence was gained from our research focus and orientation5 but also as a result of our involvement in numerous roundtables and committees that address the problem of homelessness in Ottawa and Canada.6 5 The Panel Study examined the long- and short-term factors that influ- ence why diverse persons might become homeless and also the ease with which such individuals are able to exit homelessness (Aubry et al., 2003; Klodawsky et al. 2007). A 6 They included the Steering Committee of the City of Ottawa?s Community Ca- 10 1 particular moment of significance (for this research) occurred during my participant observation of the deliberations of the Women?s Roundtable, an initiative of the City of Ottawa to encourage new, more coordinated responses to ?problematic? gaps in the City?s service and support infra- structure. The Roundtable was focused on chronically homeless women with multiple mental and physical health challenges. In this context, questions were raised about the tangible and intangible benefits of some types of ?supportive? housing in comparison with ?supported? housing. The knowledge gained through the Panel Study, about women and girls? greater propensity to report mental and physical ailments, as well as childhood sexual abuse, sexual assault and wife battering, highlighted the importance of making theoretical links with concepts of gendered and race-sensitive rights to the city (Novac, 1999; Shaw and Andrew, 2005; Klodawsky, 2006; Whitzman, 2006; Klodawsky et al., 2007). pacity Building Initiative, the Women?s Roundtable on Homelessness, the Re- search and Evaluation Working Group of the Alliance to End Homelessness, the Multifaith Houisng Initiative, the National Network on Housing and Homeless- ness, and the Research Alliance on Canadians? Housing, Homelessness and Health. In each case, the persona I have presented is that of researcher, able to offer resources (such as student and library support), relevant knowledge, and the possibility of engagement in related research activities. In a separate research study, knowledge of Canadian state responses to problems of homelessness was gained through the analysis of documents as well as in-depth interviews with 24 senior and mid-level federal and provincial government bureaucrats and politi- cians involved in helping to establish, shape and revise the National Homeless- ness Initiative. This research was conducted with other scholars who have been examining the links between public policy development and change in Canadian municipalities (Hulchanski, 2006; Young and Leuprecht, 2006). The claims about Housing First and Continuum of Care are com- plex and contradictory when considered from a social justice perspec- tive. This is not surprising, since these debates derive from multiple po- litical locations ? international, national, regional, and local ? and that even within specific jurisdictions, the framing of each approach varies as a result of diverse philosophical and practical considerations. In keeping with Bondi & Laurie?s (2005) argument that ?approaching neoliberalism as a constructed social terrain or field turns notions of inevitability into potential resources in the long-term project of ? deliberalizing? space? (399), such analysis provides additional insights on where intervention might be productive in promoting alternatives to neoliberal thinking. Contextualizing the Growing Interest in Housing First Welfare Reform Since about 1980, North American engagement with welfare reform has had a profound influence on the role of housing as a tool of social policy. Cutbacks in state-sponsored public housing have been linked with the rise of homelessness in both Canada and the United States (Hulchanski, 2002, Shapcott, 2006; Falvo, 2007; National Low Income Housing Coali- tion, 2007), and with the growing focus on explanations that emphasize the shortcomings of individuals, rather than structural matters or value- based assessments linked to citizenship rights to housing (Hutson and Clapham, 1998; Callahan et al., 2002; Shinn, 2007). Mounting evidence has challenged such explanation, suggesting instead that the majority of people who experience homelessness do so briefly and/or episodically, primarily for economic reasons (Shinn, 1998; Aubry et al., 2003; Joint Center for Housing Studies, 2007; Klodawsky et al., 2007). Arguments emphasizing the need for additional safe and secure affordable housing have not, however, been particularly effective in convincing policy mak- ers or politicians that this should be the approach of choice, deserving of far greater public resources than is currently the case (Hulchanski, 2002; Hulchanski, 2006; Shapcott, 2006; Shapcott, 2007). As in other social policy fields, the significance of proactive state in- tervention has been downplayed in favour of institutional arrangements that encourage market responses or public-private partnerships. Given the enormous gap between the material costs of building and maintain- ing housing in a country of climatic extremes such as Canada, and what poor people can afford, the housing affordability gap remains a major reason that people become homeless as well as an area of growing con- cern among municipal politicians (Layton, 2000; Moore and Skaburskis, 2004). In Canada, these problems likely have been exacerbated by inter- jurisdictional disputes between the federal and provincial levels of gov- ernment, together with structurally weak local governments with an ex- 13 4 tremely limited capacity to raise or autonomously manage the funds to which they have access (Carter and Polevychok, 2004). Chronic Homelessness: American and Canadian Responses Along with growing evidence about the significance of economic factors, it has also been found that about 15 percent of those who are homeless make inordinate demands on provincial and municipal health and social services, due to the multidimensional nature of the issues they are deal- ing with, including physical and mental health matters, substance use problems, and concurrent disorders (Tserberis et al., 2004). These indi- viduals have garnered the attention of city planners and local politicians, since the majority of ?street homeless? are drawn from these groups (Tserberis et al., 2004). As cities attempt to present themselves as ?entre- preneurial,? concerns have intensified about the visible presence of bo- dies and activities that are increasingly seen as being ?out of place? (Mit- chell, 2003). Numerous plans have been put into effect, often in the name of urban revitalization, to discipline such bodies and remove them from view (Mitchell and Staeheli, 2006). Typically, these reports do not discuss the individual rights of people who are homeless, or societal obligations to such individuals, despite mounting evidence that circumstances beyond individual control, such as childhood sexual and physical abuse, are often implicated in their circumstances (Novac 1999; Klodawsky et al, 2006; Mitchell and Staeheli, 2006; Pavao et al., 2007). Affordable housing, as an element of social policy, is characterized as a problem in need of innovative and multi-faceted responses (Carter and Polevychok 2004; Bradford 2005). Increasingly, Housing First has been promoted as innovative in just these terms. As has been the case in other instances of fast policy transfer, claims of innovativeness can be contested: key elements of the Housing First approach have been in play in both Canada and the United States for over 20 years (Kraus et al., 2006). Recent interest in this approach appears to have been sparked by a series of well-publicized research results suggesting that Housing First is both more economical and more effective than Continuum of Care (Fa- gan, 2004; Gladwell, 2006; Gray, 2006; Wente, 2006). These research re- sults have been drawn from scholarly evaluations of Pathways to Hous- ing, a New York City?based initiative that incorporated Housing First 15 with a harm reduction approach and aggressive efforts by workers to connect individuals with the services they need to stay housed, regard- less of intensity or time frame (Gulcur et al., 2003; Tserberis et al., 2004). Evaluations of that program suggest that the cost of certain elements of this approach may be lower than for a similar population of individuals using a Continuum of Care model. A key champion for Housing First who has frequently drawn upon the Pathways to Housing evaluation is Philip Mangano who in 2002 was appointed as the full-time executive director of the Interagency Council on Homelessness (ICH) by President Bush. Since his appointment, the Council has emphasized a coordinated, outcomes-based approach and promoted the development and implementation of community, state, and national ?10-year plans? to end homelessness (Burt et al., 2004). While these plans are not synonymous with Housing First, there are im- portant connections between the two: one, that efforts should first and foremost be directed at moving chronically homeless people off the street and into some sort of shelter, and two, that the focus should be on encouraging coordination among agencies and jurisdictions to prevent individuals from falling through the cracks and not being able to access the services they need to stay housed (Burt et al., 2004). Canadian debates about the relative merits of supported versus supportive housing have been somewhat similar. Historically, Canadian approaches to dealing with chronically homeless individuals involved implicit and explicit assumptions about their readiness for certain ar- rangements, typically within the context of supportive housing. Simon (2006), summarizing the situation for persons with mental illness in the 1950s and 1960s, described the dominant approach as custodial: ?client autonomy was limited, there was an emphasis on rules... and residents had little or no decision-making input? (165). However, he notes that changes began in the 1970s, often associated with self-help and tenant empowerment movements. Currently, ?best-practice guidelines stipulate that residential facilities should be small and homelike, offer more priva- cy, have more of a rehabilitation focus and be run by nonprofit societies? (167). A growing interest in supported, as opposed to supportive, hous- ing reflects ?increased emphasis on the principles of recovery and em- powerment and in particular the importance of client choice? (167). Canadian municipal government and local business interest in Housing First as an alternative to managing homelessness through Con- tinuum of Care approaches is however, a recent development. In 2006, the City of Toronto became an early champion (Gray, 2006), as the result of an emergency response to dismantling a ?Tent City? for homeless adults in 2004 (City of Toronto, 2007). Other Canadian cities with persis- tent and growing homeless populations, including Calgary, Edmonton, and Victoria, soon followed (Calgary Herald, October 1, 2006, A10; Cy- derman, 2006; Hume, 2006; Lavoie, 2006; O?Leary, 2006; Yedlin, 2007). Simultaneously, local activists began to voice reservations about the motives behind such shifts (Crowe, 2007; McQuaig, 2007). Questions have been raised about the manner in which emergency shelter funds in the city are being redirected: ?The planning and funding of homeless services are now focused on removing the visible homeless from the streets while at the same time reducing shelter beds, limiting emergency services for people who are homeless such as during extreme hot or cold weather, and seriously underfunding homeless services such as day shel- ters and meal programs? (Crowe, 2007, 1). An irony of these developments is that Canada has had a federal in- itiative on homelessness since 1999 that, among other elements, has promoted the development of community plans to better coordinate the efforts of local agencies involved in trying to address homelessness. The National Homelessness Initiative was the federal government?s response in 1999 to highly publicized efforts by the mayors of major Canadian cities to highlight homelessness as a ?national emergency.? Program funding established under its Supporting Communities Partnership In- itiative required a coordinated plan involving all of the key players (R. Smith 2004; personal interviews with senior federal government bureau- crats). In each case, either the municipal government or a designated community ?entity? took responsibility for bringing relevant players together to establish and oversee the implementation of a community plan. As in the United States, justification hinged on assumptions that such plans would aid in promoting the efficient use of community ser- vices and help eliminate duplication. In Canada, though, the motivations for the program were not only about getting the chronically homeless ?off the streets,? although there 17 8 certainly was a strong interest in doing just that. The National Home- lessness Initiative coincided with interest in reversing some of the nega- tive impacts of the ?rollback? fiscal policies of the early 1990s. A sharp rise in the number of visibly homeless people, particularly in Toronto, coupled with effective lobbying by Canada?s big city mayors, dramatical- ly raised the profile of this problem and placed a spotlight on the federal government (R. Smith, 2004; personal interviews with senior federal government bureaucrat and politicians). While this interest did not ex- tend as far as reversing federal withdrawal from building new social housing, there was a concerted effort to provide some extra resources to municipalities to deal with the problem. Minister Claudette Bradshaw, a former social worker and community activist, was charged with the homelessness file. One senior bureaucrat described the deliberations leading up to the establishment of the initiative as follows: The government found itself in a horrible quandary. All the affordable housing programs had been cut or deleted. They were trying to come up with new programs, but politically this would be impossible for the next five years... they wanted to open the door a crack...When Minister Clau- dette Bradshaw showed the video after her cross-country tour, there were tears going down faces. Cabinet Ministers did not want to be known as meanies... She gave government instant credibility... (personal interview). Recent Canadian interest in Housing First is likely an outcome of the persistence of homelessness as a visible and seemingly intractable social problem, despite the National Homelessness Initiative and the ef- forts it stimulated among many governmental and non-governmental actors and institutions. It also may be connected to two recent trends in downtown urban redevelopment. On the one hand, pressure is growing for municipal governments to attract economically productive activities in order to help pay for services previously provided by senior levels of government. On the other, disciplinary practices that attempt to reduce the visibility and the assumed negative impacts of less attractive popula- Since its beginnings in 1999, the program has been renewed three times, with some refinements in the objectives but very little change in the model?s fundamental characteristics of coordinated community planning and services and supports to help stabilize the lives of individ- uals who have been homeless. 19 tions and activities, are also on the rise. Also, there is evidence to suggest that residences such as group homes, emergency shelters and transition houses are attracting ever-greater legal oversight through municipal zon- ing and citizen opposition (Takahashi, 1998; Leamon, 2003; Feldman, 2004; Ranasinghe and Valverde, 2006). Residents? fears of adverse im- pacts on property values and neighbourhood safety have resulted in ?not-in-my-backyard? reactions so that finding a suitable location for group homes has become a key challenge and cost (HomeComing Com- munity Choice Coalition, 2005). When such impacts are coupled with downtown redevelopment pressures that reduce the possibility of group home and shelter placement in traditional locations, the ?right to be? for homeless people is stretched to include not only public and quasi-public spaces but also quasi-private ones. Meanwhile, in Canada and the United States, there is a growing consensus among housing advocates and researchers that a greater in- vestment in long-term affordable housing would address the situations of most individuals who face homelessness (Hulchanski, 2002; Shapcott, 2006; Joint Center for Housing Studies, 2007; Shapcott, 2007). In neither country, though, has the creation of additional affordable housing units been a priority, although it is frequently a recommendation of those who produce homelessness plans (National Low Income Housing Coalition, 2007). The lack of attention to building more affordable housing has led some homeless advocates to suggest that Housing First may not be as attractive in practice as it is on paper. They question the utility of concen- trating scarce resources on the hard-to-serve population, rather than ad- dressing the adverse impacts of too little affordable housing for a much broader marginalized population. A question worthy of further research is whether Housing First is attractive because of implicit assumptions that the result would be homeless people housed in ?normal? accommodations, with visible dif- ferences minimized, including the specialized facilities that cause such consternation to property owners. Mitchell and Staeheli?s (2006) argu- ments about a changing property regime with regard to public spaces is potentially relevant to the quasi-private spaces of group homes, transi- tion housing, and emergency shelters. Such institutions sit uneasily in landscapes that are increasingly being shaped by neoliberal logics. 20 1 Housing First and Continuum of Care through a Gender and Race-Sensitive Lens A Lack of Gender and Race-Sensitive Research These debates are of particular interest, because Housing First is pro- moted as being superior to Continuum of Care models, including those for chronically homeless women who are particularly likely to have ex- perienced abuse as children and adults. In Canada, disproportionate numbers of these women self-identify as Aboriginal ? such women bear the traces not only of patriarchal structures, but also of intergenerational colonial trauma (Brownridge, 2003; Peters, 2004). Empirical studies of sex and race differences are, however, almost entirely lacking. According to Kyle and Dunn (2008), one ?shortcoming of the existing research is its inability to speak to the diversity of individual factors that affect housing needs...?; they recommend that housing research ?reflect the diversity of persons with mental health problems including those with different di- agnoses and levels of illness severity, seniors, new immigrants, indigen- ous populations, and rural residents? (12). One insightful exception to the lack of a gender-inclusive approach is a 2005 study by Rich and Clark (2005). Their longitudinal research highlighted the different reactions of women and men to congregate and independent living and raised further questions about whether, for traumatized women in particular, the former may be an especially mea- ningful form of care. Rich and Clark (2005) investigated the effectiveness of two types of homelessness service interventions (similar to a Housing First versus Continuum of Care distinction) and compared single women and men with severe mental illness in this regard. The findings showed that men using the Housing First type model had longer periods of sta- ble housing, but that women exhibited more complex and contradictory outcomes (Rich and Clark 2005). Comparable studies among Aboriginal women have not been located, although Richmond (2007) suggested that Aboriginal men and women respond differently with regard to social interventions vis-? -vis health outcomes. Although critics of the Continuum of Care model can cite evidence that the model is sometimes interpreted in a rigid, inflexible manner, the 22 reality of supportive housing is sometimes far more nuanced. In Canada, several agencies have combined the best of Housing First and Conti- nuum of Care perspectives, including the combination of ?no eviction? and harm reduction within congregate settings (Kraus et al., 2006). In these cases, people have been able to connect to one another as part of a caring community, to reap the benefits that on-site staff support some- times provide, while still benefiting from enlightened policies shaped by a ?harm reduction? philosophy. Rather than focusing solely on the im- portance of access to good physical shelter and individualized support services with the aim of establishing independent living, the significance of creating home and community spaces as spaces of healing and nur- ture have been emphasized (Gurstein and Small, 2005). Chronically Homeless Women Inhabiting the City Kyle and Dunn (2008) observe that ?based on previous theoretical and empirical research, it is likely that merely having shelter is a neces- sary but insufficient condition for maintaining stable housing; in order for people with SPMI [severe and persi stent mental illness] or anyone else for that matter to be successfully housed, they must also have some experience of home? (1). Walker (2008) also notes that Canadian social policy in the 1990s has decoupled earlier efforts to join social housing and Aboriginal self-determination efforts, with negative results. In contrast to the current presentations of Housing First as a new and better way to end homelessness, an appreciation of building ?a sense of home? acknowledges that collective, appropriately situated arrange- ments (and, in the case of Aboriginal communities, autonomous over- sight) may provide therapeutic benefits over and above the ?bricks and mortar? required to construct housing. It provides an appreciation of what ?inhabiting the city? might mean for chronically homeless women. Gurstein and Small (2005), reporting on their interviews with residents of the Portland Homes Society, an innovative supportive housing organ- ization in Vancouver, powerfully articulate this perspective: [Tenants] do not perceive the acquisition of a home as part of a rehabilit a- tion process. Individuals who are homeless or who have had difficulty in- tegrating into the housing units of conventional housing agencies view creating a home as part of a wider process of personal self-healing. Heal- 23 4 ing is a process that is self-authored. It is not an intervention that is done to tenants or clients by professionals. Many do not feel confident about their sense of membership in, or connectedness to, the wider community nor do they feel confident in their full personhood... Constructing a sense of home is a social, meaningful human action. It is not simply a physical structure (2005, 732). From this viewpoint, housing is seen as a necessary social and ma- terial context within which healing can take place that allows for further engagement with others. Feelings of ?connectedness to the wider com- munity? are an outcome of positive social relations, but they also depend on a sensitivity to certain types of interactions between bodies, physical structures, and territorial surroundings. Such interactions allow a sense of being ?at home? to become established as a result of positive impacts from a multiplicity of social relations, ranging from co-residents to sup- portive staff to welcoming surroundings. From this vantage point, hous- ing is much more than a utilitarian setting: it is the starting point for es- tablishing a home base as a foundation for claiming ?rights to the city,? including efforts centred not necessarily on Feminist Perspectives employability, but on the (sometimes limited) capacities of those who have lived through trauma and abuse (Tomas and Dittmar, 1995; Bridgman 2002). The Declaration?s focus was on multi-scalar collaboration, empha- sizing particularly the significance of learning from the most vulnerable in building welcoming and inclusive cities. The declaration?s creators asserted that ?[t] he solutions introduced by women to increase safety Since the late 1960s, some feminist activists and academics have been building the case for a gender- and race-sensitive analysis in urban studies and design. Ideas have spread and networks have emerged to share strategies on how to build women-friendly and inclusive cities. In 2002, the First International Seminar on Women?s Safety became a space of sharing and elaborating diverse perspectives; it also was the moment that a formal organization, Women and Cities International/Femmes et Villes, was born. Women and men from five continents, 27 countries, and 55 cities and municipalities contributed to the group?s founding docu- ment, the Montreal Declaration (Women and Cities International, 2002). 25 and security [would] make cities and municipalities safer for all.? In oth- er words, they asserted that starting from the perspectives of those who are most affected by exclusionary cities would most likely result in broad based benefits and ?rights to the city? for urban residents more generally (Michaud, 2004; Shaw and Andrew, 2005; Whitzman 2006). This approach aligns with Fraser?s interpretation of what liberal so- cietal values should encompass. For chronically homeless women, Fras- er?s arguments would mean acknowledging the legitimacy of their claims and the simultaneous need for secure affordable housing (redi- stribution), living circumstances that signal understanding and promote healing (recognition), and efforts that invite women to articulate their needs and desires in political terms (representation). Drawing on Purcell (2008), the Women?s Roundtable might thus be interpreted as a social movement that began from the needs of chronically homeless women, not in order to achieve results for these women instead of others, but rather to bring into political discourse policies that would facilitate homeless women?s ability to inhabit the city. According to Graefe?s (2006) schema, these perspectives illustrate countervailing arguments to neoliberalization as a process promoting new forms of ?social rule.? Graefe?s (2006) framework helps unpack the diverse rationales that have contributed to Canadian housing policy dis- cussions on what to do about chronic homelessness. These complex and contradictory interpretations and prescriptions fully support Larner?s (2003) characterization of neoliberalism as not only ?operating at mul- tiple scales? but also in need of sensitivity to its ?different variants... [i n- cluding the] hybrid nature of Countervailing, Flanking, and Rollout Elements of Housing First and Continuum of Care contemporary policies and pro- grammes [and]... the multiple and contradictory aspects of neoliberal spaces, techniques, and subjects? (509). There are certainly ?countervailing? elements in initiatives such as New York City?s Pathways to Housing, where the goal is for individuals to live with dignity, with access to the services and supports they need, choices in housing, and a timeframe more geared to healing than to sys- 26 7 tem efficiencies. This initiative challenges the assumption that individu- als who use substances cannot live independently or that using sub- stances constitutes a reason for state indifference or condemnation (Tser- beris et al., 2004). However, when individuation and employability begin to supersede other goals, elements of ?rollout? and ?flanking? come to the fore, while countervailing tendencies that emphasize healing and the construction of ?home? are overshadowed. Continuum of Care can be assessed in a similar manner. When the emphasis is on dignity and real choices are available, supportive housing can be a context where countervailing values are promoted. More typi- cally though, a notion of rehabilitation that is more directive of how in- dividuals should behave and that establishes a hierarchy of suitable be- haviours and rewards, simply ?flanks? neoliberalism as a means of ?organizing? individuals who likely cannot manage on their own. De- clining resources often result in approaches that shift from flanking to ?roll-out,? where messages shift to those of ?cost recovery? and ?pro- moting independence.? The motivations of Canada?s federal homelessness strategy contain some of these same tensions. Developed in response to a ?national cri- sis,? officials rejected the argument that building additional permanently affordable housing would address the problem more effectively than other approaches. Instead, they shaped policy based on assumptions about the need for increased coordination and efficiencies tied to local circumstances, as well as greater knowledge about the sources of the problem. The coordinated community approach was countervailing in- sofar as it acknowledged that each community faced somewhat different challenges and that local knowledge was valuable in helping address those challenges. Over time though, the flanking aspect of the Initiative has become its most prominent element. Without senior government re- sources, communities could not manage the growing problems of ex- treme poverty, including homelessness, within their jurisdictions. The new interest in Housing First may be the outcome of a disillusionment with the Initiative and its ?management of homelessness? approach and a wish to find a ?quick fix? in a manner that aligns with other ?roll-out? kinds of arguments, such as those that emerged around workfare in the United States and Canada in the late 1990s (Peck and Tickell, 2002). 28 Conclusions Toronto area activists have hinted that the expectation on the part of policy makers seems to be that Housing First would be a cheaper and more direct route to moving homeless people off the streets and into self- sufficient life styles than is currently the case. A recent editorial in the Ottawa Citizen that government efforts should focus on the ?truly home- less? ? those who have lived in emergency shelters for more than 60 days ? and focus on providing coordinated housing and supports to help them stabilize their lives illustrates the concerns raised by activists. This editorial suggests that a concentration of funds on the ?truly homeless? would be a good investment, even if it meant drawing some funds from maintaining ?general-purpose social housing? ? despite numerous hor- ror stories about the extent to which that housing requires massive up- grades to address extreme mould and insect infestation problems (Otta- wa Citizen, November 19, 2007, C4). Such recommendations suggest that the motivations for promoting Housing First have more to do with re- moving visually disturbing images from places ripe for redevelopment than with concerns about the provision of decent housing or the auton- omy rights and ?rights to the city? of homeless persons. Growing interest in Housing First, and an increasing skepticism about Continuum of Care and its focus on specialized, congregate facilities, threatens marginalized peoples not only in terms of their rights to public space but also in terms of their visible presence in the city. An outcome of potential concern is that specialized, congregate spaces are re-framed as part of the problem of homelessness rather than as part of the solu- tion. Housing First is an attractive approach in theory and one that in some variants, exhibits countervailing tendencies in relation to neolibe- rally inspired social policy developments. But just as deinstitutionaliza- tion was negatively reshaped with disastrous consequences, there is a well-founded fear that Housing First could result in warehousing margi- nalized individuals in units that present a fa?ade of normalcy , but that exacerbate the isolation and exclusion of already vulnerable individuals. The lack of concerted efforts to devote substantial resources to providing permanently affordable and secure housing, coupled with urban redeve- 29 30 lopment pressures that reduce such opportunities, mean that more low- income people will struggle to maintain independence while confronting the inevitable instability of not having a home. A focus on home spaces and gendered and race-sensitive rights to the city represent alternatives to neoliberal urbanism. These perspectives would explicitly acknowledge that among the most marginalized groups, women and men may have different reasons for becoming homeless and may also therefore, react differently to the social care on offer. This focus would also highlight inevitable connections between home and neighbourhood spaces and raise questions about what physi- cal structures and community surroundings are required for inclusion efforts to be meaningful. Addressing the question of how these options affect the individuals for whom they are prescribed is one route to fur- ther investigation of the complex relations and issues raised above. Acknowledgements Fran Klodawsky is in the Department of Geography and Environmental Stu- dies, Carleton University. Bibliography Thanks to the many individuals who commented on earlier drafts: Janet Siltanen, Sue Garvey, Wendy Muckle, Alette Willis, Val Hinsperger, El- vin Wyly, Carolyn Whitzman, Nick Blomley, Rianne Mahon, Dominique Masson, and three anonymous reviewers. Aubry, T., Klodawsky, F., Hay, E., & Birnie, S. (2003). Panel study on per- sons who are homeless in Ottawa: Phase 1 results. Ottawa: University of Ottawa, Centre for Research on Community Services. Bondi, L., & Laurie, N. (2005). Working the spaces of neoliberalism: Activ- ism, professionalization, and incorporation, Introduction. Antipode, 37(3), 394-401. Bradford, N. (2005). Place-based public policy: Towards a new urban and commu- nity agenda for Canada (Report F/51). Ottawa: Canadian Policy Re- search Networks. Bridgman, R. (2002). Housing chronically homeless women: ?Inside? a safe haven. Housing Policy Debate, 13(1), 51-81. 31 Brownridge, D. A. (2003). Male partner violence against Aboriginal wom- en in Canada: An empirical analysis. Journal of Interpersonal Violence, 18(1), 65-83. Burt, M., Hedderson, J., Zweig, J., Ortiz, M. J., Aron-Turnham, L., & John- son, S. (2004). Strategies for reducing chronic street homelessness. Washing- ton: Department of Housing and Urban Development, Office of Poli- cy Development and Research. Callaghan, M., Farha, L., & Porter, B. (2002). Women and housing in Canada: Barriers to equality. Toronto: Centre for Equality Rights in Accommo- dation. Carter, T., & Polevychok, C. (2004). Housing is good social policy (Report F/50). Ottawa: Canadian Policy Research Networks, Family Net- work. Centre for Addiction and Mental Health (2003). CAMH and harm reduc- tion: A background paper on its meaning and application for substance use issues. Retrieved January 30, 2008, from http//www.doctordeluca.com/Library/AbstincenceHR/ CAMH&HR03.htm. City of Toronto (2007, Winter). Streets to Homes update. Toronto: Shelter, Support & Housing Administration. Cyderman, K. (2006, September 27). ?Homeless? czar offers inspiration: Calgary urged to develop 10-year plan. Calgary Herald, p. B1. Crowe, C. (2007, February). Dismantling downtown. Cathy Crowe?s New- sletter, 31. Fagan, K. (2004, June 14). Success in the Big Apple: New York City finds path for mentally ill: Housing homeless before treatment bucks con- ventional wisdom. San Francisco Chronicle, p. A1. Falvo, N. (2007, June 1-3). Addressing Canada?s lack of affordable hous- ing. In Interdisciplinary Approaches to Economic Issues. PEF Panel at the Annual Meeting of the Canadian Economics Association, Halifax. Feldman, L. (2004). Citizens without shelter: Homelessness, democracy, and polit- ical exclusion. Ithaca: Cornell University Press. Flyvbjerg, B. (2001). Making social science matter: Why social inquiry fails and how it can succeed again. Cambridge: Cambridge University Press. Fraser, N. (1997). Justice interruptus: Critical reflections on the "postso cialist" condition. New York: Routledge. Fraser, N. (2003). Distorted beyond all recognition: A rejoinder to Axel Honneth. In N. Fraser & A. Honneth (Eds.), Redistribution or recogni- tion?: A political -philosophical exchange (pp. 198-236). London: Verso. Gladwell, M. (2006, February 13). Million dollar Murray. New Yorker Ma g- azine. Graefe, P. (2006). The social economy and the American model. Global So- cial Policy, 6(2), 197-219. Gray, J. (2006, September 23). A home of his own. Globe and Mail, p. M1. Gulcur, L., Stefancic, A., Shin, M., Tsemberis, S., & Fischer, S. N. (2003). Housing, hospitalization, and cost outcomes for homeless individu- als with psychiatric disabilities participating in Continuum of Care and Housing First programmes. Journal of Community and Applied So- cial Psychology, 13(2), 171-186. HomeComing Community Choice Coalition (2005). Yes, in my bac kyard: A guide for Ontario?s supportive housing providers. Toronto: Centre for Addiction and Mental Health. Gurstein, P., & Small, D. (2005). From housing to home: Reflexive man- agement for those deemed hard to house. Housing Studies, 20(5), 717- 735. Homeless Community Capacity Building Steering Committee (2008, Jan- uary). Meeting Minutes, Ottawa: Author. Hulchanski, J. D. (2002). Housing policy for tomorrow?s cities (Discussion Paper F/27). Ottawa: Canadian Policy Research Network, Family Network. Hulchanski, J. D. (2006). What factors shape Canadian housing policy?. In R. Young & C. Leuprecht (Eds.), Municipal-federal-provincial relations in Canada (pp. 221-250). Montreal: McGill-Queen?s University Press. Hume, M. (2006, December 7). A home is where the hope is. Globe and Mail, p. S1. Hutson, S., & Clapham, D. (1998). Homelessness: Public policies and private troubles. London: Cassell. Joint Center for Housing Studies (2007). The state of the nation?s housing. Harvard: Author. Klodawsky, F. (2007). ?Choosing? participatory research: Partnerships in space? time. Environment and Planning A, 39(12), 2845-2860. Klodawsky, F., Aubry, T., & Farrell, S. (2006). Care and the lives of home- less youth in neo-liberal times in Canada. Gender Place and Culture, 13(4), 419-436. Klodawsky, F., Aubry, T., Nemiroff, R., Bonetta, C., & Willis, A. (2007). What happens over time: Researching homelessness longitudinally. Canadian Journal of Urban Research, 16(1), 93-111. Kyle, T., & Dunn, J. (2008). Effects of housing circumstances on health, quality of life and healthcare use for people with severe mental ill- ness: A review. Health and Social Care in the Community, 16(1), 1-15. Kraus, D., Goldberg, M., & Serge, L. (2006). Homelessness, housing and harm reduction: Stable housing for homeless people with substance abuse issues. Ot- tawa: Canada Mortgage and Housing Corporation. Lavoie, J. (2006, May 15). Crisis caught Victoria politicians off guard, de- spite years of warning. Times Colonist, p. A1. Larner, W. (2003). Neoliberalism?. Society and Space, 21(5), 509-512. Leamon, E. (2003). The zoning of group homes for the disabled... zeroing in on a reasonable accommodation. Connecticut Public Interest Law Journal, Paper 4. Layton, J. (2000). Homelessness: The making and unmaking of a crisis. Toronto: Penguin. McQuaig, L. (2007, June 1). We allowed rich to win class war. Toronto Star, p. AA8. McKeen, W. (2006). Diminishing the concept of social policy: The shifting conceptual ground of social policy debate in Canada. Critical Social Policy, 26(4), 865-887. Michaud, A. (2004). A city tailored to women: The role of municipal governments in achieving gender equality. Ottawa: Federation of Canadian Munici- palities. Available from City of Montreal Web Site: http://www.ville.montreal.qc.ca/femmesetville. Mitchell, D., & Staeheli, L. (2006). Clean and safe? Property redevelop- ment, public space, and homelessness in downtown San Diego. In S. Low & N. Smith (Eds.), The Politics of Public Space ( pp. 143-176). New York: Routledge. Mitchell, D. (2003). The right to the city: Social justice and the fight for public space. New York: Guilford Press. Moore, E., & Skaburskis, A. (2004). Canada?s increasing housing afforda- bility burdens. Housing Studies, 19(3), 395-413. National Low Income Housing Coalition (2007). Plans to end homeless- ness. Retrieved June 22, 2007, from http://www.nlihc.org/detail/article.cfm? article_id=2796&id=19. Novac, S., Brown, J., & Gallant, G. (1999). Women on the rough edge: A decade of change for long-term homeless women. Ottawa: Canadian Mortgage and Housing Corporation. O?Leary, B. (2006, November 16). A new strategy to end homelessness. Calgary Herald, p. A20. Peck, J. (2002). Political economies of scale: Fast policy, interscalar rela- tions, and neoliberal welfare. Economic Geography, 78(3), 331-360. Pavao, J., Alvarez, J., Baumrind, N., Induni, M., & Kimerling, R. (2007). Intimate partner violence and housing instability. American Journal of Preventive Medicine, 32(2), 143-146. Peck, J. (2004). Geography and public policy: Constructions of neoliberal- ism. Progress in Human Geography, 28(3), 392-405. Peck, J. (2005). Struggling with the creative class. International Journal of Urban and Regional Research, 29(4), 392-405. Peck J., & Tickell, A. (2002). Neoliberalizing space. Antipode, 34(3), 380-404. Peters, E. (2004). Suzanne MacKenzie memorial lecture conceptually un- clad: Feminist geography and Aboriginal peoples. Canadian Geograph- er, 48(3), 251-265. Purcell, M. (2008). Recapturing democracy: Neoliberalization and the struggle for alternative urban futures. New York: Routledge. Rich, A. R., & Clark, C. (2005). Gender differences in response to home- lessness services. Evaluation and Program Planning, 28(1), 69-81. Ranasinghe, P., & Valverde, M. (2006). Governing homelessness through land- use: A sociolegal study of the Toronto Shelter Zoning By-law. Canadian Journal of Sociology, 31(3), 325-349. Richmond, C. (2007). Social support and thriving health: A new approach to understanding the health of indigenous Canadians. American Jour- nal of Public Health, 97(10), 1829-1833. Shapcott, M. (2006). Framework for the blueprint to end homelessness in Toron- to. Toronto: The Wellesley Institute. Shapcott, M. (2007, January). Fourteen cents a day won?t build many homes. Ottawa: Canadian Centre for Policy Alternatives. Shaw, M., & Andrew, C. (2005). Engendering crime prevention: Interna- tional development and the Canadian experience. Canadian Journal of Criminology and Criminal Justice, 47(2), 293-316. Shinn, M. (2007). International homelessness: Policy, socio-Cultural, and individual perspectives. Journal of Social Issues, 63(3), 657-677. Simon, D. (2006). Community mental health in Canada. Vancouver: UBC Press. Shinn, M., Weitzman, B. C., Stojanovic, D., Knickman, J. R., Jimenez, L., Duchon, L., James, S., & Krantz, D. H. (1998). Predictors of home- lessness among families in New York City: From shelter request to housing stability. American Journal of Public Health, 88(11), 1651-1657. Smith, R. (2004). Lessons from the National Homelessness Initiative. In Policy development and implementation in complex files (pp. 1? 32). Ottawa: Canada School of Public Service. Takahashi, L. M. (1998). Homelessness, AIDS, and stigmatization: The NIMBY syndrome in the United States at the end of the twentieth century. New York: Clarendon Press. Tsemberis, S., Gulcur, L., & Nakae, M. (2004). Housing first, consumer choice and harm reduction for homeless individuals with a dual di- agnosis. American Journal of Public Health, 94(4), 651-656. Tomas, A., & Dittmar, H. (1995). The experience of homeless women: An exploration of housing histories and the meaning of home. Housing Studies, 10(4), 493-516. Wente, M. (2006, April 20). Real homes for the homeless. Globe and Mail, p. A21. Walker, R. (2008). Aboriginal self-determination and social housing in urban Canada: A story of convergence and divergence. Urban Studies, 45(1), 185-205. Whitzman, C. (2006). Women and community safety: A retrospect and prospect, Women and Environments, 70-71, 24-27. Women and Cities International (2002, May 19-21). Montreal declaration on women?s safety. Retrieved April 2, 2006, from http://www.femmesetvilles.org/english/ setsen/setdeclarationen.htm Young, R., & Leuprecht, C. (2004). Municipal-federal-provincial relations in Canada. Montreal: McGill-Queen?s University Press. Yedlin, D. (2007, January 12). Plight of homeless draws might of Calgary CEOs. Globe and Mail, p. B2. Chapter 4.5 Pan- Territorial Report : A Study of Women?s Homelessness North of 60 ? JUDIE B OPP , RIAN VAN BRUGGEN, SHYLAH ELLI OTT , LYDA F ULLER, MIRA HACHE, CHARL OTTE HRENCHUK, MARY BETH LEVAN, AND GILLAN MCNAUGHT O N Research about women?s homelessness in the North is critical, since women have been identified as among the fastest-growing groups in the homeless and at-risk population. In the North, all women can be consid- ered at risk of homelessness, because a small change in their circum- stances can jeopardize the fragile structure of their lives that allows them to meet their basic needs. Although everyone living in the Canada?s three northern territories recognizes that housing is a ?big problem,? few understand the complex constellation of factors, many of which go well beyond the shortage of housing stock, that conspire to keep thousands of women and their chil- dren in a condition of absolute or hidden homelessness. Those who do not live in the North have even less awareness about the despair and day-to-day suffering of their fellow Canadians. The authors of this report are convinced that the story of women and homelessness in the North must be told in such a way that it will inspire political and social will for action. Research is one way to give voice to women whose experience has so far remained on the margins of society. In 2005, a consortium consisting of Kaushee?s Place and the Yukon Status of Women Council in Whitehorse, Yukon; Qulliit Nunavut Status of Women Council and Qimaavik Women?s Shelter in Iqaluit, Nunavut; and the Yellowknife Women?s Society and YWCA Yellowknife in Yel- lowknife, NWT received funding from the National Research Program of the National Homelessness Initiative to carry out a comprehensive study of women?s homelessness in the North. Judie Bopp, Ph.D. of the Four Worlds Centre for Development Learning was asked by to serve as the study?s principal researcher. Research Purpose and Method A Study of Women?s Homelessness North of 60 was designed to address the following objectives: 1. to inform and improve the services provided by the partners of this Study related to the incidence and impact of homelessness among Northern women; 2. to influence the quality of service provided by other organizations and agencies across the North serving homeless women and those at risk of becoming homeless; 3. to inform public policy and territorial and regional program initia- tives such that they are more effective at reducing homelessness and the negative impacts of homelessness among Northern women; 4. to stimulate community action aimed at reducing homelessness and the negative impacts of homelessness among Northern women. A naturalistic research method that drew on feminist and grounded theory was chosen as most appropriate for this study. By adopting a grounded theory methodology that privileges the voices of Northern homeless women, the research process remained iterative, participatory, and action-oriented. Data was generated through focus group discussions and inter- views with homeless women and those at risk of becoming homeless, as well as the service providers in both the government and voluntary sec- tors that work most closely with them. The transcripts of these dialogues were coded according to themes that were generated from the data itself. Once all the data was coded, the material related to each theme was compiled. Theme anthologies were prepared from each of these theme compilations. These anthologies wove together the contributions from all the informants in a way that would present a coherent picture, while at the same time protecting the confidentiality of the participating women and service providers. The Nunavut Territorial Report is unique in that it also includes direct quotations from homeless women and service pro- viders. Brief excerpts from these theme anthologies appear at the begin- ning of each of the sections that describe the determinants and impacts of homelessness for northern women, as well as the policy and service en- vironment that either contribute to women?s homelessness or strive to mitigate its impacts. While this data gathering and analysis work was being done, a lit- erature review was also conducted, which provided a useful point of comparison for the Study?s findings and also added perspectives to en- rich its contribution to the field. Characteristics of the North that Contribute to Women?s Homelessness Canada?s North has special characteristics that contribute to high rates of homelessness in general, as well as among women in particular. Al- though each territory has its own unique circumstances, they all share the following: ? remote geography; ? underdeveloped infrastructure; ? a harsh climate; ? a small population base; ? a high cost of living and limited employment opportunities; ? the lack of accessible and affordable transportation systems; ? inadequate access to appropriate social services; ? the high cost of labour and materials needed to increase housing stock; ? high rates of social issues such as addictions, domestic violence, and intergenerational dependency on income support. The Demographic Characteristics of Homeless Women in the North The conduct of a homelessness count in Northern communities was well beyond the scope and means of this study. The following demographic and incidence data was compiled from the anecdotal data collected, as well as statistical information about user rates for certain types of ser- vices geared toward homeless women. ? All told, 205 women participated in interviews or focus group ses- sions (66 in the NWT, 66 in the Yukon and 73 in Nunavut). ? Of these women, 53 percent were Inuit, 30 percent First Nations, 10 percent Caucasian, 5 percent Inuvialuit, 1 percent M?tis, and 1 per- cent immigrant. ? In the Yukon and the Northwest Territories, at least one-third of the homeless women had completed high school and at least half of those have some college or university education. In Nunavut, 87 per- cent had not finished high school, and many of these did not have functional literacy skills. ? At least 80 percent of the women in all three Territories have chil- dren; about half of these children these women are in someone else?s care. ? Of the women interviewed in the Northwest Territories, 25 percent are working, but still cannot afford housing. Most of the women in Nunavut are on Income Support of some kind, since employment is very scarce. In the Yukon almost 60 percent are on income support. The women were all homeless at the time of this study. Most were cycling through the different phases of homelessness. For example, they might be currently living in a shelter, but the following week they could be living rough on the streets, then they might spend a few days with relatives or friends, end up trading sex for shelter, then because of abuse, end up back in the shelter. Variations of this pattern were the norm for these women. The study was unable to get accurate incidence figures, but infor- mants from the Northwest Territories estimate that there are could be 500 homeless women in Yellowknife alone. The figure in Iqaluit is esti- mated at 300. These estimates point to a homelessness rate across of the North of well over 1,000 women. When their dependent children are fac- tored in, the figure raises to well over 2,000. The Determinants of Homelessness for Northern Women The following thirteen themes emerged from the interviews and focus groups conducted for this study. Every woman is vulnerable Someone stole my rent money and I ended up living in my truck camper. The truck broke down and I couldn?t get to work. It was winter and too cold to be living in the camper. I went to the Shelter, but they only let you stay there a month. I was lucky and found a house sit until the weather warmed up. I didn?t mean for this to happen. The stories of homeless women across the North describe the vulnerabil- ity and insecurity of women, of how easy it can be to slide into home- lessness. The unexpected looms large and can be the final straw for women in precarious situations. Sudden illness, job loss, loss or thefts of rent money, immigration, addiction, or injury are unexpected hardships in women?s lives, throwing them off-balance and into homelessness. These events trigger a domino effect, one loss leading to many. Loss of a job can lead to loss of a vehicle, which limits job search or access to town, which leads to the loss of other possessions and any savings, which in turn leads to the loss of a home. Many women work and continued to work while they are homeless, trying to keep it together, but finding it difficult, especially if they have children. Abuse complicates the picture, taking away self-esteem as well as financial support. Women who have immigrated to find a better life in a country with a shining image can find themselves homeless and without resources. They find that Canada is not living up to its reputation or commitments to the United Nations on economic and social rights, as well as the Convention on the Elimination of All Forms of Discrimina- tion against Women. Women who had previously led comfortable lives here and abroad and thought themselves safe in their homes say, ?We are all hanging on the line.? Women?s security depends on their partners? behaviour and circumstances We get called names and other kinds of verbal abuse. For a long time I thought my partner would change. Every time after he beat me, he would cry and say he was sorry. He would promise not to do it again. I believed him because I thought I still loved him. I guess I also believed him be- cause I felt worthless and helpless. I didn?t think I could make it on my own. The security and well-being of women and children is closely linked with the behaviour and circumstances of their intimate partners. This determinant encompasses this complex web of relationships from the point of view of women who experience homelessness. Many homeless women experience physical, sexual, mental, and psychological abuse at the hands of their partners. Sometimes their children are also sexually and physically abused. Often they are trapped in the control patterns associated with that abuse. They describe themselves as slaves for their partners and without any control over financial resources. They are at the mercy of their partners, who can force them out of the home if they so choose. Sometimes partners threaten to separate women from their children as a way to control them. Addictions are an inseparable part of the abuse pattern. When part- ners are addicted to alcohol, other drugs and gambling, they are more likely to be abusive. They also spend resources that should be used to secure shelter to feed their addictions. When housing is damaged as a result of partying or violent behaviour, or when rent payments are not made, women and children lose their homes, even if they are not respon- sible for the problem. When women also suffer from addictions, this problem is even more serious and complex. Forced eviction can lead to homelessness I got evicted on more than one occasion ? three times actually. The first time it was because my first husband passed away and his name was on the lease. They made me leave. Another time my ex-boyfriend was van- dalizing and his name was on the lease, so we got kicked out. The last time I was evicted, I had a house in my community, but I came to Iqaluit for the hospital. While I was away, the house got taken away from me. There?s such a shortage of houses, they thought I had left for good so it was given to someone else. Forced eviction from social housing units was a reality for many of the women interviewed. A primary reason for eviction is that the male lists a unit under his name, exclusive of his female counterpart. If the relation- ship ends, becomes abusive, or if the woman becomes widowed, she is expected to evacuate her home. The vast majority of women who shared stories of eviction have been forced out of their homes because of their partners? actions. Tenant damage is also another reason women become evicted from public or private housing. Relocation to another community can lead to homelessness Housing in the communities is bad. I had to wait eight months to get a place back home. I ended up living with my boyfriend?s sister. It?s really hard to get a place here if you have any arrears back home. They find out and won?t give you a place here. If I could share a place it would be eas- ier. When I moved to Yellowknife, it took me six months to get a place be- cause of my arrears. I got really depressed and went to Edmonton where I could couch surf. There was an all-night coffee shop there, and the owner would let me sleep on the couch sometimes. When things were really bad, I would go to the U of A and see if I could crash with people from home. Many women move from their home communities to larger centres be- cause they think they can access better services there, including housing. Others move because they feel forced out by circumstances in their home community, including reprisals for disclosing the abusive behaviour of their partners or other men in the community. Unfortunately, many women often find that the living arrangements they were counting on in the new community were either non-existent or inadequate. Lack of an adequate support system is a factor in homelessness Sometimes family is worse than no help at all. I know that sometimes there?s so much drinking and fighting going on in a family that the only thing you can do is leave town and try to make it totally on your own. But sometimes it?s the family who cuts out the woman, especially if she is drinking. I guess you can understand how things get so that even family won?t help a woman with kids, but it?s still not right. Maybe she?s drink- ing because she has lost friends or family, or maybe she?s just trying to deal with all the pain from a bad relationship. [Women] come into town figuring that they could stay with their sister or their cousin, but that doesn?t always work out. One woman at the shelter, she said her sister wouldn?t even keep her clothes for her. She just gave them away and her mother won?t even look at her. Even if someone does take pity on them, well, they end up really resenting them. They?re always hinting about when are you leaving, and why don?t you buy more groceries. One of the key factors that allow women to survive the critical incidents in their lives that put them at risk of homelessness (such as fleeing an abusive situation, losing employment, serious illness, the death of a partner, illness or disability in a family member) is an adequate support system. When support systems do work, they often consist of an infor- mal network of family and friends, as well as an effective range of volun- tary sector and government services. There are many reasons why these support systems break down. The crises listed above may put more strain on families and friendships than they can handle. Homeless women often flee the communities in which they grew up or lived for extended periods, leaving them cut off from family and friends. Intergenerational dysfunction, often the result of intrusive forces such as residential schools, diminishes the capacity of families and friends to support each other in a healthy way. These same forces can also destroy the health and effectiveness of community institu- tions that should be there to support those members of society who are experiencing hard times. Many institutions of society also operate from ideologies that do not foster compassionate support and have policies and practices that punish women or fail to provide the support that would make the critical difference. Personal wellness and capacity is a determinant of homelessness How am I ever going to get a good night?s sleep when I?m constantly moving from place to place? I can?t think when I?m at work, I?m tired, I?m stressed out, I?m depressed. When you use all your energy going around all day looking for a place to rest, how can you have a good night?s sleep and function all day? With no food, no breakfast, you just can?t! From b e- ing so stressed out, I?m bleeding inside. I?ve had my period for three months now and I?m waiting to see the doctor again today. When I talk to him, I tell him I?m tired because I couldn?t find a place to sleep last night and I?m just depressed all the time. Another determinant of homelessness in Northern women involves wellness and capacity. A woman?s potential for improving her position in life is often inhibited by her health and/or her perception of her own personal abilities. Women often find themselves suffering from physical and emotional exhaustion, including feelings of disempowerment, which trap them in a cycle from which they can find no respite. Being incapable of sheltering/protecting themselves and their children results in feelings of worthlessness, eventually taxing every other area of their lives. They are stripped of all self-esteem, and poor health negatively infringes upon their capacity to better their situations. Many of the women interviewed stated that they have experienced a complete loss of identity, with no remaining sense of a culture and worth that brought such a great sense of pride to their forbears. Women with disabilities may face increased risks of homelessness The Handibus driver once said to me, ?Some people in a wheelchair just don?t know to stay put.?... God forbid that I should want to go to the Canada Games Centre or to the Remembrance Day parade. The attitude is, why doesn?t she just stay home? We are told that we should let able- bodied people do our part in participating in these activities. Excuse me, I work as much as I can, and so that makes me a labourer too. Women with disabilities face many obstacles that put them at risk of be- coming homeless. They have a hard time getting work in a job market that already discriminates against women. Income support payments for which they may be eligible are inadequate in light of the actual cost of living in the North. They face stigmatization and are victimized in count- less ways. If they do manage to find housing, whether private or public, it is often inadequate. Being forced to share accommodations with roommates who take advantage of them, living in housing that has not been adapted to fit their handicaps or not being able to get equipment that would allow them to function more fully, not having access to transportation or health services?these are just some of the obstacles. Women are also most often the caregivers for disabled family mem- bers. Their struggle to meet the needs of their loved ones while trying to keep a roof over their heads is frustrating and exhausting. Perhaps the most common, but still largely misunderstood, disabil- ity in the North is fetal alcohol spectrum disorder. Women born to alco- holic mothers (who may themselves have experienced homelessness) have few places to turn for help and rarely receive the support they need to live healthy and stable lives. Geographic factors play a role in homelessness It?s hard in the summer, but winter is worse. I can sleep outside in the summer, but not in winter. I almost froze my feet off last winter. I can walk around all summer until everyone wakes up. Living in the car is no problem in the summer, but in the winter I have to send my common-law to the men?s shelter, and I stay at a friend?s house. There?s so much more pressure to find somewhere to sleep in winter. Although homelessness is a global issue, Northern women face unique challenges that call for different solutions. In the North, homelessness tends to be invisible; people are not living on the street because the harsh weather prevents them from doing so. On the most frigid days of the year, the climate can reach 60? below zero, forcing penniless women to frequent coffee shops, hoping to make a cup of coffee span the day, or gathering at a friend or family member?s already overcrowded home. The northern climate, combined with lack of available housing, is why homelessness in the North shows itself in the average number of people per dwelling. With so few shelters for women in the territories, women rely on extended family to house them. A further determinant of homelessness is the very geographic area of the North, most of which is not accessible by road. (For example, only nine of the 33 NWT communi- ties are on a road system. Nunavut communities are all ?fly-in.?) Women are not able to return to their home community, or escape it, without ex- tensive financial, emotional, and practical resources. They end up home- less, living in a shelter, on the street, or with family or friends who do not want them, because they cannot easily travel. Community institutions and structures can contribute to homelessness I am from a small community, and I don?t want to live there any more. In my community there are two different tribes, and one looks down on the other. I am from the lower one, so no one will help me or my sister. The town supported my ex-husband ?cause they said he was a good person. He works for housing, so I will never get a house there. The Housing Au- thority there said I made false statements. But I know it?s because my ex works there, and his family is influential too, so I can?t get a place there. The policies and practices of community institutions and other structures can contribute to homelessness for women. Many of the women inter- viewed for this study reported that they were never able to access hous- ing in their home communities, as the housing always went to the friends and family of housing authority staff. If these women left an abu- sive situation, the man retained control of the home, and the women and children were the ones without shelter. When a woman?s husband dies, she can also end up homeless, since the unit they were living in might be assigned to someone else. The women who were interviewed felt that community leadership is reluctant to address their problems, and that it is much easier for them to simply dismiss homeless women as ?bad.? In addition, they found the housing, income support, child welfare, and other services (in both their home community and the capitals) inaccessible, confusing, and unsym- pathetic. The powerlessness of women is a factor in homelessness This problem of homelessness for women has a lot to do with women having useless boyfriends. Women are doing what their boyfriends say and they end up being a slave to their boyfriend. I did the same thing in some ways. Every time I left he would get me back with sweet talk. I took off from him many times, but listening to his sweet talk turned my head, plus I couldn?t find anywhere to live. Now I will never go back to him. I want my own place. Many of the women interviewed for this study discussed how powerless they felt. They feel that no one values them as wives, workers, mothers or citizens. They feel that they have been abandoned by their families and communities. They are trapped in abusive relationships in which financial, psychological, physical and mental control are used to keep women feeling powerless. They end up engaging in prostitution and criminal behaviour, because they feel that they have no other option. Every service provider interviewed described the traumatic impact of homelessness on the women they serve. They described the high lev- els of family break down and the chronic state of crisis these women en- dure. They point out that homeless women, and their children, live in a state of high anxiety, and are always tense and afraid. This elevated state of anxiety, tension and fear contributes to the sense of powerlessness that homeless women experience, often on a daily basis. The cost of living in the North and business sector practices can contribute to homelessness I am getting sick and tired of not being able to make it from cheque to cheque, of having to borrow from friends, and of that snowball of trying, you know? I?m always coming up short on the groceries because I have to pay bills like the rent, the electricity, childcare. It?s hard. You have to con- stantly find new ways to make ends meet. Although the gap between the rich and the poor is widening steadily almost everywhere in Canada, the consequences of this trend is espe- cially noticeable in the North. The cost of all of life?s necessities is ex- tremely high in Northern Canada: housing, food, transportation, insur- ance, dental care and prescription drugs, childcare ? you name it. For homeless women and those at risk of homelessness, trying to meet their basic needs is a full-time job. And things are getting worse, not better. The women interviewed in this study were unanimous in their opinion that income support programs simply do not provide enough income to meet basic needs. The wage economy, where jobs exist, all too often rele- gates women to part-time or seasonal work that also leaves women try- ing to choose between shelter, food and clothing. There?s never enough for all three. When they have children, the choices are especially bitter. Landlords play a role in women?s homelessness Last year, I was looking for a place calling around saying I was looking for a job, stuff like that. One of the first questions somebody asked me was, ?Are you on social assistance? Because there?s no way I?m going to let anybody move in here on social assistance.? I was so mad! And you try to tell Welfare that and they don?t believe you. Then there?s racism. I?d phone for an apartment, get right on it, and then when I?d show up, sud- denly the place is rented. And that is what hurts, what makes me mad. Landlords exert tremendous power over low-income women?s lives. They are the gatekeepers between homelessness and housing. Some landlords are helpful, giving a woman a break when she is late with the rent and making repairs when needed. But by far, women reported that their experiences with landlords were negative. Their stories reveal land- lords who discriminate against First Nation women and women on so- cial assistance. They encounter landlords who advertise ?no pets, no children.? Landlords who withhold damage deposits when women did no damage. Landlords who, at best, ignore requests for repairs and, at worst, evict women who complain about the conditions of their housing, conditions ranging from leaky ceilings, to mice, mould, no locks, and no heat. Often when landlords do make repairs, they raise the rent, forcing low-income women to find other accommodations. Absentee landlords have told women they were unaware of the un- safe or unhealthy conditions of their property, an unacceptable argument at best. Women?s perceptions are that absentee landlords do not care about what goes on with their property as long as they make a profit. Women are living in unsafe, unhealthy conditions because there are no alternatives. In our affluent society, it is incredible that women and chil- dren live in such intolerable, substandard conditions. Some landlords walk in unannounced on tenants? units and check things out when they please; others sexually harass their tenants. Land- lords control whether women can find a place to live, determine their living conditions, and can evict them at any time without cause with two weeks? notice. Landlords are part of the problem and could be part of the solution to homelessness with a little imagination and kindness. Societal indifference/punitiveness toward the homeless (including racism) is a barrier to progress They judge you from the way you were, not the way you are. Emotion- ally, it really hurts your mind trying to understand why this happened to you. Facing reality. Feeling suicidal. Worthlessness feeling. I don?t ask for help from anyone. I think we have to live like the old days. It would be a lot better. It was more community oriented in the past. We need to help everybody out like our ancestors did. It?s easier to live in the South than in Nunavut. I am not proud to say that, but it?s reality now. Regardless of where you live in Canada, the homeless tend to be nega- tively stigmatized by other members of society. Homeless persons are often judged and mistreated, based on the stereotypes. The situation in Nunavut illustrates the systemic nature of this discrimination. The vast majority of Nunavut women interviewed believe it would be more bene- ficial to go back to their traditional way of living. It was expressed in numerous interviews that, ?Qallunaat [southern Canadians] are taking over our land.? While most women maintain that they feel no prejudice toward White people, they feel the ?White way of life? does not fit their traditional lifestyle and has further complicated their living situation. Several women suggested by simply looking at the homes owned by Inuit versus that of Qallunaat, that the message is clear as to who is val- ued the most. This gap continues to increase, suggesting a systematic failure. Service providers in the Northwest Territories point out other as- pects of societal indifference and punitiveness toward homeless people. Downtown areas in cities are becoming increasingly hostile places to be for those on the streets. As well, some service providers note that not all government employees understand the circumstances and realities that many women face, and develop policies and programs that penalize rather than help. The Impacts of Homelessness on Northern Women It is difficult to separate the impacts of homelessness from its determi- nants, as these two sets of factors are often cyclical. Our data generated the following five themes. Family separation After my common-law went to jail, I ended up not being able to pay the rent. Income Support wouldn?t help me because they counted the money my common-law made in the month before he went to jail as income, even though I didn?t get any of it. On top of that, I found out he hadn?t paid the rent for two months. The landlord gave me a note to get out in ten days, so I ended up on the street... Child welfare said they would put the kids into foster care until I could find a place to live. It just broke my heart. My kids cried and screamed at me. They weren?t even in the same house together. I went to housing right away, but they told me there was a huge waiting list. Then they said I couldn?t even get on the waiting list because my kids weren?t with me so I didn?t qualify. Isn?t that a joke? I need to get a place so I can have my kids, but I can?t get a place because I don?t have my kids. I went three months without seeing my kids, and without having a place to live. I just couldn?t face them. The separation of family, whether it is partner relationships, parent-child relationships or sibling relationships, is a common challenge associated with homelessness. Partners often separate as a result of family violence, youth may be ousted from their homes by parents, and children may be apprehended by child welfare authorities from parents who are home- less or are living in violent situations. Often a temporary separation leads to a larger breakdown of the family structure, which then results in permanent parent-child separation as well as family members living apart from the support of extended families and communities. Children?s well-being I remember when we stayed with my sister for a few months after I left my partner. I didn?t have any money to rent a place, but I just couldn?t handle the drinking and abuse anymore. Of course I?m grateful that she let us live with her, but it was really tough. My kids and I slept in the liv- ing room, and so we had to roll up all the bedding and put it behind the couch every morning. There was constant tension about things that got broken because my kids played with their cousins? toys or dropped a glass. Everything was always upside down, with so many of us in one lit- tle space. We had no privacy and my nerves were constantly on edge. I was always yelling at the kids, but it wasn?t really their fault. They were just being kids. One of the most difficult aspects of being homeless for women is seeing how the life they lead is affecting their children. The women who par- ticipated in this research project were eager to talk about their struggles to keep their children safe, healthy and happy. They spoke about the im- pact of poverty on body and spirit; of the inappropriate, over-priced and inadequate housing they are forced to accept so that their children will have a place to sleep; and of how the restrictions imposed by landlords and other tenants place unreasonable demands on family life. They also spoke about how their relationships with their children?s fathers were shaped by their homelessness. They anguished about the teasing their children had to endure and the things their children worry about that other, more fortunate, children would not even understand. They see their children acting out and losing self-esteem. Loss of resources through the vicious cycle of homelessness When I first lost my place, I was working at the hotel. I missed a few shifts because I was trying to find a place to crash and store some of my stuff. I needed to find a babysitter too. I phoned everyone I know, but by the time I found a place and someone to watch my kid, I had missed too much work. My boss was pretty understanding at first, but after things at my friend?s house didn?t work out, I was in the same situation again. My boss said he understood that I was having personal problems, but if I couldn?t be relied on that he would have to find somebody else. I was up- set to lose my job. I understand why I was fired. When I was couch surf- ing with my kid I would be tired because people would be partying and then I would be late, or not be able to come in at all. But, after I lost my job I couldn?t get a place at all, and my friends got sick of having me around all the time. On top of that Income Support told me they wouldn?t help me for two months because I had been fired. Homeless women suffer a myriad of losses. Lack of stability in housing and poverty create a maze of dead ends for homeless women, who, once they are in that situation, fear and plan for when it will happen again. The lack of amenities creates a physical appearance for women that acts as a barrier to employment, which is the most cited way that women try to improve their personal circumstances and gain a home. Physical and mental health It?s hard when there?s no food. I come from the old days when you were shy and embarrassed to ask for help or for money or food, so I can?t even ask my son. I have one child and he?s twenty-eight and I?m embarrassed to ask him for food. I sometimes sneak in a hint and say, ?Oh, I haven?t had anything to eat.? Sometimes you can go to relatives and have a little bit of bannock and some tea and that?s great but, how long can you live like that? There are some days you just can?t get through. I haven?t eaten a meal in over a week. An obvious result of poverty is malnutrition. Women are often forced to go days without sustenance for their already weary bodies. Poor nutri- tion results in countless physical ailments, which further complicate the capacity of these women to better their position. Without proper re- sources (e.g., homeless shelters), personal hygiene is also a compromise women are forced to make. Illness is common for women living in overcrowded conditions. The North?s dire overcrowding issue (particularly in Nunavut), combined with building design flaws, contributes to a variety of respiratory and communicable diseases. Continued feelings of worthlessness also affect a woman?s mental health. Many women cite depression as a common emotional response to their unfavourable situations. Several also admitted to feeling suicidal. Women with children were particularly hard on themselves. Feeling as though they had failed at motherhood was the most painful emotion they endured and often led them into severe depression. Survival sex and criminalization When you?re desperate, you go with this man even though you don?t want to. You don?t love him, you don?t like him, but he has a bed to sleep on. You have no choice but to follow him, because you need a place. You get kicked out when the bars close, so you go to sleep in an alcoholic?s house. But, if you?re not willing to have sex, you get kicked out of there too. Poverty-stricken people are often forced to prostitute themselves in a variety of ways to meet the basic needs for survival. Women are abused in different ways than men, as women are often forced to engage in sex- ual relationships in exchange for accommodation. A community pastor in Nunavut expressed his concern by saying, ?One girl told me she?s been prostituting herself since she was a teenager. There are young girls coming up learning this same thing and will eventually take her place. I?m really heartbroken for them. I see what they are forced to do to pro- vide for themselves.? Policy and Bureaucratic Environment When the interview and focus group data was coded and compiled, the following 15 themes emerged. Inappropriate income support policies and services Income Support does not help at all. When you have kids the money is just too small, and if you don?t go by their rules, they cut you off. No good. Oh yeah, and they deduct everything! Even if you get money from your relatives, they deduct that too. Bingo earnings even! The homeless women interviewed all expressed frustration, confusion, despair and anger with the policies and bureaucratic practices of the ser- vices allegedly at their disposal. In general the respondents do not per- ceive that these agencies, and their employees, are interested in helping them or will do anything other than make their already difficult lives more unbearable. They were particularly distressed about their interac- tions with Income Support programs. They feel the rules that guide these programs are punitive, onerous and opaque. Waiting times are too long, and have to be restarted every time someone reapplies. Even when women do qualify for support, the level of their benefits is not sufficient to cover basic living expenses. For example, food money often only lasts for two weeks. Women feel that they can?t break the cycle of homelessness. For ex- ample, many homeless women have lost their housing because of rental arrears, and Income Support policies make it difficult to get caught up, or to secure enough money at any one time to cover a damage deposit and the first month?s rent. Women living in a shelter cannot receive income support, and they may face a waiting period when they leave, so how can they make the transition from the shelter to rental accommodation? Women also com- plain about the attitudes and actions of income support workers, who seem to care more about the rules than people and who sometimes in- trude into what women consider their personal lives. Service providers who work outside the system are no less critical of the income support program, which they describe as inadequate, unresponsive, unprofes- sional, unethical and irresponsible. In the final analysis, service provid- ers wonder to whom income support is accountable. Jurisdictional issues I?m from a small community in Nunavut, but I came to Yellowknife to get help with my addictions. I brought my kids with me. I just couldn?t leave them behind. I?d just miss them too much, and besides, I don?t trust any- one else to take care of them. Well, I found out things aren?t easy in Yel- lowknife. I couldn?t get help from Income Support, and I couldn?t find anywhere to live. I stayed at the Women?s Centre for a while, but you can?t live there forever. The waiting lists for all the housing are a mile long! I?m at the bottom of the list anyway, because I?m from Nunavut. Everyone says I should get help from the Nunavut Government, but they didn?t help me when I lived in the community, so they sure won?t help me now that I?m in Yellowknife. I just don?t understand the rules. How can I get the help I need? Despite the challenges of living in a large centre, many women continue to move to capitals from smaller communities in the Yukon, Northwest Territories and Nunavut. The women migrate to flee intolerable situa- tions and to access services they need. Relocating from one community to another can leave women in a jurisdictional ?no man?s land.? For ex- ample, Aboriginal women who leave their home communities often lose the support of their own Bands, and they do not qualify for support from the Band government in their new community. In addition, women from Nunavut are not eligible for some services in the Northwest Territories unless specific funding agreements are in place between these two juris- dictions. Women also have a hard time understanding the different poli- cies and rules that may be in effect in their new communities. The cost of travelling within the North means that women who leave their commu- nities have a difficult time returning home. Service providers cited the following situation that arises because of jurisdictional issues: Nunavut sometimes sends women to NWT on shared services agreements or Nunavut women are hired to baby-sit in Yellowknife, but end up drinking and on the street; Yellowknife Health and Social Services will pay emergency rent only for these women, and the woman herself has to find the place. Lack of support for 16 to 18 year olds I have a teenage son who doesn?t want to stay at home because my part- ner is so violent. I really worry about him. He has nowhere at all to live, and no one will look after him. He told me that some men have been giv- ing him money and are driving him around in their trucks. That really scares me to death! Every day I worry about him committing suicide. I meet him uptown everyday and bring him food. Every day he looks worse and worse. The Salvation Army says he?s too young, and the social worker only looks after kids up to sixteen. Nobody wants him. He?s living in a truck now. I?d like to leave my partner and take him out of Yellow- knife, but I have no money and no one will help us. Young people between the ages of 16 and 18 don?t fit the criteria for most programs. In some cases they are too young to access services. They are still considered ?children,? and are therefore not eligible for services geared for adults. If they leave home because of abuse or other family problems, they are expected to go into a foster care situation and receive services in that way. On the other hand, they may be considered too old to be eligible for some services. If they have ?graduated? from foster care, they do not get services under Child Protection programs, but are still too young for Income Support benefits. Many other programs designed to assist homeless people (e.g., those of the Salvation Army) will not serve them. Some informants indi- cated that some local hotels don?t ask teenage girls for identification, making it possible for these girls to use their premises for prostitution and drugs. Child protection policies and programs I worry about the neighbours. They can phone Child and Family Services on you. Then they check it out and can end up taking my son if I don?t have enough food or the place isn?t clean enough. For some women, child protection has played a role all through their lives. They have been in government care as children and now their own children are in care. Women who are homeless can have their children apprehended by Child Protection Services. Women living in unsafe housing situations, staying with relatives or friends or in substandard housing, live in fear that their children will be apprehended. They do not feel the child welfare system is there to help them. Other support services My neighbour is having a hard time. The last place she was in, she was evicted because she was asking for repairs; happens all the time. He kept her damage deposit too. Anyhow, the landlord has a reputation as a real scam artist. My neighbour went to the Human Rights Commission be- cause there were some discrimination issues as well, but nothing ever happened. Then she went to see the Landlord and Tenant people, but they said there was nothing they could do because he gave her a timely notice. What kind of use are they? We don?t seem to have any rights. Homeless women access many support services with varying degrees of success. Some services are helpful and others contribute to their prob- lems. Most women interviewed found the rules that govern service pro- vision to be inflexible, particularly in government agencies. This inflexi- bility prevents sympathetic workers from giving women the kinds of support that would enable them to climb out of homelessness. Employment Insurance is not helpful for women who are self- employed or under-employed in part-time, contract, seasonal, or low- wage work. Women feel discouraged and further marginalized when services they turn to for help do not help them. Women with disabilities have fewer options and can feel frustrated and constrained by the ser- vices meant to assist them. When support services work for women, they can make their lives and their children?s lives easier and their poverty seem less demeaning. Public housing policies and programs It is really impossible to get a place to live in Yellowknife. I have applied to Yellowknife Housing three or four times to get an apartment or some- thing. Right now I am living in an emergency unit at the YWCA Transi- tional Housing Program. But I can?t get points with YK Housing and move up on their list if I live at THP because then they say I already have a place, so they don?t care. But you can only stay at THP for a year and then what? I heard that YK Housing is only for families and there is a three-year wait if you?re single. Many of the women interviewed despair of getting a place to live. Some do not qualify for subsidized housing, because they have rental arrears or debts for damages to their former housing hanging over their heads. Although in most cases the arrears and damages were the result of their partners? behaviour, they are still accountable for these debts and will not be assigned another unit until they clear up these charges. If the women are single, they are placed so low on the housing lists compared with families, they know their turn will never come. The waiting lists for all subsidized accommodation are just too long. Women also feel that the housing authorities in some communities are not impartial. Women get denied housing as a punishment for leav- ing a man from an influential family, even if he abuses them or if he gets violent and damages property or for speaking out about the injustices they perceive in the system. Service providers also expressed a good deal of frustration with Housing Authority policies. The Yellowknife Housing Authority says prospective tenants have to be back on the street after staying at the Transitional Housing Program (THP) before they can be admitted to social housing. The Housing Authority works on a points system. Women only get points if they are homeless. Staying at THP doesn?t count as homeless to the Housing Authority. According to ser- vice providers, this policy does not help anyone. The Landlord and Tenant Act I don?t understand why there aren?t rules for the landlord. Like they?re getting all this money from the Government from Social Assistance for rent and they don?t spend even $20.00 to fix anything. There was a prev i- ous tenant in that building who put in a complaint to the Human Rights Commission but nothing ever happened. Women are homeless because of this system. We need a better law that doesn?t give the landlord all the rights and power. The Landlord and Tenant Act in the Yukon is an antiquated piece of legis- lation created in the 1970s and never updated. It has virtually no protec- tion for tenants. Landlords have the right to evict a tenant with two weeks? notice any time of the year with no cause. The act does nothing to protect a tenant?s basic human right to adequate, safe shelter, especially in the winter. This leads to abuses by landlords who refuse to make re- pairs and can evict ?troublesome? tenants without fear of repercussion. Women will stay in unsafe and sub-standard housing for fear of eviction. Many women live without leases on a month-to-month agreement with the landlord, so there is no protection there for them. The act has a direct influence on women?s absolute and relative homelessness. The act needs to be updated to create a fair balance between the rights of land- lords and the rights of tenants. It is a stumbling block impeding positive change for women in low-income housing. Addictions treatment services I got sent to the treatment program in High Level. It?s a great program, and for a while I stayed sober. But I guess I haven?t dealt with all the is- sues underneath the drinking. I haven?t really healed from the sexual abuse and from all the violence and stuff I witnessed as a kid. I?m also grieving for the loss of my child and for all the deaths in my family be- cause of substance abuse. I wanted to go back to the High Level program, but for some reason they won?t send us there anymore. There just doesn?t seem to be much help for someone like me. There is a drastic shortage of appropriate addictions treatment programs for women in the North. Small communities may have almost no ser- vices beyond a wellness worker who is not trained to provide counsel- ling and can only offer referral services. There are almost no programs, even in larger centres, to refer women to. There is one residential addic- tions treatment program in the Northwest Territories, but it is co-ed. Some women commented that the program in High Level, Alberta, was a help to them in the past, but that this option is no longer available. An- other challenge that they face is that, if they place their children in care while they are undergoing treatment, they could face difficulties regain- ing custody. Women also cannot receive income support if they are in a residential program, and so cannot maintain a household to support their children or to come back to once they finish the program. Aboriginal government services When I left the community, the Band took my house and gave it to the Chief ?s niece. They didn?t even tell me. I asked the Band to get me a house in Yellowknife because I have to live there [owing to a disability]. I can?t move back. I told them, ?If I lived at home, you would have to help me. I could own my house, and I wouldn?t be living poor and on the street.? My Band gets lots of money from impact benefit agreements, but they won?t help me. They don?t care about me. They say Yellowknife is responsible to take care of me because I live here now. They forget the hard work I did all those years in the community, and I am mad that I live as a beggar in my own land. Some women find that their Band offices were generally considered more approachable and accessible than territorial or federal government programs. Once women are out of their home communities, they often lose that source of support. Other women complained that services run by their Band offices are open to corruption. Powerful families control the programs, and if, for some reason, you are out of favour, you have no recourse. These women complain of favouritism and a lack of transpar- ency in the administration of benefits. Inuit organizations All these non-Natives are getting all kinds of benefits. I?ve called NTI and told them they should be straightening up their policies to better serve their people, and why are so many non-Inuit getting this and that. They told me they would look into it, but that was over ten years ago. But I re- member In 1993, the Inuit, the government of Canada and the government of the Northwest Territories signed the largest Aboriginal land claim agree- ment in Canadian history. At the same time, legislation was passed lead- ing to the creation of a new territory of Nunavut on April 1, 1999. The new territory was to have a public government serving both Inuit and non-Inuit. Various private corporations were founded to guarantee that the lands claims agreement was upheld, and the rights of all Inuit per- sons preserved. The Nunavut Tunngavik Incorporated (NTI) represents the 21,000 Inuit of Nunavut; Qikiqtani Inuit Association (QIA) represents the inter- ests of the Inuit of the Baffin Region, the High Arctic, and the Belcher Islands; Kitikmeot Inuit Association (KIA); and Kivalliq Inuit Association represent those respective regions respectively. The Inuit Tapiriit Kana- tami (ITK), a Canadian-wide body, represents the four Inuit regions of Canada and has at heart the interests of the Inuit at the national level. All organizations have as their objective to work to improve living condi- tions, both socially and economically, for all Canadian Inuit. All the women interviewed in Nunavut knew of the existence of these organiza- tions, but many were cynical about their usefulness. Several women felt the organizations that were established to represent them were not communicating with the public, and information gathered was not easily accessible. The women expressed an increasing frustration with Inuit associations and felt racism was taking place against their own people. Municipal government policies and services A lot of the Yukon Housing houses are out of town. ... I?ve never learned how to drive. I?ve never had a vehicle and the transit system, well that?s a big question mark in my mind. ... I?m the kind of person who could rent a cabin for $50 a month in the bush with no running water, no electricity and I would be fine. I know how to do all of those things. I could grow a garden but how do I travel? You have to have a vehicle. You can?t rely on the bus system. I would go live in a tent, but you can?t do that in White- horse. I?ve learned that you have to go outside the City of Whitehorse, and then you?re no longer safe. I could afford to live on my income then. You can?t win either way. The situation in the Yukon illustrates the issues for homeless women related to municipal government. Yukon has one city, Whitehorse, which is not directly involved in providing services to homeless people. How- ever, the City has directly contributed to homelessness by the demolition of the cabins in the Shipyards area. People living there as a ?lifestyle choice? managed to find other places to live, but people with fewer op- tions have not fared so well. One First Nation elder had camped on the banks of the river for many years in the summer. This was a traditional practice, not a lifestyle choice, which is no longer open to her. The City?s plan for gentrification of riverside property did not take into considera- tion the poor and thus inadvertently created homelessness. The City of Whitehorse has a bylaw prohibiting living in tents within city limits. This has forced out homeless women who see this as a viable option during the warmer months. Living in a tent is not only a chance to live within their means and within social assistance rates, but also a traditional prac- tice for First Nation people. However, women who want to do this are forced outside city limits that are quite large. This creates further problems, such as transportation and safety. The City?s transportation system creates problems for all those who rely upon public transportation, including the Handibus that serves disabled people and seniors. There is no service at night and lim- ited service on the weekends. The Handibus seems anything but handy with many rules and limited service for people in wheelchairs. In 2002 the number of people using wheelchairs in Whitehorse was estimated at 57, but there are only two spaces on the Handibus for wheelchairs and this is a population that has no other affordable option. The City?s bus system has been the object of extensive studies and many recommenda- tions have been made for improvements, however, nothing seems to change. Limited resources and cuts to the voluntary sector There hasn?t been any meat at the Women?s Centre for two months now. No one has made any donations of meat. There never seems to be enough food or anything else and Income Support doesn?t give us any money as long as we are living at the Centre. Government support for the voluntary sector services that so many homeless women depend on for help with many aspects of their lives is limited and becoming scarcer. Programs are trying to meet the needs of an ever-growing population of women who are either on the street or who are staying in intolerable situations, just to have a roof over their heads. Some of these women have severe mental and physical health issues. Existing services cannot afford staff with the right qualifications to serve these women. Shelters are overcrowded and understaffed. They may even run out of the commodities they need, like food, hygiene products, and furnishings. Minimum wage policies I make $6 per hour at a job and it is not enough to live on. It is the mini- mum wage and I can?t make it on that. It would be a lot easier if I had a place to stay so I could make more money. My sister has three jobs, and she is raising her grandson too. It still isn?t enough, and she?s afraid Yel- lowknife Housing will evict her because she is sometimes short on the rent. Homeless women are affected by the low minimum wage in the Territo- ries. They cannot afford even a small apartment at market rental rates without holding down several jobs that pay more than minimum wage. The justice system In my community, they?re going back to traditional laws. That scares me. It?s who has power in the community and who doesn?t. It?s harsh. Some communities have circle sentencing and some have traditional justice. That?s not so bad. But I don?t think a homeless woman gets much help from any justice system. I knew a woman who was trying to get her property rights in her community and they were just siding with the hus- band. She was left with nothing and there wasn?t a thing she could do about it. The man has all the rights. Homeless women do not feel that the judicial system is there to help them. Women do not know how to navigate around the justice system. Whether in matters of separation, child custody, damage deposits, dis- putes with landlords or First Nations justice systems, women feel power- less and without the necessary resources to represent their interests. They found the Neighbourhood Law Centre staff helpful, but bound by the legislation. Women have lost faith in the justice system. Bureaucratic indifference My sister and her husband finally got an apartment. He was always beat- ing her up, though. The last time he almost killed her. She has tried to kick him out, but he keeps coming back and she can?t stop him. She tried to get Housing to change the locks, so she could feel safe at night. Because his name is on the lease, they said she had to get a legal separation or a divorce before they could force him out. It just doesn?t make sense that a rule should put her life in danger. Women who are trying to get help finding accommodation, securing adequate income support to meet basic needs, caring effectively for their children or regaining custody if they have been apprehended, dealing with legal issues, getting out of an abusive situation, accessing educa- tional opportunities, or healing physical and mental health issues, have to deal with program officers of some sort. Homeless women, or those at risk of homelessness, report that this experience is often frustrating and disempowering because of the bureaucratic nature of these interactions. It feels to them that they are dealing with a ?system? rather than with another human being who understands their circumstances. They feel that if they could get the right help at the right time, they might be able to move out of the often destructive patterns of life that they now find themselves in to a better life. Instead, the system just seems to keep pushing them down. Service Provision in the Territories When the interview and focus group transcripts for this study of home- lessness among Northern women was completed, the following five themes emerged. Physical environment of housing services I just can?t seem to find a decent place to live that?s affordable in this town. The first place I spent a winter in was unbelievable. The furnace broke down and it went to minus forty. It was really cold and the room started to ice up. The landlord didn?t do anything. The physical environment of low-cost housing appears to be largely sub- standard. Very few women we interviewed were satisfied with their housing. Stories of mould, dirt, mice, thin walls, inadequate heat, poor maintenance, and leaky windows that do not lock were abundant. Women living in low-income housing and social housing related that their neighbourhoods were rife with drugs and alcohol and they did not feel safe. Some women found used needles and drugs in the hallways and yards of their buildings and were worried about their children?s safety. These conditions applied to social housing and private rentals alike. Homeless women and their children are forced to live in unsafe, unhealthy sub-standard housing because there is not enough decent, affordable social housing. There are few wheelchair units in social or private market housing. The units that many women in wheelchairs live in are not fully wheelchair adapted, creating unsafe and inconvenient living conditions. Decent housing for all women is a basic human right. However, so- cial housing units cannot be properly maintained and repaired unless governments are willing to make this a priority and expend adequate funds. In the private rental market, low-income housing is not main- tained because it is not seen as profitable and there are always enough poor people to fill vacancies. As for women in desperate need of emer- gency housing, the few emergency shelters are overcrowded and are not always gender-specific. Until governments acknowledge the right to de- cent housing, women and children will continue to live in conditions that most Canadians would not tolerate. Service effectiveness I used to go live at Mary House when I had nowhere to go, but now they don?t have that open. One time I went to Kaushee?s, but I wasn?t being abused that time, didn?t have a boyfriend, so they kicked me out. That was really bad that time. I know some women don?t like it there because they can?t bring their boys there if they are fourteen years old. I don?t think that is right. It?s just separating your family again and that is sup- posed to be where you go to get help. They should have a place where you can take your family. Homeless women use a number of services in the larger centres and ru- ral communities. All have policies and requirements determining who can use the service, how long the service can be used and the length of stay for shelters. Women interviewed related both positive and negative comments for most services. Many found regulations restrictive and not responsive to their needs. The lack of emergency shelters for women puts the women?s transition homes and detox centres in the position of having to turn away women who are seeking shelter but do not fit their mandate. Both services are usually operating at capacity and do not have room for flexibility. There is very limited second-stage housing for women leaving the transition house. Emergency shelter for youth outside Child and Family Services facilities is non-existent. Most services regarded as helpful were provided by non-governmental organizations. Most negative comments were the result of lack of funding for the agency to adequately meet the needs. For example, the Salvation Army in Whitehorse has only 10 emergency shelter beds on a first-come, first-served basis. There are no beds for women with children. Men usually get there first and women feel intimidated. When beds are full, people can sit in the dining area. They can lay their heads on the tables and sleep. The Salvation Army is unable to offer day program- ming or a regular food bank. In the Yukon, services such as the Women?s Advocate at the Victoria Faulkner Women?s Centre (VFWC), the Fetal Alcohol Society of the Yukon (FASSY), and the Committee on Abuse in Residential Schools Society (CAIRS) are able to provide more flexible services and were consistently found helpful. Lack of housing options There is no housing in my home community. I had to live with my three kids at my boyfriend?s sister?s place. But I might have to go back there be- cause there is nowhere in Yellowknife for me to live and Income Support won?t help me. There is a critical lack of affordable housing in all three territories. This means that for some women the emergency shelter has, in fact, become their permanent home. One obvious gap is that there are few apartments for single people. The lack of affordable housing has a devastating im- pact on women with children. If they are only getting rent from income support for a single room, as they do if for any reason their children are not in their care for a period of time, they cannot get their children back because Child Welfare policies stipulate that the mother has to have ade- quate accommodation (that is, an apartment). Food security I use Social Services and the soup kitchen to eat mostly. Going to your sis- ters or your brothers and collecting a little bit of food from them helps too. [My children] don?t want to go to school anymore because they don?t get any sleep and they don?t want to go hungry. Feeding our children is the biggest challenge. The provision of food is one of the biggest challenges faced by the home- less. With little income, it is a constant daily struggle to find their next meal. The only obvious difference to these women between a healthy and unhealthy choice is the price, and food with no nutritional value provides a greater quantity at a lesser cost. A poor diet inevitably leads to various health complications, which further hinders a woman?s capac- ity. A woman often has many mouths to feed, and will go hungry herself to feed her children. These women experience feelings of disempower- ment when they continually fail to feed their children and themselves. Lack of Specialized Services for Women The New Horizons Centre is really geared for men. They do things there like go fishing and do other men-type activities. But there is food there and you can sleep there during the day and they help with resumes too. Some women talked about services being geared largely towards home- less men. Service providers also point out that there are no specific ser- vices targeting homeless Aboriginal and Inuit women. Day program- ming for homeless women is also an issue, because some shelters close during the day and the women do not necessarily know about or want to go to other programming. This is because the women find some day programs ?too white? or insufficiently advocacy-oriented. Recommendations The following recommendations were developed by the research part- ners after a careful review of the findings of the study carried out in all three territories. Recommendation 1 ? Create a national housing policy that takes into account the special circumstances and needs of vulnerable women: (a) Creation of a Na- tional Housing Policy instituted by the federal government that is inclusive of women and lives up to human rights obligations under the International Cove- nant on Economic, Social and Cultural Rights guaranteeing a right to an ade- quate standard of living and adequate housing, and ( b) Ensure that women?s housing needs across their lifespan are met. Recommendation 2 ? Increase the supply of decent, safe low-income housing: ( a) Ensure an adequate supply of a variety of low-income housing stock is available for women and children in environments that can be kept safe and secure, and (b) The federal government must provide funding mechanisms to encourage and support the development of low-income housing in the territories. Recommendation 3 ? Increase supportive housing options: (a) Implement a contin- uum of supportive housing options, and ( b) Encourage service providers to identify and develop potential supportive housing options as new initiative proposals. Recommendation 4 ? Increase the number of emergency shelters and improve the quality of their services. Recommendation 5 ? Increase second-stage housing options. Recommendation 6 ? Implement housing authority policies that remove barriers for women living in violence and those who are homeless or are at risk of becom- ing homeless: (a) Apply a cultural and gender analysis to housing authority policies to ensure human rights obligations and the needs of homeless women are met in a way that is measurable and makes the agencies accountable, and ( b) Create priority-housing policies for women leaving abusive relationships. Recommendation 7 ? Address landlord and tenant issues by reforming Territo- rial Landlord and Tenant Acts. Recommendation 8 ? Implement poverty reduction strategies: (a) Improve exist- ing social security programs, and (b) Introduce new programs and policies that are designed to prevent and reduce poverty. Recommendation 9 ? Provide services that address the full range of determi- nants of women?s homelessness: (a) Implement a continuum of care model, and (b) Enhance the capacity of service providers to work effectively with homeless women. Recommendation 10 ? Reduce barriers to accessing services for homeless women. Recommendation 11 ? Ensure appropriate funding for a range of front-line ser- vices: (a) Ensure front-line services are adequately and appropriately funded to build capacity to function effectively, and (b) Recognize the value and contribu- tions of service delivery through the voluntary sector. Recommendation 12 ? Enhance access to education and training programs: (a) Increase access to educational programs, and (b) Increase access to affordable daycare so that women and participate in educational programs. Recommendation 13 ? Ensure access to affordable childcare. Recommendation 14 ? Develop mechanisms for collaborative and creative solu- tion building: (a) Nurture the creation of collaboratives that are dedicated to addressing the full range of determinants of women?s homelessness and build their capacity to function effectively, and (b) Ensure that all relevant stake- holders are ?at the table? when public policy related to women?s homelessness is being developed and when government program decisions are being made. Recommendation 15 ? Collect, manage and share information: (a) Design and implement interagency protocols and tools for collecting, managing and sharing accurate and relevant information as well as for designing and tracking clear outcomes indicators, (b) Provide adequate funding to service agencies to allow them to keep appropriate records and to access and share information, and ( c) Conduct further research. Recommendation 16 ? Enhance public awareness and facilitate attitude change. This chapter is based on You Just Blink and It Can Happen: A Study of Women?s Homelessness North of 60, prepared for the Qulliit Nunavut Status of Women Council, the YWCA of Yellowknife , the Ye llowknife Women?s Society and the Yukon Status of Women?s Council by the Four Worlds Centre for Devel- opment Learning, in November 2007. To view the co mplete report, go to http://www.ywca.ca/Northern_Territories_Reports/PAN - TERRITORIAL_PDFS/PanTerritorial%20_FinalReport.pdf . Chapter 5.1 Living on the Ragged Edges: Latin Americans and Muslims and the Experience of Hom elessness in Toronto JASMIN Z INE This examination of housing and homelessness in the Latin American and Muslim communities attempts to map the realities of those living on the unstable peripheries of our society. Research on homelessness among these particular populations is virtually non-existent. This exploratory study hopes to open the door to future research and community devel- opment focusing on the housing needs of these communities. The purpose of this study was to (1) identify the social, economic, and political conditions that contribute to the marginalization and disen- franchisement of Muslims and Latin Americans in West Central Toronto; (2) analyze how the interlocking systems of oppression based on race, class, gender, religion, sexuality, age, mental health status, and disability affect their ability to access and maintain stable housing; (3) uncover the specific housing needs within these groups; and (4) explore the dynamics of informal housing networks. Integrated framework for understanding homelessness Homelessness results from the interaction of social, cultural, economic, and political factors. These include the lack of affordable rental units be- ing built in Ontario, long waiting lists for social housing, low vacancy rates, decreases in social assistance support, legislation that favours lan- dlords, and the lack of strong political will to end the housing crisis. These factors in turn are mediated by an individual?s immigration status, race, gender, language, age, religion, sexuality, mental health, and disa- bility level as additional barriers to securing and maintaining adequate affordable housing. Immigrants and refugees must cope with the difficulties associated with settlement; the shifting narrative of ?home? as being a strange and unfamiliar place, lacking the cultural capital to negotiate the institutional systems in Canada, language barriers, the lack of accreditation for pro- fessional skills, being channelled into lower-paying jobs and often social or emotional isolation. These barriers are compounded when mental health problems, domestic abuse, disabilities and discrimination on the basis of race, ethnicity and religion are involved (Chambron et al., 1997; Murdie et al., 1996). An integrated framework involves examining the inter- connectedness of these multiple factors that mediate homelessness. It is essential not to homogenize the variety of situations along the conti- nuum of hidden to absolute homelessness as circumstances that occur independent of issues of race, class, gender, and other forms of social difference. People live those experiences differently according to their particular social locations. The issues affecting homeless women, for ex- ample, cannot be fully understood if we view women as a universal so- cial category unmediated by class or race. Therefore, homelessness as a wide-ranging social phenomenon must be understood through the ways it is related to race, class, and gender. Informal housing networks A significant focus of this research is the understanding of how informal housing networks operate within these communities. People in need of housing help turn first to their family members and then to their friends and other community members (Rose et al., 2000). These contacts pro- vide access to resources that can help with housing-related concerns. However, many newcomers lack these supports and also lack access to formal housing support systems provided by housing help centres. This study examines the possibilities of interventions to strengthen both in- formal as well as formal channels to ensure eliminate barriers to access- ing housing help. Latin Americans and Muslims in Toronto Latin Americans and Muslims in Toronto represent ?communities in exile? that have created local diasporic societies. Whether fleeing war, civil strife, or persecution, reuniting with families or looking for better economic opportunities, newcomers from these communities are recreat- ing a home in Canada and rebuilding the framework of community through new social networks. In 1996, 61,655 Latin Americans were living in Toronto. More than half had immigrated to Canada in the 1980s (the majority between 1982 and 1991). Latin Americans are therefore among the youngest immigrant groups in Canada. As a group, they tend to have less formal education, low labour force participation, and a high rate of unemployment. The 1996 Census data did not provide population statistics accord- ing to religious groups, therefore it was difficult to determine the current numbers of Muslims in Canada.1 1 Census data on population by religious groups are only tabulated every 10 years. According to the 1991 census, there were 105,970 Muslims living in Toronto; current projections place this figure at 325,000 to 350,000 in 2001. The Muslim community is ethnically and racially diverse including South Asians, Africans, West Asians/Middle Easterners, Iranians, Afghanis, Indonesians, and Malay- sians, as well as many from the Caribbean. There are also many Anglo- Canadian and African-Canadian converts to Islam. Method Quantitative methods A survey of housing needs was developed in consultation with the pro- ject?s advisory committee. The survey was translated into Spanish, So- mali, Urdu, Arabic, Farsi, and Dari and disseminated through ethno- specific agencies and LINC (Language Instruction for Newcomers) classes in the West Central Toronto area. The survey was also adminis- tered during two housing seminars conducted by a project advisory member from Syme-Woolner Neighbourhood and Family Centre A total of 300 surveys were completed between October 2001 and March 2002. Specific sites were chosen to administer the survey, includ- ing LINC classes, seniors groups, women?s organizations, youth pro- grams, and mental health support groups. Qualitative methods Interviews and focus groups were conducted with three categories of informants: (1) people experiencing situations of absolute or hidden homelessness in the Latin American or Muslim communities in West Central Toronto; (2) service providers in agencies dealing with home- lessness among these populations; and (3) people involved in informal housing networks, such as family members, friends, or members of church groups and mosques. Ten interviews were conducted with indi- viduals in the absolute and hidden homeless categories: three were Latin Americans and seven were Muslims. Among the seven service providers interviewed, five were Latin American and two were Muslim. The sample Of the 300 participants surveyed, 44 percent were from the Latin Ameri- can community and 56 percent from the Muslim community. The Latin American participants came from Central America (19 percent), South America (21 percent) and the Spanish Caribbean (4 percent). Within the Muslim community, the largest number came from South Central Asia, including Iran and Afghanistan (17 percent) and the Horn of Africa re- gion, including Somalia, Eritrea, and Ethiopia (14 percent). Other regions represented include West Asia (Turkey, Azerbaijan, 6 percent), East Afri- ca (Kenya, Tanzania, Sudan, 5 percent), South Asia (Pakistan, Kashmir, and India, 4 percent), the Middle East (3 percent), and the Balkans (Bos- nia, Kosovo, Albania, 2 percent). A small number (0.3 percent) came from each of these regions: Central Asia, West Africa, and the Caribbean. Length of time in Canada Of the sample, only 2 percent were Canadian-born. The majority were relative newcomers, with 70 percent having lived in Canada five years or less. Of the most recent newcomers, 36 percent had been in Canada one year or less, while 34 percent had lived in Canada two to five years. A smaller number (14 percent) had lived in Canada for between five and ten years and 14 percent had lived in Canada more than ten years. The predominance of newcomers is due to the sites where the survey had been distributed, such as LINC classes and through ethno-specific set- tlement agencies. Immigration status Of the sample, 40 percent were landed immigrants. The highest propor- tion, 49 percent, were refugee claimants, 6 percent were sponsored refu- gees, and 4 percent were waiting for status. According to the Golden Re- port on homelessness (1999), refugee claimants are the most at risk of homelessness. Of the refugee claimants, 42 percent were Latin American, or 52 percent of the total Latin American sample. In the Muslim community, 53 percent were refugee claimants, or 46 percent of the total Muslim sample. In the sponsored refugee category, 82 percent were Muslims and 18 percent were Latin Americans. In the overall sample, 83 percent of Latin Americans indicated that they were waiting for status, while 17 percent of the Muslims were in this category. Reasons for immigration War and civil strife were the most significant reasons for immigration, with 30 percent of the sample indicating this as the reason they fled their homeland, while 22 percent cited political reasons for migration, and 18 percent were fleeing personal violence. Another 18 percent had left their country to find better economic opportunities, while 10 percent cited ?other? reasons. Of those fleeing personal violence, 59 percent were Latin American and 42 percent were Muslim. Of those fleeing war and civil strife, 26 per- cent were Latin American and 74 percent were Muslim. Of those partici- pants citing political reasons for leaving their homeland, 49 percent were Latin American and 52 percent were Muslim. Of those leaving their ho- meland in search of better economic opportunities, 61 percent were Latin American and 39 percent were Muslim. Languages spoken Overcoming language barriers is a necessary first step to integration. In this sample, 46 percent indicated that could speak English; 60 percent indicated that they understood English to varying degrees, though not all were able yet to speak fluently. The sample included 43 percent Span- ish-speaking individuals. Among the Muslim participants, the language groups represented included: Arabic (10 percent), Urdu (4 percent), Farsi (18 percent), Somali (10 percent), and ?other? (21 percent). In some cases multiple language facility is noted as with many Afghans, who often speak Urdu due to being displaced in Pakistan, and Somalis, who often also speak Arabic and Italian. Education background The highest level of education achieved by most of the participants was high school, comprising 48 percent, while 15 percent had graduated from a community college, 10 percent had a university undergraduate educa- tion, and 18 percent had attained a graduate degree. Only 6 percent stated they had no formal schooling. Cross-tabulation of results on the level of education show that 30 percent of those whose highest level of education was high school were men, while 70 percent were female. While gender differences at the post- secondary level were not significant, among those with ?no formal schooling,? 94 percent were women, or 8 percent of all women in the sample, versus 6 percent of men, or 1 percent of all men represented. Clearly, women have had less access to education than the men in this sample, although those who had access achieved academic levels on par or higher than men, particularly at the community college level. Employment While 49 percent of the participants in this study indicated that they had training in a professional field, only 4 percent were working in this field in Canada. The majority of those who were professionally trained were women, comprising 65 percent; professionally trained men comprised 35 percent. This is consistent with the higher number of women in the sam- ple who had attended community college. Most participants had re- ceived their education in their homelands. Only 3 percent had received their education and professional training in Canada. Only 18 percent of the participants were currently employed. Of this 18 percent, 52 percent were Latin American and 48 percent were Muslim. In total, 21 percent of the Latin American participants and 15 percent of the Muslim were working. Level of education and employment status The findings show a negative correlation between the level of education and employment in Canada. The results are counter-intuitive, as they indicate that the higher the level of education, the lower the employment status. For example, the highest number of those currently employed (57 percent) had only a high school education, and 26 percent had attended a community college. Only 4 percent of those currently employed had a university degree and 10 percent of those working had a graduate de- gree. Of those currently employed, 6 percent had no formal schooling whatsoever. Income sources The primary sources of income for participants in this sample were: wages and salaries (25 percent); social assistance (69 percent); income from self employment (2 percent); child support (5 percent); economic support from family and friends (3 percent); and the Child Tax Credit (10 percent). Of those receiving their income from wages and salaries, 54 percent were Latin American, representing 31 percent of the total Latin Ameri- can community in this sample and 46 percent were Muslims, or 21 per- cent of the Muslim community. Among those receiving social assistance benefits as their primary source of income, 37 percent were Latin American, or 58 percent of the Latin Americans in the study and 63 percent were Muslim, or 77 percent of the Muslims in the sample. The percentage of the sample receiving social assistance benefits de- creases as the length of residency in Canada increases, as would be ex- pected. For newcomers who had been in Canada less than one year, 77 percent were receiving social assistance. For those who had been in Can- ada for two to five years, the number decreases to 70 percent, however it remains at 70 percent for those who had been in Canada for five to ten years. The number decreases to 48 percent for those who have lived in Canada more than ten years. Despite the decline with years of residency, the numbers of those on social assistance is still very high. Even after ten years, almost half of the participants were unable to find employment. According to the Ornstein report (2000), Central Americans, Afg- hans, Arabs, West Asians, Pakistani, Bangladeshi and Somalis were among those categorized as the most severely disadvantaged, with an unemployment rate of 20 to 29.9 percent, as opposed to the overall aver- age of 9.4 percent. Income levels The average annual income of participants in this study ($13,468), is well below the low income cut-off (LICO) or poverty line. Examining the data through quartiles shows that: 25 percent earned below $7,440, 50 percent earned below $13,200 (the median incom e level), and 75 percent earned below $18,000 a year. A total of 90 percent of all participants earned less than $21,600 a year. These income levels are also considerably lower than those cited for the average household income in the catchment area of this study, the former City of York ($43,192). The a verage income level in this region is significantly lower than other municipalities in the Greater Toronto Area (Wallace & Frisken, 2000). Difficulties in finding housing When participants were asked how difficult their last housing search had been, 62 percent said their housing search had been ?very difficult,? 32 percent found it ?somewhat difficult,? and 5 percent found it ?somewhat easy.? With vacancy rates as low as 0.9 percent in Toronto, it is not sur- prising that for the majority, the housing search was ?very difficult.? Lack of income, source of income, and being on social assistance were the most prevalent reasons for why the housing search was diffi- cult. Number of children, lack of transportation, and the need for refer- ences and a guarantor were also barriers to finding housing. Many faced numerous refusals by landlords and ended up accom- modated by ?slum landlords? in sub-standard housing in dangerous areas of the city. According to Dion (2001), being housed in substandard circumstances reduces access to other social and economic needs such as education, health care, and employment. Housing discrimination Participants were given a definition of housing discrimination and asked to indicate whether they felt that they had experienced discrimination on the various grounds that were listed. The definition read: ?Sometimes lan- dlords or their staff refuse to rent to people for unfair reasons and/or people have to pay higher rent than others for no valid reason. This is housing discrimina- tion based on things like ethnicity, gender and income. Do you feel that you faced any in discrimination in finding housing?? In response, 68 percent of participants indicated that they had expe- rienced some form of housing discrimination. Of these, 54 percent were Latin American (80 percent of the Latin Americans in the study) and 46 percent were Muslim (57 percent of the Muslim participants). Level and source of income were the most noted forms of housing discrimination, although the narratives revealed strong examples based on racism and Islamophobia. For example, the Refugee Housing Task Group in Toronto (2001) reported that since September 11, 2001, there has been a reduction of Muslim clients attending programs due to fear of backlash, and some landlords are openly refusing to rent to Muslims. According to a report by the Toronto Police Services (2001), there was a 66 percent increase in hate crimes in 2001; the largest increase was against Muslims. Of the 121 hate crimes linked directly to September 11, 45 incidents were perpe- trated against Muslims, 20 against Jews, and 38 against other groups. In addition to these individual acts of violence, racial profiling at airports, train, and bus stations has also been reported. Finding housing in this political climate is difficult for many Muslims already disadvantaged by the barriers of race, ethnicity, and lower socio-economic status. Racialized groups face constant rejection from some landlords, who have unstated policies of racial exclusion and the ?selection? of tenants from groups seen as socially acceptable and desirable. One participant gave an example of the various circumstances in his housing career that were interrupted by issues of racism and social difference: [When I called one lady about a place,] she sounded optimistic. I told her that I was taking one or two night school courses and I was working part- time. I told her how old I was. I didn?t believe she knew about my race or she didn?t have any preconceived notions over the phone after hearing my voice. But when I showed up there, she had like a shocked look on her face. I don?t know. I guess she had a different face in mind when I told her my name over the phone. She didn?t even let me come in. She told me that somebody had already come by and they were giving the place to them. She said basically, ?Oh, sorry,? [and that] somebody else already had the place. For this individual, who had converted to Islam, Islamophobia among some landlords forced him to use his given Christian name rather than his Muslim name to landlords over the phone: Sometimes I would get different responses in regard to my name and how I would sound on the phone?my Islamic name. So at times I would use my given name?my birth name?just at least to get my foot in the door. Interlocking systems of oppression and housing Interlocking systems of oppression affect daily interactions and broader systemic barriers. These are based on discrimination related to race, class, gender, sexuality, religion, mental health status, age, and ability. According to Dion (2001), ?Housing discrimination is an integral part of an overarching, interlocking system of discrimination that Sidanius and Pratto (1999) have aptly termed ?the circle of oppression? ? (p. 536). Gender The service providers interviewed for this study affirmed that all of the issues related to housing and homelessness were experienced in a more pronounced way by women, particularly women already marginalized by race, poverty, and language barriers. Sexual harassment and abuse sometimes involved landlords taking advantage of single women. Women in the Latin American and Muslim communities must therefore contend with the barriers associated with their gender as well as other forms of social difference, such as race and class that lead to multiple marginalities. Meanwhile, single mothers face multiple burdens as providers of shelter and support for their families. Homophobia Although issues of sexual orientation were not noted as factors of con- cern for participants in this study, there are broader concerns in both communities, particularly regarding lesbian, gay, bisexual and transgen- dered (LGBT) youth who leave home due to their family?s negative reac- tion to their sexual orientation. It is estimated that up to 75 percent of Toronto?s street youth are gay and fleeing family violence and rejection (Wolfe, 2002). A focus group with Latin American Service providers revealed that the social stigma and discrimination surrounding homophobia can be a barrier to seeking formal help, since many young people do not feel safe ?coming out,? given the lack of tolerance they have encountered. The lack of gay-positive support available for those who need housing help leads to greater isolation for those members of the community. Disability Finding affordable subsidized housing was a barrier for people with dis- abilities who may also lack employment opportunities. People with cer- tain physical disabilities face barriers in locating housing that is wheel- chair-accessible. Seniors The number of homeless seniors is increasing as the result of changing family dynamics, whereby elders are brought to Canada to help with grandchildren and to rebuild an extended family network of support. As families grow and change and there is less need for support from elders, many find them themselves in need of alternative support and shelter. Or if overcrowding occurs as the result of families doubling up with grandparents, there is the possibility of eviction. Some seniors also suffer elder abuse from the family members with whom they are housed. Those who are reluctant to disclose the problems they face remain in abusive situations. Youth Racial profiling by police and the negative stereotypes particularly asso- ciated with Latin American youth reduce the varied experiences of these youth to a single negative referent. This social positioning has serious implications for their access to housing. Many landlords see youth as irresponsible and undesirable to house due to their age and poverty. Mental health Research has shown that there is an over-representation of people with severe mental problems among the homeless population. An estimated 33 percent of men in hostels and as many as 75 percent of women have serious mental health problems. There is also significant evidence that homeless people with mental health problems remain homeless for long- er and have less contact with family and friends than other homeless people. They encounter more barriers to employment, tend to be in poorer physical health, and have more contact with the legal system than the rest of the homeless population. Cultural stigmas around serious mental health problems may prevent some sufferers from disclosing their problems to others and seeking professional help. The impact of post-traumatic stress affects many who had to flee situations of war and violence in their homeland. There are few commu- nity-based supports for counselling and housing help. For example, Somali youth who had been displaced from their fami- lies back home and migrated to Canada as refugees, faced many emo- tional stresses associated with the danger and instability of homeless- ness. Living on the streets created fear and anxiety; they also suffered from the stress and anxiety of not knowing what had happened to their families back home. Religion is a spiritual lifeline for these youth, who found solace in spending time at the mosque. While spirituality is an important mechan- ism grounding these youth, who remained committed to their Islamic practices despite their transient and unstable lifestyle, traditional servic- es geared toward the homeless population place little emphasis on this aspect of emotional and spiritual survival (see also Kappel & Ramji, 2002). Also, many Somalis favour indigenous health practices to those based on a Western model. Muslims have a different cultural framework for understanding mental or physical illness that proceeds from a more spiritual worldview. There is insufficient knowledge about Islamic practices within the mental health care system. Cultural or religious practices can lead to the stigmatization of Muslims within the mental health and criminal justice system. Many of these individuals leave the prison system and end up homeless. Certain religious practices are pathologized and seen as evi- dence of mental illness. The events of September 11, 2001, prompted people in the Federal Correctional System to focus on Muslims in the care of the government; this surveillance is not limited to Muslims behind bars. Some have been denied parole and their presence in the system is lengthened. There has been targeting, stereotyping, violence and an anti-Arab, anti-Muslim backlash. Muslim men discharged from the criminal justice system have diffi- culties finding adequate housing, and this problem affected their mental health. The trauma these former inmates encounter in trying to reinte- grate into society is often too much to bear and many prefer to remain institutionalized. In the Latin American community, the loss of occupational status many men have experienced in their transition to Canadian society can lead to mental health problems (Dunn et al., 2000). For Latin American men, this results in the loss of identity and in the traditional role of the man as provider, and leads to feelings of inadequacy, anxiety, and isola- tion, a complex referred to as ?nervios? or emotional distress (Bakshi et al., 1999). The loss of social networks as the result of migration was also seen as contributing to mental health problems for both Latin American men and women. Within both the Latin American and the Muslim community, stigma surrounding serious mental health problems was a barrier to disclosure. While people might report experiencing depression, they are far more reluctant to speak of more serious mental health problems. Although some of the surveys were administered in a mental health support group, only 7 percent indicated that they needed mental health support. Mental health problems may be misdiagnosed. One community mental health rehabilitation worker noted: ?Somebody from the upper classes would be diagnosed with neurosis. If it is somebody from the working class ? or from a different ethnic group, it will be ps ychosis.?2 People with serious mental health problems are often labeled ?hard to house? and face barriers to housing based on the stigmas related to people experiencing these mental health difficulties. The need for more housing that provides support services to people with serious mental health problems was highlighted in the survey. However, supportive housing does not always attend to the religious and cultural needs of clients, therefore the issue of cultural and religious diversity in this sec- tor of service delivery also needs to be addressed. An ability to ascertain a client?s mental health condition and needs was one of the many skills that housing help workers said they needed to provide the kind of holistic care required. Ongoing training and support for housing service providers was cited as a need in order to meet the growing demands of their work. 2 Neurosis is understood as ?being overly in touch with reality? (e.g., overly aware of risks in the environment), while psychosis is a break with reality. Psychosis is perceived to be a more serious condition and is often misdiagnosed when it is ascribed to people from racialized and lower-income communities. Rent and housing Participants in this study were paying an average of 64 percent of their income toward rent. For Latin Americans, the rent-to-income ratio was 60 percent and for Muslims the ratio was 67 percent. By comparison, in low-income subsidized units, rent geared to income is set at 30 percent. Newcomers were paying the highest proportion (69 percent) of their income toward rent. Among those who had been living in Canada for two to five years, the proportion is 65 percent, and among those in Can- ada for between five and ten years, 66 percent. There is a significant drop in the rent-to-income ratio (49 percent) among those who had lived in Canada 10 years or more. While those who had lived in Canada for less than on e year had the lowest monthly income at $902 , there was less income difference be- tween the other categories. In other words, while the length of time in Canada does lower the average rent-to-income ratio, higher levels of in- come are not the reason. It is important to take into account the high number of participants receiving social assistance. While the number of those on social assistance decreased over time, almost half of the sample were receiving welfare benefits after having been in Canada 10 years or more. This is a sad commentary on the difficulties faced by immigrants and refugees from racialized communities in achieving consistent eco- nomic growth, despite having lived in Canada for longer periods of time and overcoming the initial barriers of language and employment. Access to affordable housing Only 14 percent of the participants in this sample lived in social housing with a rent-geared-to-income ratio of 30 percent. The majority (76 per- cent), paid rent to a private landlord. Of those living in public housing, a very low number (9 percent) were Latin Americans, or 3 percent of the Latin American participants in this sample. Muslims represented 90 per- cent of those who paid rent to a public landlord, or 24 percent of the total Muslims in this sample. A total of 42 percent of participants had applied for subsidized housing. Of these, 30 percent had been on the waiting list for up to five years and 4 percent had been on the waiting list between six and ten years. The majority (58 percent), had not applied for subsidized housing. Current waiting lists for social housing in Toronto are 10 years or more with approximately 100,000 people currently waiting for a placement. Being at risk for homelessness Participants were asked whether they felt they were at risk of losing their home and 55 percent said that they were at risk of homelessness. The primary reason given for being at risk, was that their housing was too expensive, with 42 percent of the sample indicating this reason. Not be- ing able to pay rent on time was also a key factor. A total of 19 percent of the participants indicated that they had been without a permanent home in the last five years and 22 percent were currently without a permanent home. The most prevalent reason given for their unstable housing condition was that housing was too expensive. Living in shelters and on the street Many shelters do not accommodate religious dietary restrictions. As a result, many Muslims cannot eat the food provided in the shelters. Par- ticipants in the study also cited concerns with overcrowding, privacy, safety, and cleanliness in the shelter system. The overcrowding in many shelters violates the United Nation?s requirements for refugee camps, which call for 4.5 to 5 square metres per person (Gillespie, 2002). Few Latin American youth use shelters. Many are suspicious of dealing with ?the system? and when they find they are not getting a friendly reception, they back off. Both Latin American and Muslim youth who become street-involved may be more prone to live in overcrowded and inadequate housing. Among the Somali street-involved youth in our study, some had been displaced from their families when they came to Canada as refu- gees. Displacement was causing them emotional stress, compounding their already difficult situation. These youth were interested in staying in school, but found that difficult due to their transient lifestyle. They went from living in shelters, to staying with friends or sleeping rough on the streets or in coffee shops. Housing adequacy Adequate housing was defined as meaning that ?your home has heating, lighting, ventilation, sanitation and washing facilities and does not need major repairs.? By this definition, 63 percent felt that their living condi- tions were adequate, and 34 percent felt they lived in inadequate condi- tions. Living in housing that is in a state of disrepair was given as a pri- mary reason for the housing being inadequate. Other factors included improper heating, ventilation, problems with appliances, and cockroach- es or mice. Informal housing networks While the majority of participants in this sample had made housing ar- rangements with either family or friends before coming to Canada, only 26 percent had sustained ongoing housing help from relatives. Support from family members Of those receiving housing help from relatives, 31 percent were Latin American, or 19 percent of the total Latin American community in this sample, and 69 percent were Muslim, or 32 percent of the Muslim com- munity. The most common form of support was help in finding a place to live. There is a gap in the housing help available from family after the in- itial housing support provided to family members who are newcomers. This is consistent with the findings of Rose et al. (2000) who note that family support systems are not sufficient in providing mid-to long-term support for housing needs. For many participants, other family members may themselves be relative newcomers who do not have the resources needed to provide adequate housing support. Nevertheless, Murdie (1999) notes that in research among Somalis in Toronto, proximity to rel- atives and friends was the most important factor in searching for the first permanent residence. Maintaining bonds with social networks based on family friends and community group is an important means of facilitat- ing integration to a new society, but may not provide the kind of longer- term support needed to negotiate the various systems in Canadian socie- ty, in this case, access to housing. Support from friends and social networks A larger number, 39 percent, of the sample said they received some form of housing help from friends. This is consistent with social network theory, in that social contacts outside the primary groups relationships may provide a broader gateway to housing information and resources. If friendships are cultivated outside the immediate circle of friends and family through community or faith-based organizations, language classes, or workplace relationships, individuals have more diversified opportunities to accessing knowledge related to housing. The findings also relate to the fact that many of the participants are refugees who may rely on support from friends in the absence of family ties in Canada. According to our findings, of those receiving support from friends, 47 percent were from the Latin American community, or 41 percent of the Latin Americans participants, and 54 percent were from the Muslim community, or 30 percent of the Muslim participants. The most common form of support was help in finding a place to live. Mosques as informal housing providers Mosques operate as informal centres for the dissemination of various kinds of information related to housing, education, and employment. Mosques are, however, not formally structured to address housing or employment needs, and have little funding to engage in more formalized systems of support to meet these needs. However, when family networks break down, the mosque becomes a surrogate for support. Living in a mosque or church can be a temporary arrangement. One participant had been living in a mosque for the past two months, as he tried to save money to support himself and a daughter who was chronically ill. Formal agencies providing housing help Relatively few participants went to formal agencies to access support for their housing needs. Most had never approached a formal agency for housing help. This finding is troubling, as it would seem that few partic- ipants have any form of support, either formal or informal, with their housing needs. Participants indicated that ?finding a place to live? was the greatest service that formal agencies could provide, followed by help ?under- standing tenant rights.? This reflects a desire to engage in political self- advocacy by building knowledge of the legal and institutional frame- works that would protect their interests as tenants. While informal social systems can provide some knowledge based on people?s own experience, it is clear that participants are looking for formal support in this area. New strategies for housing help outreach During the period of this study, two housing seminars were conducted in partnership with the Afghan Women?s Organization and the Arab Community Centre to provide public education on housing rights, evic- tion prevention, and access to affordable social housing. Fifty-six at-risk homeless people attended, and four service providers. The housing se- minars were conducted by Syme-Woolner Neighbourhood and Family Centre?s Homeless Outreach Coordinator and the Informal Housing Network Project?s (IHNP) outreach workers. Volunteers from the organ- izations hosting the seminars translated into Arabic and Dari. This public education service was critical in disseminating informa- tion and resources to other organizations that require additional know- ledge on housing to help meet the needs of their clients. Information packages were prepared by Syme-Woolner?s Homeless Outreach Coor- dinator and distributed at the seminars to provide additional support to both clients and service providers. These seminars helped develop ca- pacity among people in the at-risk homeless category as well as service providers, through access to information on housing. An alternative model of service delivery proposed by the IHNP in- volves a decentralized approach, including the development of mobile housing clinics that can provide training and workshops in the areas of housing support identified in the research, as well as housing referral services. These clinics would be made available in various culturally ac- cessible sites. This represents a departure from the current centralized approach to housing help that require people in need of housing support to obtain help from the location of the centre. This strategy also allows for information to be communicated in first languages with the support of community partners on site. This de- centralized process of providing housing support will also considerably strengthen the ability of informal housing networks to provide housing information and support through less formal channels. Selected recommendations The following are some of the key recommendations from our study. ? Undertake greater grassroots capacity-building by coordinating for- mal and informal housing networks to provide temporary housing when necessary and information and resources on housing needs. ? Provide adequate funding and organizational support to ethno- cultural and neighbourhood associations to allow them to participate in partnerships, develop funding proposals, and meet community needs. ? Implement mandatory Employment Equity and Service Equity Leg- islation Implementation of Employment Equity and Service Equity Plans within community, organizational and business practice both in the public and private sectors. ? Ensure that refugee claimants, government-sponsored refugees, and sponsored refugees have rights in relation to meeting basic human needs (e.g., access to financial and settlement services). The process must be culturally accessible and transparent (e.g., people at all the borders get information about the refugee process and settlement workers are on site to give information and any necessary emergen- cy help). ? Ensure that when sponsorship breaks down because of violence, women can still access the same financial and settlement services in tandem with culturally accessible counselling supports. ? Ensure that all newcomers have access to affordable housing, lan- guage support, food, income, employment, and self-defined support. ? Fund programs that provide a process for newcomers to access a guarantor, two-month deposit, information, basic needs allowance, and the start of credit rating. A system must be developed and im- plemented for people to live up to reasonable expectations within their means, for example, interest-free loans. ? Provide pro bono legal support for class-action legal suits against various forms of housing discrimination so that people from margi- nalized communities have support to take on landlords who system- ically exclude tenants on the basis of level or source of income, fami- ly size, race, ethnicity, gender, or mental health problems. ? Train service providers so that they are able to provide more holistic assessments of clients? housing needs based on an integrated anti- racism, anti-Islamophobia and anti-oppression framework, and cul- turally accessible service delivery, and they understand issues of mental health, addiction, and incarceration in the Latin American and Muslim communities and their connection to homelessness. ? Provide funding for holistic housing help in ethno-specific organiza- tions that generally lack the requisite resources to provide adequate housing help along with other linguistic or settlement services. ? Ensure that housing help centres are adequately funded to provide information and support in culturally and linguistically accessible ways for information on support to find a place to live; help main- taining a home; social assistance; shelters and drop-ins; food banks; and tenant rights. ? Develop and provide cross-cultural services in shelters, including provision of halal meals, prayer or mediation rooms as a recognition of the important of spiritual support for those facing housing crises. ? Extend the Reintegration Project delivered by the Ontario Multi-faith Council on spiritual and religious care, which seeks to house inmates discharged from the provincial prisons, to the federal system. ? Provide funding for alternative model of service delivery proposed by the Informal Housing Network Project (IHNP) that involves a de- centralized approach, including the development of mobile housing clinics that can provide training and workshops as well as providing housing referral services. These clinics must be available in cultural- ly accessible sites. ? Provide public education in the Latin American and Muslim com- munities about mental health, addiction (including Harm Reduction vs. Abstinence approach) and incarceration; undertake anti-stigma campaigns in areas that incorporate an anti-oppression framework. ? Develop greater cultural and political literacy among people in mar- ginalized communities to better understand and advocate for hous- ing rights. ? Develop public education strategies for landlords. Encourage more communication between landlords and tenants groups; ensuring that there is continued funding for tenant services, like the Tenant Hot- line and the Federation of Metro Tenants ? all those bodies that pro- vide legal representation and information to tenants. ? Educate refugee claimants about the Canadian political system (e.g., income support programs). Encourage people to assert their political rights and contact their local politicians about housing concerns, identifying themselves as future voters. Jasmin Zine is an Associate Professor of Sociology at Wilfrid Laurier Universi- ty. Her research interests include critical race and ethnic studies, postcolonial theory and research methodologies, education and social justice, representation, Canadian Muslim studies, and Muslim women's studies . This paper was origi- nally presented at the Metropolis International Conference, University of Gene- va, Switzerland, in 2004. Acknowledgements This project was funded by the Supporting Communities Partnership Initiative (SCPI) and sponsored by the Informal Housing Network Project a partnership of the York Hispanic Centre, Islamic Social Services and Resources Association (ISSRA), Syme-Woolner Neighbourhood and Family Centre and the Community Resources Consultants of Toronto (CRCT). The author also acknowledges the support provided by Dana Milne, Sharon Allen, Lorrie Simunovic, Rosamaria Andino, Hasrat Ga- foor, and Robertina Saavedra. References Bakshi, L., Rooney, R., & O?Neill, K. (1999). Reducing stigma about mental illness in transcultural settings: A guide. Melbourne: Australian Transcultural Mental Health network, Department of Psychiatry, University of Melbourne. Canada Mortgage and Housing Corporation (2001). Rental market survey: Fact sheet. Toronto: City of Toronto. Chambron, A. S, Hulchanski, D. J., Murdie, R. A., Teixeira, C. (1997, May). Access to housing in a Canadian city: Experiences of three immigrant groups. Urban Af- fairs Association Conference, Toronto. Dei, G. J. S. (1996). Anti-Racism education: Theory and practice. Halifax: Fernwood Press. Dion, K. L. (2001). Immigrant perceptions of housing discrimination in Toronto: The Housing New Canadians Project. Journal of Social Issues, 57(3), 523-539. Dunn, S., Pottie, K., & Mazzeo, M. (2000). Central American immigrant men and mental health: Problemas con tus nervios? Q ue puedes hacer? Problems w ith your nerves? What can you do? . Toronto: York Community Services. Gillespie, K. (2002, June 12). Vacancy rates more than doubles. Toronto Star, p. B1. Golden, A., Currie, W. H., Greaves, E., & Latimer, J. E. (1999). Taking responsibility for homelessness. City of Toronto: Mayor?s Task Force on Homelessness. Kappel, B., & Ramji, Z. (2002). Common occurrence: The impact of homelessness on women?s health. Toronto: Sistering: A Woman?s Place. Murdie, R. A. (1999). The housing careers of Polish and Somali newcomers in Toronto?s rental market. Toronto: York University, Department of Geography. Murdie, R. A., Chambron, A. S., Hulchanski, D. J., & Teixeira, C. (1996). Housing issues facing immigrants and refugees in Greater Toronto: Initial findings from the Jamaican, Polish and Somali communities. In E. M. Komut (Ed.), Housing question of the ? others? (pp. 179-190). Ankara: Chamber of Architects of Turkey. Ornstein, M. (2000). Ethnoracial inequality in the City of Toronto: Analysis of the 1996 census. Toronto: Access and Equity Unit, Strategic and Corporate Policy Di- vision, Chief Administrator?s Office. Refugee Housing Task Group (2001, October 16). Meeting minutes. Toronto.. Rose, D., Carrasco, P., & Charbonneau, J. (2000). The role of ? weak ties? in the set- tlement experiences of immigrant women with young children: The case of Central Americans in Montreal. Toronto: Centre for Excellence in Research on Immi- gration and Settlement. Siddiqui, H. (2001, April 15). Census draws portrait of multi-hued country. To- ronto Star, p. A1. Toronto Police Services (2001). Annual hate/bias crime statistical report. Toronto: Author. Wallace, M., & Frisken, F. (2000). City sub-urban differences in government responses to immigration in the Greater Toronto Area (Research paper 197). Toronto: Centre for Urban and Community Studies, University of Toronto. Wolfe, J. (2002). Situational assessment: Political, economic, social-demographic, tech- nological and legal context for community mental health in Toronto: Best practices. Toronto: Community Rehabilitation Services, Community Resources Con- sultants of Toronto. Chapter 5.2 Taking Care of Their Own? Or Falling B etween the Cracks? Absolute and Relative Homelessness Among Immigrants, Refugees, and Refugee Claimants in Vancouver DANIEL HIEBERT, SILVIA D?ADDARI O , AND KATHY SHERRELL In July 2002, the Greater Vancouver Regional District (GVRD) released its report on Homelessness in Greater Vancouver. The findings were alarm- ing: the population at risk of (economic) homelessness in Greater Van- couver increased dramatically between 1991 and 1996. This increase may be attributed, in part, to worsening conditions for both renters and own- ers. The period between 1991 and 1996 was characterized by increasing property values and rental rates, low vacancy rates, and a decline in (real) household incomes. The report also revealed that immigrant populations are dispropor- tionately affected by changes in the housing market. For newcomers, the challenges faced by Canadian-born population in accessing affordable housing are compounded by both economic and social barriers. In light of the declining fortunes of immigrant households over the last decade, these barriers are insurmountable for some. We undertook this study to examine the extent and profile of abso- lute and relative homelessness among immigrants, refugees, and refugee claimants in the GVRD. There is little systematic knowledge on this topic, partly because marginalized populations are poorly recorded in key data sources. Basic social surveys, such as the census, do not neces- sarily include all groups. Members of some groups, including Aboriginal people, may refuse to acknowledge the census. Others, including those without shelter, can easily fall below the notice of census enumerators. In conducting this study, we sought to: 1. generate basic knowledge, and if possible a realistic estimate, of the number of immigrants, refugees, and refugee claimants experiencing relative or absolute homelessness in the Greater Vancouver Regional District (GVRD); 2. understand the degree to which these communities provide in-group assistance to homeless individuals and families; 3. understand the ways that service organizations (NGOs) provide as- sistance to homeless individuals and families. We defined homelessness broadly as a spectrum of conditions that ranges across the following categories: ? Housing stress: households spending more than 30 percent of their income on shelter. ? Critical housing stress: households spending more than 50 percent of their income on shelter. ? Relative homelessness: individuals or households who are in tempo- rary accommodations, such as ?sofa surfing? or ?camping out? with family members or friends. ? Absolute homelessness: individuals or households who are in shel- ters or, worse, living without shelter. Method In light of the complexities in defining and enumerating homelessness, we adopted an evidence-based, multiple-points-of-contact study design, combining both qualitative and quantitative methods. The project was composed of three sub-studies, each of which highlights a particular as- pect of homelessness. The first sub-study focused on the immigrant and refugee popula- tions using emergency shelters and transition houses. This sub-study involved 12 semi-structured interviews with key informants from emer- gency shelters and second-stage transition houses in the GVRD; and the compilation and analysis of data collected by shelter personnel over seven 24-hour periods between October and December 2004. In total, we received 261 completed shelter data collection forms. The second sub-study explored the housing situation of refugee claimants who have recently received a positive decision enabling them to stay in Canada. Thirty-six individual interviews were conducted with successful refugee claimants (SRCs) in the GVRD. The interviews were semi-structured and explored the housing situation of claimants both before learning of the positive decision, and in the first six months since learning of it. In addition, four interviews were conducted with settle- ment workers. In this summary chapter, we focus particularly on the results of our interviews, since they include the voices of refugees, and provide startling insight into their experiences since arriving in Canada. The third sub-study examined the profile and extent of relative homelessness among immigrants, refugees, and refugee claimants. We hoped to generate a basic estimate of the ?sofa surfing? or ?camping out? population among recent immigrants, as well as to identify in- group systems of support through questions about the provision or re- ceipt of housing assistance. This sub-study is mainly focused on the Im- migrant and Refugee Housing Survey (IRHS), conducted October 4-8, 2004. In total, we received 554 completed surveys. The shelter study: Where immigrants are a small minority The shelter survey conducted as part of this project reveals a sample population that is overwhelmingly English-speaking (91 percent) and Canadian-born (82 percent). Although immigrants and refugees form 38 percent of the population in Greater Vancouver, they account for only 18 percent of our respondents, and probably an even lower proportion of the total shelter population. It was suggested that when facing a lack of secure housing, mem- bers of established ethnocultural groups stay with family or other ac- quaintances, instead of relying on emergency shelters. For those lacking secure accommodations, these networks may be tapped to provide tem- porary accommodations. While established ethno-cultural communities may have the ability to ?take care of their own,? other groups who lack extensive social networks, including recently arrived individuals and refugee claimants, may fall through the cracks. We also learned that almost one-quarter of the individuals staying in shelters reported some form of employment, either full-time, part-time or casual. While some respondents in our shelter survey describe their last long-term housing as unstable (e.g., living in single-room occupancy ho- tels, staying with friends or family, or renting a friend?s couch), others reported having owned their last form of long-term housing. Key infor- mants spoke of a cycle of homelessness in which people move between insecure housing and ?rooflessness? on a regular basis. Low social assis- tance and shelter allowance rates, combined with institutional rules that limit shelter stays to no more than 30 days, produce a cycle in which whatever housing that is obtained is temporary. In the rush to obtain housing, and with limited means, clients settle in unsafe housing located in marginal and inexpensive areas of the city, perpetuating a cycle of social marginality and homelessness. Often, key informants report, these clients return to the shelter system within six months to a year. The most frequently cited causes of homelessness were financial (e.g., job loss, eviction); substance abuse (e.g., drugs, alcohol); mental health issues; family breakdown; and physical or emotional abuse. When we combined all the immigrant categories (refugee, immigrant, perma- nent resident, and citizen), the three most frequently cited responses for homelessness are physical or emotional abuse, family issues, and mental health issues; while for non-immigrants, the three most frequently cited are financial crisis, substance abuse, and mental health issues. In sum, we learned from the shelter survey that few immigrants and refugees appear to use the shelter system. We also found that despite being homeless, many respondents reported having some form of em- ployment. Finally, shelter personnel expressed concern about the struc- tural barriers that affect all clients, including limited shelter and trans- portation allowances, as well as the time limits on stays (in light of current waiting lists). Interview findings: Unique obstacles and social isolation Income, rent, and making ends meet Until they obtain employment, successful refugee claimants depend on basic welfare provisions. All but three respondents in this study were dependant on welfare alone during the initial stages of settlement. Rental rates in Vancouver have been increasing, while the basic welfare allow- ance has not, leading to a critical affordability problem. Unless people share accommodations, refugee claimants and their families can rarely, if ever, afford larger units. During the initial settlement stages, 32 out of 36 refugee claimants found themselves spending between 50 and 74 percent of their income on housing in the initial settlement phase, and 4 respon- dents spent more than 75 percent. We heard many stories about particularly difficult circumstances. For example, Ali arrived in Canada by himself in 2003 from Afghanistan. At the airport, Ali?s appendix burst and he was rushed to the hospital and underwent emergency surgery. Without any knowledge of the medical system or social support, Ali was charged medical fees and was required to take antibiotics after his surgery which cost him well over $200 a month. After one month of staying in Canada, Ali received his first welfare check totalling $500. He found housing with a friend an d they each paid $350 a month in rent. Ali?s medical expenses exceeded the $150 that he had left over after paying his share of rent. In addition, Ali spoke of shrapnel wounds that he had endured during the war. He said that for months he required medical attention that he just could not af- ford. With no money left over, Ali said that he went hungry and thirsty for months. He stated that he had fainted several times owing to starva- tion and dehydration. Finding work Gaining entrance into Vancouver?s labour market has been a trying and emotionally difficult experience for many interviewees. While some de- cided to wait and upgrade their language and education, those who sought work were stymied. The greatest barrier and point of frustration for those who had searched for work has been employers? expectation of Canadian work experience. For example, Paulo, a 51-year-old man from Mexico, had been a very successful businessman in his country and had worked in film pro- duction and within the media for over 25 years. In addition, he had taught related subject matter at a university. Despite his training, skills, and excellent English, Paulo was unable to attain work after his arrival here in 2000. He said that he was discriminated against because he did not have Canadian experience in his professional field. However, when Paulo attempted to look for low-skilled work in the local caf?s, he re- ported that he was then discriminated against for being too old and overqualified. Paulo spent years volunteering in a variety of places in order to attain Canadian experience. ?But now I can?t be a volunteer anymore, I need money. ? I don?t have time to be a volunteer. I need a job, a real job. So the housing could be a problem for me and my future.? Paulo was desperate and felt he had nowhere to turn. Without employ- ment income, his housing situation is very unstable. While many have had to deal with a labour market that does not recognize foreign experience or skills, some have accepted downward occupational mobility in order to attain employment. Of the 16 claimants who possess a postsecondary degree, and who had attempted to attain employment, none were able to utilize their education, either in a practi- cal application or in order to obtain a skilled job. Language barriers Lack of fluency in English also proved to be a significant barrier inhibit- ing access to adequate housing for refugee claimants. Only 5 of the 36 participants had arrived in Canada with fluent skills in English. Ten claimants spoke about the issue of discrimination spontane- ously. Of particular concern were issues of social insurance tagging, wel- fare discrimination, and a widespread reluctance of many landlords to rent to households with children. All the respondents reported that there was little, if any, support of- fered upon entry to Canada and in the subsequent days after arrival. Claimants said they experienced feelings of confusion and fear upon ar- rival and in all cases, they were left on their own. A young woman from Albania noted, ?[The Immigration officers] make it clear to you that this is what their responsibilities are and where they end and so you are on your own from that point on.? Five respondents arrived at the airport in Vancouver with French- language skills. These claimants were surprised that, upon arrival in a bilingual, French-English speaking country, there was no one who could communicate with them. The participants said that there was no one at the airport who spoke French, and that this made the refugee claim process much more difficult. A 25-year old male from Togo told us, ?When I came here I could hardly speak any English... [the language] was difficult. The worker started to ask me a bunch of question and I can?t speak English. I tried to speak, I asked if they speak French, and they didn?t know. I got so angry.? Arrival Participants were also asked, ?Where did you stay the first night you arrived in Canada?? For several claimants, the first night ? and in some cases the first several nights ? were spent at the airport. Some were de- tained for lack of documentation, while others spent the first night at the airport because they had no other place to go to and were unable to find proper accommodations that first night. For example, one respondent from Nigeria arrived in Vancouver in 2003. She was eight months preg- nant and accompanied by two children, aged one and four. She recalled her experience with the immigration officers. ?They said that I had to go? I said, ?Where do you want me to go?? [The officer] said, ?An y- where.? ? [I s aid] ?I don?t know anywhere? you have to tell me.? I [asked] if I can sleep on the floor. She said y es. So I slept on the floor? I am pregnant.? Another woman from Congo had a similar experience. On her first night in Vancouver, eight months pregnant, she said, ?I had to sleep on the chair, because I don?t know where I am.? When discussing her first few days in Canada, one woman from Albania expressed her frustration over the lack of support during such a trying time. ?There are too many processes going on at the same time: you have find a house, you have report to immigration, you have to find a lawyer, you have to do your welfare papers, you have to go do immi- gration exam ? and then you have to go apply for the work permit and then you have to go apply for social insurance number and then you have to go and apply for a job, and then your hearing comes and ... it?s too many things to do at the same time ... And you only have 20 days to do everything and what if you don?t have your lawyer at that time? And what if the lawyer asks for too much money and you don?t have the time to collect all of that money?? The initial settlement period was trying for all of the participants. A single mother from Sri Lanka expressed the gender differences and cul- tural considerations that are sometimes overlooked. She stated, ?Guys, they can go around and get the information, but ladies, in our country, we are taught that it is scary and especially because we don?t know the language and we cannot trust anybody, so we cannot find the informa- tion right away.? One settlement worker reflected on the capabilities of settlement agencies to provide initial information on housing to claimants. He stated, ?Unfortunately, the situation for a lot of settlement workers is that we don?t have that many resources to offer in terms of housing. We can?t say to our clients, ? By the way, there is this specific way where you go to get all of the information and they will help and give you assis- tance and inform you about housing and where to go.? There is no such thing. [Housing] is an area that the settlement sector has not put that much attention to, and it?s the key thing from the beginning.? Social networks Only one participant had arrived to Canada with a family member who was already established here in Vancouver. All 35 claimants arrived without any pre-existing social networks. Settling in a new country without any social support can make the housing situation for newcom- ers even more difficult. One settlement counsellor stated, ?If they are very honest and tell them that they are a refugee claimant, then most probably the landlord won?t rent a place to them first. They don?t know much about refugee claimants. In their mind it?s always someone very desperate, no job, maybe experienced violence in their home country or their personality is unknown and also they don?t have networks here, so if anything happens they have no other sources to help these tenants. Stigmatization is very serious.? Although the claimants did not have anyone to assist them in the first few days after arrival, some claimants did manage to tap into ?eth- nic resources.? One settlement counsellor noted, ?[ Claimants] will turn to people that seem familiar to them. Familiarity. If they speak their lan- guage then they will approach them ... people who look like their group ... they are looking for a face or words that will lead them to a place.? A 29-year old female from Sri Lanka recounted that she felt the saf- est approaching someone from her own ethnic group. ?On bus I met some Sri Lankan Singhalese lady, my language. She said, ? Do you know about Inland Refugee Society, they help refugees. Go and talk to them?? Then I go and I try to find them, but it was difficult. We don?t know any information, especially BC housing, we don?t know anything.? A 32-year old man from Cameroon was able to find housing by networking with other refugees and African migrants. ?I met this friend from Liberia. Then I spoke to him that I was looking for accommoda- tions. In fact I was with one African guy that just came at the same time. So we were both looking for accommodation. So we happen to meet this guy who is from Liberia, then that?s when he invited me to meet [a set- tlement worker] at church with the possibility of how I can get acco m- modation.? Another settlement worker credited the settlement of claimants to their creative survival methods. He said, ?They have been so creative, they develop these kinds of networks amongst themselves ... they start talking about living in such a place and they know the landlord now ... so it helps, but it also brings its own problems because people end up being in places that aren?t necessarily the best.? While discussing his observations in dealing with refugee claim- ants, another settlement worker asserted, ?The more supported a refugee claimant is, not only with housing, but with relationships ... they have the support they need to pull it off and they settle in more quickly, gen- erally find jobs more quickly. Refugee claimants are totally discon- nected.? Living conditions Inadequate and substandard living conditions constitute a major com- ponent of relative homeless. Although all of the claimants reported hav- ing a space in which to live, their dwellings were often of low quality. One interviewee spent many of his nights sleeping on the floor or on old mattresses that had been discarded on the street. In the second place that he stayed, he recounted, ?Until that time we didn?t have blankets, we were sleeping on the floor ... we didn?t have anything until four or five months we were sleeping on the floor. We had no pillows, no mattress ... nothing. We didn?t even have cups, we were drinking water out of our hands.? A prominent housing outcome for many refugee claimants was overcrowding. A number of problems led to ?doubling up? strategies in order to access rental units in Vancouver. Vacancy rates for larger ac- commodations are higher, but so are prices. Therefore, newcomers are forced to seek out smaller and more affordable accommodations. This quotation is representative of many: ?It was just a one bedroom and it was very hard for us; my son needs a bedroom and also me and my hus- band need one bedroom ... for our culture it is very important for our child. My son got the bedroom and me and my husband sleep in the liv- ing room. It was really, really, really hard time.? A male claimant from Sri Lanka has lived in six units since his arri- val, and he is currently searching for his seventh. He commented on the crowded conditions of a one-bedroom suite that he lived in for six months. ?The whole house was filled with beds, like two beds in the room and one bed on the outside.? In addition to crowding, claimants also mentioned substandard conditions, describing much of these places as dirty, smelly, and requiring upgrades. Owing to financial constraints, claimants are settling in areas that have low rental rates or are compromised in quality. Alongside low- income and cheap housing complexes are also crime, drug abuse and prostitution. This issue can be seen in the experience of a family from Nigeria. Nia was eight months pregnant and had two small children with her. Her husband Joseph was not able to reunite with them until several months later. Nia and the children spent their first month in an emer- gency shelter. She said that, ?The [emergency shelter] was very dirty and there was a lot of smoke, marijuana ... the smoke come inside, it was mouldy and the kids were getting a lot of problems.? Nia was very con- cerned about the health and safety of her children while living at the shelter. Once Joseph arrived in Canada, the family was able to move into BC housing in the Downtown Eastside. The move out of the emergency shelter and into more stable housing did not increase feelings of safety for this family. Both parents are still quite concerned for the wellbeing of their children. Joseph said, ?We are so much worried about our own children. When we take them out, there are the drunks and drug ad- dicted? it is not a good place.? In response to concern for their chil- dren?s safety, the couple applied for a transfer. They reported being told that a transfer is only possible after having lived in the housing complex for more than one year. A 43-year-old female claimant from Russia commented, ?Safety is always jeopardized, especially for the refugee claimant. Canadian, well- educated intelligent women tried to use me as a free housecleaner, baby- sitter ... they don?t care; [she says] I kick you from the country. And guys are always looking for how to use women. And no one cares.? A place in Chinatown The refugee claimants who have come from China told us an important and unique story. While many claimants discuss one or a combination of the above factors (crowding, substandard conditions, and safety), the claimants from China speak of a housing situation that includes all of these factors in an alarming combination. Six of the seven claimants from China found their first accommodations in Chinatown, and all six still reside in this area of the Downtown Eastside. As newcomers to the country, they arrived alone, without any fi- nancial resources or English skills. Without any knowledge of the hous- ing market, all six found themselves wandering the streets. Several Chi- nese claimants relied on the advice of strangers, which led them to seek accommodations in Chinatown. The respondents said that once they ar- rived in Chinatown, they were all able to find somewhere to sleep; all five claimants found themselves in similar rundown accommodations that are geared towards newcomers from China. The interpreter/settlement worker acknowledged one specific hotel as the same place that nearly all of her refugee clients from China find themselves. According to these six participants, the conditions here were nothing short of horrendous. A male claimant aged 49 from China gave these details, ?Things there are in a mess ... there were cockroaches eve- rywhere. But the rent was cheap. There were a lot of seniors living there; they are dirty and have a lot of personal belongings, so things are in a mess. A lot of cockroaches. Dirty, stinky.? The description of crowding varied slightly between respondents, but the image remained the same. Four of the claimants noted how this site for Chinese refugees allots one washroom and a small kitchen area for 20 to 30 people. Electricity and heating work sporadically at best. Each participant detailed the same list of unhealthy and unsanitary con- ditions, which include dirty, smelly, infested rooms. In the Downtown Eastside, safety is also a major factor. A female claimant from China, age 65, became very emotional during the inter- view as she discussed her first reaction to living in Chinatown. ?First, it?s very noisy. Second there is drug trading inside the hotel and some peo- ple using drugs and there is different mixture of people living there like refugee claimants, very low-income people, or long-term residents and there is a gambling room for people to go gambling.? This woman stated how unsafe she felt in an environment where there was rampant drug use and dealing as well as illegal gambling. She still resides in Chinatown five years after her arrival. All six of the claim- ants who have lived or are still living in Chinatown felt compelled to reside in this precarious environment primarily because they cannot see any other option. Without the ability to speak English, and without in- formation or how to learn English, these claimants felt that there was no other way to communicate with others or even perform basic tasks such as shopping for groceries, unless they were in a culturally homogenous environment. Chinatown provided a familiar environment during a very daunting and unstable period. Financial constraints are also a key factor in the clustering of claimants in Chinatown. The participants all quoted the same rental rate of $325 per month and all were restricted to the wel- fare allowance of $510 per month. Spending more than 50 percent of their income on housing, these five were all in critical housing need. One settlement counsellor who works with Chinese claimants was asked about the settlement patterns of her clients. She discussed China- town as a likely starting point for claimants who are new to Vancouver. She stated, ?These refugee claimants don?t have many resources to look for other places, and? the lan dlords don?t like to rent a place to refugee claimants, so they are stuck in a hotel in Chinatown and the living condi- tion is very bad. I heard from my clients that there are mice, and people break in and steal their stuff and also the facilities, shared kitchen and shared bathroom and very it?s noisy and also people are gambling ... so the whole environment is not very healthy.? The vulnerability associated with refugee status, as well as the bar- riers faced by all immigrants, results in a high degree of homelessness in one form or another. The situation is more extreme for claimants, who face deeper levels of deprivation than the average immigrant; most claimants depend on welfare rates that are far below the poverty line. In sum, we learned that refugee claimants face unique obstacles that are symptomatic of their immigration class. By virtue of their means of entry, claimants have a greater disadvantage in the housing and labour markets. Furthermore, there is a discrepancy between what refugee claimants are receiving in basic aid, and the average cost of renting an apartment in Vancouver. In addition, the vacancy rates for smaller, more affordable accommodations are particularly low. Inadequate and sub- standard living conditions, overcrowding and safety concerns represent three major components of relative homeless among SRCs. Finally, claimants tend to be socially isolated. The minimal levels of financial, documented human, and social capital of individuals in this group is associated with extreme vulnerability to homelessness. The situation is quite different for those immigrants and refugees who have access to social networks and support systems. Survey fin dings: Giving and receiving assistance The IRHS was intended to be a representative sample of all clients who sought the services of immigrant and refugee-serving agencies on either a phone-in or an in-person basis during the week of October 4-8, 2004. Sixty-four percent of our respondents were female. Although respon- dents came from 61 countries, 52 percent of the respondents were born in 4 countries: China (19.1 percent), India (13.2 percent), South Korea (10.5 percent), and Iran (9.2 percent). Many of the respondents who have arrived within one year are lo- cated outside the traditional immigrant receiving areas (the east side of the City of Vancouver). Thirty percent of respondents reported their status as being Cana- dian citizens, 60 percent are Permanent Residents (they arrived as either Economic Immigrants or through the Family Class), 5 percent are Gov- ernment Assisted Refugees (GARs) and 5 percent are refugee claimants. Those who arrived as immigrants were more likely to be geographically dispersed than those who arrived as refugees/claimants. Although 90 percent of survey respondents report having some form of housing, 7.5 percent were living in what could be considered temporary or unstable living conditions (e.g., staying with friends or family, living in single-room-occupancy hotels, or emergency shelters) over the three-month period directly preceding the survey. Twenty-eight percent of respondents in the IRHS are receiving some form of help with housing, while 15 percent of the respondents who are not receiving help indicate that they have provided assistance to someone other than their parents or children. Almost one-quarter of those receiving help were staying with friends and family. Almost half (44 percent) of those who are receiving help have been in Vancouver four years or longer. We learned from the survey that the socio-economic profile of re- spondents who are providing assistance does not differ significantly from those who are receiving assistance. Respondents in both groups reported high numbers of people who officially landed in Canada within the last three years, as well as living in households with no one em- ployed. Moreover, those who are providing help often do so despite liv- ing in precarious situations. Over 61 percent of those providing assis- tance, for example, are in core housing need, while 25.6 percent are in critical housing stress. We also learned that there is a disconnect between the length of time people expect they will need assistance, and the length of time peo- ple have provided assistance. Despite the perceived need for help over long periods, those who have provided assistance report having done so for relatively short periods of time. Finally, there is a lack of similarity between type of help received and type of help provided. Those provid- ing assistance generally help their guests obtain housing, while those receiving assistance say that they receive help paying the rent. This point reveals the fact that our sample included few immigrants who are rela- tively well off. Overall conclusions: Hidden homelessness and social capital The phrase that has come to represent our understanding of the situation is ?hidden homelessness.? This is another concept around which all three sub-components of the study converge. The SRCs in our sample tell a fairly consistent story of their first encounter with Canadian society as confusing and full of anxiety. Their initial housing experience was typically in the cheapest accommodations available, in poor residential environments. They coped by sharing rents and crowding. Nearly all continue to be dependent on social assistance and nearly all are in situa- tions of housing stress. But they are not ?on the streets,? in large part because of their coping strategies and ? in number of cases ? help ex- tended from social organizations or other members of their ethnocultural community. The existence of bottom-up self-help was even more apparent in the survey of clients of settlement agencies. In this part of the project we found a significant sharing of resources that mainly occurs within famil- ial networks and ethno-cultural or religious communities. About 15 per- cent of those using settlement services are receiving some form of hous- ing assistance, which ranges from help locating housing, through financial help, to the provision of housing (often temporary, but occa- sionally long-term). Nearly all of this activity occurs ?below the radar? of the Canadian welfare state. Those who are helped, in essence, are able to avoid the services of homeless shelters. Significantly, even those who are living in precarious circumstances extend whatever help they can to oth- ers in their close networks. The findings of this project also highlight the fact that the housing situation of newcomers to Greater Vancouver is heavily influenced by the social capital of existing ethno-cultural communities. As a result, the extent of relative and absolute homelessness among immigrants, refu- gees, and refugee claimants is less than would be expected, given the income levels of these groups. This is not to say that the delineated groups are well housed. In- deed, many individuals and families are living in crowded, substandard conditions. However, the social networks operating among immigrant, refugee, and refugee claimant communities appear to mitigate the worst forms of homelessness, and the groups of people we studied are actually underrepresented in the population using homeless shelters. Our findings lead to several conclusions. First, our study suggests that current levels of social/shelter assistance are exceedingly low, espe- cially in light of the lack of affordable housing. When clients settle in un- safe housing, in inexpensive and marginal areas of the city, they tend to enter a cycle of homelessness, needing help from others. Second, help is available. The positive side of the story is the extent to which mutual aid is provided. This is a clear example of what is vari- ously labelled ?ethnic resources? or ?social capital? in the academic and policy literatures. But systems of reciprocity do not include everyone ? which is our third basic finding. Refugee claimants, given the combination of their uncertain legal status, lack of language facility, and lack of familiarity with Canadian society, are the most likely of all newcomers to ?fall be- tween the cracks? of both ethnocultural communities and the welfare and housing provisions of the state. Fourth, we re-emphasize the phrase ?hidden homelessness.? Immi- grants, refugees, and refugee claimants appear to be particularly suscep- tible to relative homelessness, so their difficulties in the housing market are essentially invisible. Fifth, as we increasingly come to understand the fact that homeless- ness is a spectrum of conditions, rather than a single absolute state, it is logical that there also needs to be a spectrum of policy responses to homelessness. Daniel Hiebert is Professor of Geography, University of British Columbia. Silvia D?Addario is a PhD candidate in Geography at York University. Kathy Sherrell is a PhD candidate in Geography at the University of British Colum- bia, and also Manager of Community Settlement Programs, at the Immigrant Services Society of British Columbia. Based on ?The Profile of Absolute and Relative Homelessness Among Imm i- grants, Refugees, and Refugee Claimants in the GVRD: Final Report,? prepared for the National Secretariat on Homelessness. The principal investigator on the project was Sherman Chan. Chapter 5.3 At Risk in Canada?s Outer Suburbs : A Pilot Study of Immigrants and Homelessness in York Region VALERIE PRESTON , ROBERT MURDIE, JANE WEDLOCK , MIN JUNG KWAK, SILVIA D?ADDARIO , JENNIFER LOGAN , ANN MARIE MURNAGHAN, SANDEEP AGRAWAL, & UZ O ANUCHA The suburbanization of immigrant settlement means that immigrants must look for housing in submarkets that offer mainly single-family, de- tached, owner-occupied housing. In York Region, rental housing and social housing are especially scarce, as are social services, including set- tlement services to help newcomers secure affordable, adequate housing. Immigrants in York Region may face precarious and overcrowded hous- ing arrangements, as households double up, move into unlicensed rental units, or try to maintain homeownership without a stable income stream. The resulting stress may delay successful settlement. In York Region, a region known for its affluence, homelessness among immigrants is a hidden issue, as immigrants double up with friends and family and tend not to use the shelter system. Alternatively, some immigrants in York Region are forced to take on the burden of homeownership before the household has a stable and adequate income. These households cling to their housing, but the face increased financial and person stress. These findings represent a challenge to the conventional assump- tion that homeownership is evidence of a progressive housing career, in which the quality, size, and location of housing improve over time. The findings also underscore the variable geography of homeless- ness within a single metropolitan area. Information about housing issues at the level of the metropolitan area must be disaggregated if we are to understand the dynamics of homelessness in the suburbs where popula- tion and employment growth are concentrated. Methods This research draws on both primary and secondary sources: ? a comprehensive review of policy documents and academic litera- ture about immigrants? housing circumstances and their vulnerabili- ty to homelessness; ? an analysis of 2001 and 2006 census information on household com- position of immigrants in York Region, residential tenure, housing costs, and affordability; ? 13 key informant interviews with key actors in the immigrant set- tlement, housing, and social service sectors; ? 9 focus groups with 62 immigrants in the area experiencing chal- lenges finding and keeping suitable, affordable, and adequate hous- ing. York Region: an affluent suburb York Region is one of Canada?s fastest-growing suburban areas and an important destination for immigrants. Its current population is about 1,000,000. Between 2001 and 2006, the population of York grew by 22.4 percent, largely as a result of immigration. In 2006, approximately 43 percent of the region?s population was foreign-born. In some municipalities within the region, such as Mark- ham, the proportion of foreign-born residents is higher than 50 percent. The housing market in York Region is dominated by owner- occupied, single-family houses. Single-detached dwellings account for 72 percent of the housing stock, semi-detached and row houses make up 16 percent of the stock, and the remaining 12 percent are apartments. More than 86 percent of all dwelling units in York Region are owner-occupied. Housing is expensive in York Region. In June 2005, the average sales price for a single detached dwelling in York Region was $477,000. Within the rental stock, rents are very close to those in the City of Toron- to. The region has approximately 11,000 units of subsidized housing, made up of municipal housing, non-profit housing, and cooperative housing and housing where tenants pay only 30 percent of total income for housing. r capita for social services in York Region is lower than in any other part of the metropolitan area. To put the numbers in perspective, it may be helpful to compare York Region to Peel Region, the suburban region to the west of Toronto, which has a slightly larger population (1.16 million). Single, detached housing is cheaper in Peel, rental units account for about 20 percent of the housing stock, there are more than twice as many social housing units, and more than twice as many households receive housing subsi- dies. Immigrants need affordable housing Affordable housing is a prerequisite for successful settlement. Suitable and affordable housing in a comfortable neighbourhood represents a base from which immigrants can address the challenges of settling suc- cessfully in Canada. There is, however, growing evidence that newcomers are more like- ly than their Canadian-born counterparts to suffer housing affordability problems. And although affordability problems tend to decrease over time for immigrants, data from the Longitudinal Survey of Immigrants to Canada show that housing affordability persists as an important issue for almost four out of ten immigrant households after four years of resi- dence in Canada. Affordability problems also differ according to ethnic origins. A To- ronto study found that the proportion of Latin American, Arab, West Asian, and Black immigrant households spending at least 30 percent of income on housing are higher than for the population as a whole. By contrast, the percentages of Chinese, South Asian, and Italian immi- grants that spend 30 percent or more of their incomes on housing are below the metropolitan percentage (Hiebert et al., 2006). The hidden homelessness of immigrants Homelessness describes a continuum that ranges from those who are living in shelters or on the street, to the hidden homeless who stay with friends and family, to those who are vulnerable to homelessness because they spend a high proportion of their income on shelter. Very few immigrants are living rough, and few use shelters or so- cial services. Immigrants are more likely to be hidden homeless. They stay with friends and family, sometimes sending their children to differ- ent households to reduce the inconvenience for the families with whom they stay. Even this option may be difficult, as many immigrants are iso- lated with small social networks of friends and family members. What we can learn from the census An analysis of census data from 2001 and 2006 indicates three important characteristics of immigrants in York Region. First, in York Region, recent immigrants are more likely to live as couples with children than people born in Canada and less likely to live alone or in lone-parent families. The preponderance of couples with children among recent immigrants is a direct result of immigration selec- tion policies that favour people aged between 25 and 45 ? the prime years for child-rearing. Second, immigrant households in York Region are more likely to be multiple-family households than Canadian-born households. In 2006, 31 percent of recent immigrants? households included more than one household. Many immigrant households double up so they can afford to pay the rent or make homeownership affordable. Third (and this point can be inferred from the other two), average household size is substantially larger for recent immigrants than for the Canadian-born residents in York Region. Households headed by a per- son born in Canada have an average of 2.9 persons, while the average household size for households headed by immigrants who arrived be- tween 1996 and 2001 is 3.7 people. The large size of immigrant households and the predominance of couples and multifamily households in York Region represent a housing challenge. Housing must be large enough to accommodate large families. Yet large dwelling units tend to be expensive, often costing more than many immigrants can afford, particularly during the initial years of set- tlement. Large households in a region of homeowners Given the housing stock available in York Region, homeownership is of necessity the predominant form of housing tenure for immigrants. In any case, the monthly costs of renting are much closer to the monthly cost of ownership in York Region than they are in the City of Toronto, so immi- grants have an extra reason to own. At the same time, with lower household incomes than their Cana- dian-born neighbours, higher proportions of immigrant households spend at least 30 percent of their incomes on housing than the Canadian- born. In some cases, the proportion is as high as 50 percent. In 2006, while slightly more than one-quarter of all households in York Region were spending at least 30 percent of monthly income on housing, this is true of almost one-third of all immigrants and close to half of all recent immigrants. These numbers are higher than the corres- ponding numbers for the City of Toronto. The situation varies according to the municipality within York Re- gion. It is most acute in six census tracts in York Region (in southwestern Markham, and central Richmond Hill), where 26 percent to 34 percent of immigrant households are at risk of homelessness because they are spending more than half the household income on housing. Behind the numbers: the effects of housing shortages Our interviews with non-profit and municipal housing managers, as well as representatives of settlement agencies, non-profit agencies pro- viding housing support, and social agencies serving homeless and mar- ginalized groups in York Region shed light on the lived experiences of immigrants in the area. The interviewees confirmed the lack of affordable housing in York Region and pointed to two typical outcomes of the gap between availa- ble housing and the low incomes of immigrants. First, many recent immigrants live in unlicensed secondary suites, usually basement apartments. These units seldom satisfy the require- ments of the building code, and many are unsafe, badly maintained, or in poor condition. Secondary suites in some neighbourhoods are iso- lated, located far from routes served by public transit. Second, the shortage of rental housing often contributes to over- crowding that increases the risk of homelessness. Overcrowding occurs when a large family tries to fit into a small apartment, when families sub- let space so that they can afford mortgage or rent, or when families try to help each other by sharing accommodation with relatives. Overcrowding can lead to family stress and conflict. The worst-off are often seniors, who have the fewest alternatives and the fewest social connections. However, youth may also be at risk; young people in con- flict with their parents may end up leaving the family home. These situa- tions are by no means unique to York Region, but the shortage of afford- able housing, shelter beds, housing services, and settlement services heightens the vulnerability of these elderly and young immigrants, ei- ther to homelessness if they leave, or to abuse if they stay. Few places to turn The limited availability of all types of social services in the region adds to immigrants? difficulties. Housing information may not be available in all immigrants? languages. Newcomers may not know anyone who can write the letters of reference (in English) that are often requested by lan- dlords as part of the application. Many recent immigrants rely on friends, relatives, and religious in- stitutions for negotiations with landlords, but those they turn to may not fully understand landlord and tenant laws and regulations. At the Lan- dlord and Tenant Tribunal, immigrants often rely on interpretation of- fered by strangers, since interpretation services are limited. Services available in Toronto, such as housing counselling, financial assistance with first and last months? rent, and a worker to accompany newcomers to hearings when they face eviction are not available in York Region. Funding for settlement services in York mainly supports em- ployment information and job training. Settlement workers usually refer clients to the York Region Housing Access Unit and offer housing sup- port only when time permits. Immigration status affects access to social housing. Social housing is available only to immigrants who are legal residents. Refugee claimants who are denied refugee status can remain on the waiting list for social housing for as long as their claims are being appealed. If the appeal fails, the household is no longer eligible for social housing. Services such as the Homelessness Prevention Program and the rent bank, which offers assistance to households facing eviction, are available only to the work- ing poor. The pros and cons of social capital Key informants had different views about the social capital available to immigrants and its merits. Several key informants suggested that immi- grants were isolated from friends, relatives, and services. Others sug- gested that recent immigrants often relied on family members, friends, and religious institutions to secure housing. However, newcomers who rely on members of their own communi- ty for information and advice often try to resolve housing issues without accurate or complete knowledge of their rights as tenants. And newco- mers who rent from members of the same ethnic community may be re- luctant to exercise their rights during instances of conflict if the landlord is a member of the same community. Immigrants may leave housing units rather than argue with the landlord. Given the shortage of afforda- ble housing in York Region, they may then become homeless. Discrimination Although many key informants agreed that immigrants experience dis- crimination in their neighbourhoods or from landlords, no one men- tioned discrimination on the part of realtors, mortgage lenders, or home sellers. The grounds for discrimination include household size, source of income, immigration status, ethnoracial identity, and age. Landlords often do not want to rent to large households, regardless of whether or not they are immigrant households. In response, immi- grants may lie about the size of their families and then move into too- small dwellings that become overcrowded. Overcrowding becomes even more acute when immigrants sublet rooms so they can afford the rent. Single mothers, youths, and seniors who rely on government trans- fers are particularly vulnerable to discrimination on the part of landlords who do not want to rent to recipients of social assistance or small public pensions. Immigrants who earn the minimum wage or who have a bad credit history may also be subject to discrimination. Many small landlords are immigrants themselves with limited knowledge of landlord-tenant legislation and regulations. In some cases, the arrangement is made orally, with nothing in writing. Key informants knew of cases in which immigrants who are long-time residents and are now small landlords have taken advantage of newcomers from their home country, financially, emotionally, or even physically. When things go wrong, such tenants flee their housing, unaware of their rights, and may end up homeless. The vulnerability of caregivers and spouses Newcomers who arrive in Canada under the federal Live-in Caregiver program are particularly vulnerable to hidden homelessness. This pro- gram recruits caregivers (most of them women, and mostly from the Philippines) to live in an employer?s household. The caregiver may ap- ply for citizenship after two years of live-in service. Although the care- giver is usually well housed, she is in an extremely vulnerable situation. Caregivers tend to stay silent if they experience physical, emotional or sexual abuse to avoid losing their residence, employment, and means to stay in the country. Some newcomers arrive intending to marry a Canadian citizen; oth- ers marry Canadian citizens overseas and plan to apply for citizenship when they arrive. If women who endure physical and emotional abuse after arrival leave the housing of their sponsor, they may relinquish their right to residency in Canada and thereby lose access to all public assis- tance, including transitional housing and affordable social housing. Therefore some women stay with abusive spouses to avoid homeless- ness. Listening to the voices of immigrants We conducted 9 focus groups in which 62 immigrants participated. They came from the following immigrant communities: Farsi- or Arabic- speaking, Russian-speaking, Filipina caregivers, South Asian groups, Cantonese-speaking, and Korean-speaking. Within certain communities there was considerable socio-cultural diversity. For example, the South Asians came from India, Sri Lanka, the Middle East, and Eastern African countries. Two immigrant groups described themselves as well housed ? the Hong Kong Chinese and Koreans. Responses from South Asians varied: some described serious housing deficiencies and difficulties paying for housing, while others have obtained affordable, adequate, and suitable housing. The Russian-speaking, Farsi- and Arabic-speaking immigrants, and live-in caregivers from the Philippines were at greater risk of home- lessness, largely as a result of the high cost of housing relative to their low incomes. Filipina live-in caregivers emerged as the group most vulnerable to homelessness. People admitted to Canada as refugees ? those from Iraq, Iran, and Sri Lanka in our study ? also struggle for a long time to obtain adequate and affordable housing. Among the South Asians, a very diverse group, there was generally steady improvement in housing conditions over time. But the Arab and West Asian immigrants and those who spoke Farsi, Arabic, or Russian struggled in the housing market even after long periods of residence in Canada. Immigrants told us they had moved to York Region for housing that they considered less expensive than that available in other parts of the Toronto area. Five or six years ago, new houses spacious enough to ac- commodate large immigrant households were available at lower prices in York Region than in the City of Toronto. Immigrants were also at- tracted by York Region?s quiet neighbourhoods and good schools. Many immigrants also moved to York Region because they had friends and family living there. Struggling to pay for housing A common refrain among focus group participants was the stress and strain of living in housing they could not afford. We heard comments such as: We bought this house because we didn?t have any choice [but] I know that I have to sell it. It is quite depressing. I?d like to sell it if housing mar- ket is okay. If I sell it now, I will lose so much money. That?s why I can?t sell right away. I am stuck. (Korean participant) Our current place is newly built, there is no asphalt, and there are no parks or services at all. There is nothing we can do right now, we have to live there. In the summer it is possible to handle a mortgage, but in the winter it is harder because there is less work and there is not enough money. (Iranian participant) The lack of rental accommodation and the problems of large families came through clearly. When we started to look for housing, we found that we have no choice. There is no rental housing whatsoever. It is very hard to rent in Richmond Hill. You have no option but buying a house here. (Korean participant) I couldn?t and cannot find a proper housing that meets my family needs. The apartments reject us because we have too many children. I have no money to buy a house. The option is so limited. That?s why I had to rent this house. But? it is too expensive. I have no proper job. I am so worried about my financial situation. (Korean participant) We also heard about immigrant families living in substandard basement apartments, and about homeowners who had no money for maintenance or repair on the houses they had bought. I have had this leaking water problem so long in my house. But I can?t fix it, because I can?t afford the cost. I don?t know when it is going to collapse over my head. My insurance doesn?t cover [it]. My mortgage is full and I can?t get any more loans from the bank. (Korean participant) Overcrowding was often mentioned. I have even seen 16 people sharing a basement. It is so crowded that they set a house rule on bathroom use on certain time. You can?t use a bath- room if your time is up. This is not a joke. That is a reality and it is on- going situation in our community. (South Asian participant) Overcrowding accelerates the deterioration of housing units and increas- es the risk of homelessness for tenants. In overcrowded dwellings, ten- sions rise. In a tight rental market such as York Region, immigrant te- nants may find themselves without shelter when conflicts occur. Lack of information and landlord discr imination Focus group participants emphasized the need for more and better in- formation about the housing and labour markets. They suggested that detailed information would have helped them make more informed housing decisions. Rather than purchasing housing, many would have looked for rental accommodation in other locations. Immigrants also confirmed that they need more information about their rights and responsibilities in the housing market, and described conflicts with landlords (some of them also immigrants) over problems such as pre-existing damage to rental units. Many had experienced discrimination because of immigration sta- tus, ethnocultural background, and household size, or a combination of factors. Sometimes they don?t give you because you are Iraqi or Indian. I know they do this. And ? most of the immigrants have kids, at least two. So when they see that those people have kids, they refuse? My brother ? found so many places, but they say, ?Oh you have five [children]? No, we want four.? (Arabic-speaking participant) The future isn?t what it used to be Immigrants used to relocate to the outer suburbs as part of a progressive housing career in which the move to the suburbs was evidence of eco- nomic and social upward mobility. In York Region today, however, ho- meownership is a risky housing strategy for many immigrants. Some observers question the significance of this financial risk. They suggest that immigrant homeowners are just like many other homeown- ers who stretched financially to purchase their first house. But the recent housing and economic crises in the United States, and their conse- quences for Canada, have left all households so financially stretched that illness, job loss, or even a moderate increase in interest rates could threaten their access to secure housing. In today?s economic climate, researchers may need to reconsider the conceptualization of a ?progressive housing career? in which ownership is viewed as the ultimate goal. The research also calls into question assumptions about immigrants? reliance on social networks ? many who did so found those resources limited, or experienced different forms of discrimination or exploitation at the hands of more well-established immigrants. Hidden homelessness, hidden patterns In York Region, homelessness among immigrants is a hidden issue. The spacious single-family detached houses that predominate in the region mask pockets of poverty in which recent newcomers may be at risk of losing their shelter. Furthermore, our findings would not have emerged from a study of homelessness in the Greater Toronto Area as a whole. Information about housing issues at the level of the metropolitan area must be disaggre- gated if we are to understand the dynamics of homelessness in the sub- urbs, where population and employment growth are concentrated. Finally, the affordability challenges that immigrant homeowners encounter in York Region force us to reconsider the meaning of ho- meownership. For some immigrants in York Region, homeownership is thrust upon them or assumed before the household has a stable and ade- quate income. These households cling to their housing with all of the attendant stress that delays successful settlement. Recommendations 1. Secondary suites should be legalized in every York Region municipality. Regulating secondary suites makes them subject to regulation and in- spection that will improve the safety and quality of the secondary suites. It provides tenants and landlords with more secure rights and access to legal protections. Secondary suites may also assist homeowners strug- gling with housing that they cannot afford. Finally, secondary suites can be legalized at relatively low cost. 2. Information about housing should be distrib uted more widely to immigrants, ideally prior to arrival in Canada. Recent immigrants want accurate and detailed information about hous- ing costs, the incomes that immigrants can expect during the first years of settlement in York Region, and the availability of rental housing, ideally before moving to Canada. 3. Additional housing supports and interpretation services need to be made available to recent immigrants. Immigrants often need housing information in their own languages, in- terpretation and translation of legal documents, assistance with first and last month rent, and workers to accompany them to legal proceedings when they face eviction or foreclosure. All agencies that serve immi- grants should be funded to provide housing supports or to refer immi- grants to these services at accessible locations in the region. 4. The supply of shelters and other forms of eme rgency and transitional housing in York Region should be expanded. A small number of shelter beds are now available in the northern half of the region, but immigrants at risk of homelessness are concentrated in the southern half of the region. The number and location of shelter beds are inadequate, and policy needs to address both the total number of shelter beds and the spatial mismatch. 5. The supply of affordable housing in York Region should be expanded. A national housing strategy that involves all three levels of government is needed, including policies that stimulate the production of private ren- tal housing. If the supply of multi-family rental units expands, the num- ber of rental subsidies in the region should be increased in response. Areas for future research The pilot study raises four particular questions for future research. First, additional research is needed to explore the housing situations of newcomers in other outer suburbs and determine how immigrants negotiate diverse environments of risk outside the centre of metropolitan areas. Second, future research should include information about the grow- ing numbers of immigrants from China and Afro-Caribbean immigrants in York Region, as well as the housing experiences of immigrant women who are victims of violence. Third, by detailing the housing trajectories of individual immigrant households, researchers can better understand affordability problems. In some cases, affordability problems decrease over time for immigrants, but in others they increase. Researchers need to know how and why dif- ferent households experience different trajectories. Finally, additional research exploring the significance of homeow- nership is called for. The findings raise questions about the conventional notion that the attainment of homeownership indicates success in the housing market. For some immigrants in York Region, homeownership is a precarious status that threatens their ability to settle successfully. Valerie Preston is a professor in the Department of Geography at York Univer- sity in Toronto. Robert Murdie is also a professor in the Department of Geo- graphy, York University. Jane Wedlock works with the York Region Alliance to End Homelessness, in Newmarket, Ontario. Sandeep Agrawal is a professor in the School of Urban & Regional Planning, Ryerson University, T oronto. U zo Anucha teaches in the School of Social Work at York University. Silvia D?Addario, Min Jung Kwak, Jennifer Logan, and Ann Marie Murnaghan are all associated with the Department of Geography, York University. This project received funding from the Homelessness Partnering Secre- tariat, Human Resources and Social Development Canada (HRSDC), The Metropolis Secretariat, Citizenship and Immigration Canada, and York University. The research and recommendations are the responsibility of the authors of the report and do not necessarily reflect the views of the Homelessness Partnering Secretariat, Human Resources and Social De- velopment Canada (HRSDC). Reference Hiebert, D.J., Germain A., Murdie, R., Preston, V., Renaud, J., Rose, D., Wyly, E., Ferreira, V., Mendez, P., and Murnaghan, A.M. (2006). The Housing Situation and Needs of Recent Immigrants in the Montr?al, T o- ronto, and Vancouver CMAs: An Overview. Ottawa: Canada Mortgage and Housing Corporation. Chapter 6 .1 Housing for Aboriginal Children and Youth : The Need for a Holistic Approach NATI ONAL C OUNCIL O F WEL FARE Now, I understand what you?re saying in terms of looking at housing as a key factor that affects the development of children and youth. That?s an honourable undertaking you might say, but I always insist that people try to overcome the western European approach of categorizing and indivi- dualizing various areas of social existence. You have social control. You have the related criminal justice system. So you have the judge and the policemen, lawyer and so on. Then you have the church and all its related exercises. Then you have the schools, universities and so on. And they don?t seem to be really related. I have to remind people that our tradi- tional approach to our existence has been based on recognition that every- thing is related and inter-related and I don?t think you can really get away from that. In that context, then, I would not choose to speak of how hous- ing, by itself, affects children and youth. But I think we have to take the approach that housing is one of the key and basic essentials of life. (Char- lie Hill, Executive Director, National Aboriginal Housing Association) Inadequate Aboriginal housing can be viewed as both cause and effect of poverty, low educational attainment, high unemployment rates, poor health, and outcomes involving children in care and the justice system. Canada?s Aboriginal population is younger and growing at a faster rate than the total population. This factor contributes to an even greater need for new housing. Additionally, some Aboriginal households re- quire housing for an extended family which generates specific housing needs. Another important consideration which is part cause and part effect of the housing crisis is the demographic trend indicating that Aboriginal people are changing residences both within cities and in and out of cities at a more frequent rate than the overall population. This high rate of churn among the urban Aboriginal population is having adverse effects on individuals, families, communities, and service providers. As noted in one recent study: Many social programs that provide services to urban Aboriginal popula- tions, such as health, family support and counseling, and education, are designed on a neighbourhood basis to ensure a coordinated response to multi-faceted family and individual needs. Frequent mobility among Ab- original families can result in discontinuity or disruption of service provi- sion, with negative consequences for the family and service provision agencies. Discontinuity in service provision can be especially pronounced among high-need families such as those of lone female parents, who are among the most mobile, yet often in the most need As the independent indigenous submission to the United Nations Committee on Economic, Social, and Cultural rights in response to Can- ada?s periodic reports has noted, (Clatworthy & Norris, 2007, p. 228). The issue of aboriginal housing is not simply a matter of differences in living standards. Overcrowded and dilapidated houses pose a significant threat to the physical health of aboriginal people to TB, diabetes, and ob- esity. Psychologically inadequate housing among aboriginals reinforces a sense of marginalization and hopelessness. Furthermore adequate and af- fordable housing is essential to the stability children need to perform well in school; the need to move frequently hurts a child?s social and academic development (INET, 2006, p. 36). Aboriginal housing needs must be viewed holistically; all too often however, they are not. Nor do housing programs and policies for Abori- ginal peoples even provide adequate shelter. Aboriginal Households (Non -Reserve) While the percentage of Aboriginal households in need has declined since 1996, it remained extremely high at 25 percent of all households in 2001. The absolute number of households in core need has risen from almost 70,000 to 74,000 between 1996 and 2001. Inuit households, as of 2001, were twice as likely to fall into core housing need as non- Aboriginal households (Canada Mortgage and Housing Corporation (CMHC) Households that have to spend more than 30 percent of their in- come to find adequate and suitable housing in their local housing market are considered to be in core housing need. Adequate housing means the dwelling has basic plumbing facilities and is not in need of major repair. Suitable housing means there are enough bedrooms for the size and make-up ? age and sex ? of the family (City of Toronto, 2006, p. 21). , 2004a, p. 19). Housing costs are greater as a percentage of income for Aboriginal people, mainly because, on average, Aboriginal household incomes lag those of non-Aboriginal households. According to the 2001 Census, Ab- original households reported an average 19.9 percent less than non- Aboriginal households. Yet Aboriginal households average shelter costs ($705) were only 8 percent lower than the average shelt er costs ($766) of non-Aboriginal households. As a result, Aboriginal households spent more of their income on shelter relative to non-Aboriginal households, and a greater percentage fell into core housing need (CMHC, 2004b). At the same time, the quality of Aboriginal housing is far inferior to that of non-Aboriginal households. In an examination of non-reserve housing, the Canadian Mortgage and Housing Corporation (CMHC) showed while non-reserve Aboriginal families represent only 2.8 percent of all families, they make up 4.8 percent of those in core housing need. This figure rises to over 22 percent in Manitoba and Saskatchewan, where about 10 percent of all households are Aboriginal (CMHC, 2004b, p. 3). In large cities, the need is even greater than in small urban centres. Almost 25 percent of Aboriginal households in census metropolitan areas and census agglomerations were in core housing need in 2001 compared to 20 percent in small urban centres (CMHC, 2004b, p. 3). Inuit people have been particularly hard hit in housing, as funding is often apportioned by population numbers and as a small proportion of all Aboriginal peoples, Inuit often receive little money. Lack of local ma- terials for housing, as well as high transportation costs, make housing construction even more expensive than in other areas. Inuit are most likely to face overcrowding. In 2001, 20 percent of Inuit households were crowded followed by 10 percent for First Nations on-reserve and only 2 percent for the non-Aboriginal population (ITK, 2005). As Okalik Eegeesiak, Director of the Socio-Economic Development Department of Inuit Tapiriit Kanatami, commented, ?It is kind of hard to study for anything when you have three or four generations living in one house? [and] with so many people living in the house and poor circula- tion, that compounds the health status of everybody in the house.? Maria Wilson, Training and Development Coordinator of the Socio- Economic Development Department of Inuit Tapiriit Kanatami, adds: I think when speaking of Inuit and the North, most Canadians do not realize how different it is to do anything there. It?s very expensive to bring building materials if one has to fly everything in. It is estimated that $9 billion is required to meet current need for housing in Nunavut. So how much of it is actually to buy building supplies and how much is to pay to bring it all into communities? And the population growth as you may have heard, 60 percent of the population is under the age of 25. Overcrowding leads to stressful homes and no space for children to study, so they skip school more often and have poorer education results. Overcrowding also contributes to spousal abuse and other crimes. ITK (Inuit Tapiriit Kanatami) has linked overcrowding to poor health, especially for infants, and to the transmission of infectious dis- eases such as tuberculosis (25 times the Canadian average for Inuit) as well as increasing the risk of injuries, mental health problems, family tension, and violence. ?These stressors are powerful triggers for negative coping behaviors such as dependence on alcohol and drugs? (ITK, 2005, p. 5). A report prepared for ITK also noted, ?Many Inuit offenders had difficult home environments during childhood, including exposure to violence and substance abuse? (Trevethan et al., 2004). Alastair MacPhee, a childcare expert with the Congress of Abori- ginal Peoples, sums it ups succinctly: ?Access to good-quality housing is essential for the well-being of Aboriginal children and the federal gov- ernment should undertake a national housing strategy for all Aboriginal peoples.? Homelessness Young people make up a crucial component of the homeless population and Aboriginal youth are over-represented in this group. Studies indi- cate Aboriginal people as a whole are over-represented among all home- less people ?in every major city where statistics are available? (CMHC, 2004a, p. 3). In the North and in other rural areas the problem is often hidden and overcrowding or ?couch surfing? are the methods by which it is concealed. As noted by Dr. Cathy Richardson, a M?tis psychologist, ?So what are people on the lower end doing? Many of these people are becoming homeless and that certainly is affecting a lot of M?tis.? Interviews conducted with Aboriginal street youth for the Royal Commission on Aboriginal Peoples (RCAP) found that for Aboriginal youth, their cultural background, history, structural conditions, and ex- perience on the street were different from those of other street youth. Many experienced racism, in addition to the stigma encountered by all street people. The youth also spoke of identity confusion and self-hatred, dislocation from home, difficulty in reunification and ignorance of Ab- original rights, history, and culture (RCAP, 1997). Some of the key indicators are: ? in Calgary, Aboriginal people make up 2 percent of the city?s popu- lation but 17 percent of the homeless population; ? in Edmonton, Aboriginal people make up 4 percent of the city?s population but about 37 percent of the homeless population; ? in Ottawa, Aboriginal youth are 18 percent of the population of homeless male youth and 19 percent of homeless female youth, but only 2 percent of the population of Ottawa is Aboriginal (CMHC, 2001). Garry Jobin, Coordinator of BladeRunners in Vancouver (a program that matches construction labour needs with disadvantaged, street- involved youth, about 95 percent of whom are Aboriginal), observes: A lot of our kids are dealing with homelessness. So, you have to know all the community agencies. If you?re dealing with a kid that?s kicked out on the street at 1:30 in the morning and is at the corner of Main and Hastings with all his clothes, you need to come in and know how to deal with that right away, and know who to call to get that young person housing, or you?re going to lose them immediately. Adds Charlie Hill: You have to admit, and the research supports the fact, that Aboriginal people are the worst housed in Canada. There are programs that were helpful, but they were capped. Like the CMHC cap on new social housing in 1993. It was a successful program. The membership of the National Aboriginal Housing Association acquired about 11,000 housing units up until ?93 but after that then there was a waiting list at the time but it has grown and grown. More and more people have become homeless. Home- less people start to include single parents and families, then it?s really time to take a hard look at how come this is resulting. Now this isn?t just limited to Aboriginal peoples. But the Aboriginal people are the worst of the worst. After 1993, the existing subsidies at that time still continued ex- cept when the mortgages are terminating, then there?s no more subsidy. So then the rents have to go up which has an impact on the families. In that case, it?s a vicious circle. Aboriginal youth are over-represented among the homeless popula- tion in Canada. Given the demographics of a burgeoning youthful Ab- original population combined with a shortage of adequate Aboriginal housing, there is an immediate need for increased housing supports, or the Aboriginal homeless population may be expected to grow. First Nations On -Reserve According to Hill, between housing on- and off-reserve, ?there?s not all that much difference. If you?ve got the economic base, you can support a house. Without the economic base, you?ll find it very, very hard to support a house. I?m not talking about a home, I?m talking about houses. The major difference, of course, is the type of tenure. The fact that Canada holds the title to the land. I mean, I don?t know how they arrived at that. They came over and said, ?This is our land, but you can stay here if you want.? It seems kind of ironic, but anyway, that?s what happened and they put us on reserves. People say that was to protect us, but it wasn?t really. They fully expected us to die off as a people. If we wanted to leave the reserve and live elsewhere then we had to become what they call enfranchised and you had to pretend that you were not an ?Indian.? And so when people say, ?Well, yeah, you chose to go and live on the reserves,? that?s not true. We were forced to go on the reserves and if you look around anyplace right now, it?s all the worst land and we were forced into those territories. A lot of people don?t know that up until 1951, we had to have passes to leave the reserve. CMHC noted that in 2001, 22.5 percent of on-reserve Aboriginal households were living in inadequate housing and in core housing need compared to 2.5 percent of non-Aboriginal households (CMHC, 2004a). The federal government has openly acknowledged the shortfall in re- serve housing. In October 2006, Indian and Northern Affairs Canada (INAC) reported: Overcrowding and inadequate housing are of particular concern on- reserve, where there is a current housing shortage of between 20,000 and 35,000 units. The shortfall is growing by an estimated 2,200 units a year (INAC, 1997). CMHC also noted, As of March 2004, INAC [Indian and Northern Affairs Canada] r eported a total of 95,479 dwelling units on-reserve, of which 16,878 required major repairs and 5,199 needed replacement. On-reserve housing shortages are currently estimated at 20,000 units, with an additional 4,500 new units needed annually to meet the requirements of new households (CMHC, 2004a, p. 4). In recent years, in spite of these pressing needs, new on-reserve housing has declined or stagnated. In 2002?2003, only 1,889 homes were built, down from 4,254 in 1993-94. The number of renovated dwellings has barely changed, from 4,126 in 1993-94 to 4,224 in 2002?2003 (INAC, 2004, p. 61). Housing challenges include such inexcusable conditions as bad wa- ter and lack of sewage services. Indeed, the Assembly of First Nations has indicated, Currently [in 2006], almost 12 percent of First Nations communities have to boil their drinking water. Six percent of First Nations homes ? over 5,000 homes ? are without sewage services. Almost 1,600 homes lack hot water, cold water, or flushing toilets (CBC News, 2006). There were 89 First Nations communities under either ?boil water? or ?do not drink? advisories as of June 8, 2007 (Health Canada, 2007). These conditions have negative effects that go beyond the physical health of the individual and extend to all elements of community well- ness. Aboriginal Women And Housing Many First Nations women and their children face a particular legal is- sue in terms of housing. Arising from the distribution of powers in the Constitution Act, 1867 , provincial or territorial law governs how assets of a marriage or common-law relationship are to be divided upon break- down, including real property such as a house. The legislation generally provides for equal division between spouses. However, these laws do not apply on-reserve, as a result of subsec- tion 91(24) of the Constitution Act, 1867 which gives the federal govern- ment exclusive law-making authority over ?Indians, and lands reserved for the Indians.? this has been interpreted to mean that provincial and territorial matrimonial property laws do not apply to real property on- reserve. Since there are no federal provisions in the Indian Act or else- where that fill in this matrimonial property gap, people living on-reserve generally have no legal system for resolving issues relating to land and houses upon a breakdown of their relationship. Thus, First Nations women currently have no right in law to certain assets on-reserve where their marriage breaks down, unlike all other women in Canada; they and their children are therefore left with no legal claim to occupy the family residence. They may be forced to leave the matrimonial home and due to acute housing shortages, may also have to leave the reserve. Where family violence is involved, the woman and her children are rendered all the more vulnerable by this gap; sometimes remaining with the abuser for lack of an alternative (Mann, 2005). The human rights of First Nations women and their children are vi- olated and they are discriminated against when they are unable to exer- cise rights they would have off-reserve (Mann, 2005). First Nations women want an avenue of redress and effective en- forcement mechanisms for matrimonial matters involving real property on-reserve. Some Aboriginal people want the repeal of provisions in the Canadian Human Rights Act preventing its protections from applying o- reserve. That would allow women some recourse if they believed a Band Council?s decision involving housing was discriminatory. As Dr. Richardson observes, for Aboriginal women everywhere, vi- olence, poverty, and housing are closely linked, Courts are deciding that if a woman is in a transition house because she had to leave her family home that is pretty unstable, so often they would decide to put children with the assaulting spouse, the father, the child?s Dad, because he might still have a home. So women are bearing a dispro- portionate responsibility for violence in families which is often used against them. So again, the issues of being vulnerable or subjected to vi- olence they often relate to housing that is safe. You don?t have a good place to live; you know children are removed from mothers due to pover- ty as well as being the victims of violence. For me, I would call that another human rights abuse. These issues are all related to violence and child welfare, human rights and how we are dealing with families. It is also noteworthy that Inuit women are often tenants of their homes, with the man?s name on the lease, also resulting in great difficul- ty in removing a male perpetrator of violence from the home (Mann, 2005). Lone parenting Aboriginal women are more likely to be lone parents compared to non- Aboriginal women. This means more will be in core housing need due to high urban housing costs for single-income families. Young women are often the most affected as lone parents. Nearly half (47 percent) of Ab- original lone parents experience core housing needs. Dr. Richardson observes the change in M?tis family stru ctures: So what happened, in a very short time our families moved from being what we would call extended families or living more communally both in communities but within houses it was quite normal that grandparents might live with a family or an aunt or uncle. And so we moved from that situation maybe quite quickly to a period of a nuclear family to a time when many of our families actually now are single-parent families living in urban settings or in small resource towns, so quite vulnerable to the flux of industries and facing a lot of issues related to poverty. So when I think of what are the issues, well one issue certainly is housing. Architecture has never really helped the M?tis, since 1885. So when fam i- lies come to a city like Vancouver and they need to get an apartment, of- ten it is one-bedroom where a mother will live with her children. And it?s really hard to find apartments that are suitable for larger groups of people like an extended family. So I think housing is seriously related to other is- sues that link isolation, the need for quality childcare, peer support, and family support and for communities. Those kinds of things could really be supported in advance through proper community-based housing. Inadequate housing, family violence and underlying poverty issues are closely inter-related with Aboriginal children being taken into care. Comprehensive Approaches As with all indicators of Aboriginal child and youth poverty, housing does not stand alone but must rather be viewed as interconnected to all other elements of the experience of aboriginal peoples. Hill notes, So I mentioned the social aspects, the cultural-linguistic aspect, the eco- nomic aspect, and the education aspect. All of these play a part in the suc- cess or lack of success in terms of housing. Over and above that is the whole question of racism that still exists. It?s not just housing. You can?t isolate housing and say that if you?ve got a good house that things will be great. There?s this kind of stuff that you have to put up. There?s that fact that the people, by and large, don?t have any economic base with most of them coming from isolated communities so there?s a language barrier. Low education. But there?s racism on top of everything else. The result, Hill notes, is that Aboriginal people who do obtain a well-paying job and move into a neighbourhood that?s ?all WASP,? often run into discrimination, while what usually happens is that Aboriginal people wind up in the worst housing in a city: ... that lends itself to this repression and being put into a ghetto-type of situation. So if you?re forced into a negative housing situation, there?s overcrowding, there?s a struggle to pay the rent, because you can?t get a good job and there?s also the social repression. One more thing is that where there is a low-rent housing situation, usually that brings with it a bunch of other social ills. People who have not had a great opportunity whether they?re white, Indian or black or what. This in turn lends itself to fostering not a very good attitude in terms of social responsibility. That?s a nice way of saying there?s a lot of troublemakers that hang around in low-rent areas. Now, this is not stereotyping, it?s through experience. Having said that, you still want to hang out with your own people. You don?t want to abandon your own people and go and try to live elsewhere because you?re not well-received. You get pretty darn lonely so you stay with your people, your culture. I think that it is very important that our people start speaking up more for themselves. As always, funding, intergovernmental co-operation, and aboriginal control are major factors in the success or lack thereof in Aboriginal housing. According to Hill: There?s two things that I recommend that are essential. One is the increase of resources to help us acquire more housing units but the other thing is that it is essential that we be recognized as a people who can support our- selves in terms of administering houses and so on. I guess putting it another way is, Indian-controlled or Aboriginal control over their housing program. And a holistic approach to housing means addressing more than just Aboriginal people having a place to call home, says Hill: One of the things that the Urban Native Housing Program incorporated into its activities was the role of tenant counsellor. I think this role is one of the keys and is one of the major differences between the mainstream housing and the Aboriginal housing is that we do have tenant counsel- lors. I think that the tenant counselling is one aspect, but there also have to be special provisions made to provide counselling for the kids by other Ab- original people. I mean education that will bring to the attention of our Aboriginal kids the fact that all of this stuff has happened but here we are and we have to make a fresh start. To help them understand why people are like they are on one hand, but why they are treated the way they are on the other. So, with regards to the youth I mean we are all children at one time, we were youth and we are adults. The difference is that we get to be bigger and stronger and you can start acting out. The one thing that I know has been instrumental in helping a lot of people find their feet again is their re-grasp or the revitalization of traditional teachings in their day-to-day life. I think that this is one of the areas that really has to be stressed. Of course the housing groups are not in a position to provide that kind of instruction but if the tenant-counsellor position was either expanded or rendered more specific to dealing with children and youth then that might be one element that could be incorporated into the day-to-day ac- tivities of the tenant counsellor or person who is supposed to work with the tenants. In addition to policy/program approaches pertaining to tenancy, says Hill, Some people over the years have been able to move into well-paying jobs so they do have an economic base. I think that many are now in a position where they could carry mortgages if they don?t have enough money to pay to buy a house outright, so I think this is the other area the federal government should really look at. In devising ways, working with banks and real estate people to give Aboriginal people a break in terms of relief in terms of interest and things like that, so that would help them to enter the home ownership area. I think that homeownership, in turn, makes a person feel better which is passed along to the kids but also people, non- native people start taking a different view, ?Oh they have a house, they?re just like us.? But we?re not like them, but we have to live in the same en- vironment. Keep in mind, the other thing is that we have been specifically excluded in the sharing of the resources of this country. The resources have been turned over directly to the provincial governments and it has only been recently, like the diamond mines, where our people have really insisted that they have a share of the resources that are extracted. In this context, if we could share the resources that are extracted then our people would then be, in fact, better-off and I think you have to admit that one reason Canadians have been well-off is because they had access and they took all the resources. So we?re only a drop in the bucket in terms of population so it would be a simple matter to share some of the benefits from various resources which in turn then would help housing which in turn would help the kids. The well-being and future prospects for Aboriginal children and youth are so tied to housing that only a comprehensive long-term ap- proach will fill the need. Conclusion Aboriginal housing is substandard and inadequate at rates disproportio- nate to that experienced by the non-Aboriginal population; Aboriginal people and youth particularly are also over-represented in the homeless population. As with all indicators of Aboriginal child and youth poverty, hous- ing does not stand alone but must rather be viewed as interconnected to all other elements of the experience of aboriginal peoples, including rac- ism and discrimination. Aboriginal housing issues must be approached through a holistic strategy. Adequate funding, intergovernmental co- operation, collaboration with aboriginal communities and a focus on community economic development are all important for housing ade- quacy and sustainability. We have to build a more inclusive society where Aboriginal people can benefit. If we do not, an increasing rate of human misery?and mili- tancy?is almost certainly to be expected. This doesn?t serve anyone well. The National Council of Welfare is not an Aboriginal organization, but it has a unique mandate to advise the federal government on poverty and social development. From this perspective, we are making what we believe are important recommendations. We are not experts on First Na- tions, M?tis , and Inuit issues, and our goal is to support and complement the recommendations already made in Royal Commission on Aboriginal Peoples and beyond by Aboriginal organizations. Canada is a rich country. Collectively, we have choices ? far more choices than most Aboriginal children and youth can even imagine. The NCW wants the federal, and other, governments to choose to: 1. Adopt a national anti-poverty strategy, as outlined in the NCW?s solving poverty: four Cornerstones of a Workable national strategy for Canada. Aboriginal poverty cannot be solved in isolation from other Canadians who are also impoverished due to factors like disa- bility, racism, sexism and lone-parenthood. And as highlighted re- peatedly in this report, challenges facing Aboriginal people cannot be solved in isolation; everything is interconnected. The only way forward is a comprehensive strategy. 2. Adopt, within this national strategy, a specific long-term vision for Aboriginal peoples along with targets, timelines, indicators, intergo- vernmental coordination, and accountability to Aboriginal Cana- dians for results. 3. Include First Nations, M?tis, and Inuit people in creating every part of the strategy, and guarantee that women have an equal say, espe- cially in the interests of children, and that young women and men and girls and boys themselves will also have their voices heard. 4. Immediately invest sufficient resources to meet the basic needs of every Aboriginal child and young person, regardless of their status or where they live (food, clothing, drinkable water, safe housing, ear- ly learning and care, access to education and health care) through in- creased income and services that foster autonomy and dignity. Poverty is a political choice. Social inclusion and well-being are equally possible. There are no excuses in this country to give in to the status quo. This is a summary of Chapter 6 and the concluding sections of Time to Act: First Nations, M?tis and Inuit Children and Youth, a 200 7 report by the National Council of Welfare. The full report is available in both English and French at www.ncwcnbes.net. Reproduced with the permission of the Minister of Public Works and Government Services Canada, 2009. References Canada Mortgage and Housing Corporation (2001). Environmental scan on youth homelessness. (Socio-economic Series, 86). Ottawa: CMHC. Canada Mortgage and Housing Corporation (2004a). Aboriginal housing back- ground paper: Part II. Ottawa: Indian and Northern Affairs Canada. Canada Mortgage and Housing Corporation (2004b). 2 001 Census housing series: Aboriginal households, Issue 6 rev. (Socio-economic series No. 04-036, p. 3). Ottawa: CMHC. CBC News (2006, November 21). Aboriginal peoples, 10 years after the royal commission. Retrieved November 23, 2006 from http://www.cbc.ca/news/background/aboriginals/status-report2006.html. City of Toronto (2006). Perspectives on housing affordability. Toronto: City of Toron- to, City Planning Division. Clatworthy, S., & Norris, M. J. (2007). Aboriginal mobility and migration: Trends, recent patterns, and implications: 1971-2001. In J. P. White, P. Maxim, & D. Beavon (Eds.) Aboriginal policy research: Moving forward making a difference (Vol. 4). Toronto: Thompson Educational Publishing. Health Canada (2007). Drinking water advisories. Ottawa: Author. Retrieved June 18, 2007, from http://www.hc-sc.gc.ca/fnih-spni/promotion/water- eau/advis-avis_concern_e.html . Indian and Northern Affairs Canada (2004). Basic departmental data 2003 . Ottawa: Indian and Northern Affairs. Indian and Northern Affairs Canada (2006). Fact sheet: Aboriginal housing. Ottawa: Author. Retrieved June 29, 2007, from http://www.ainc- inac.gc.ca/pr/info/fnsocec/abhsg_e.html . Indigenous Network on Economics and Trade (2006). Independent indigenous sub- mission to the United Nations Committee on Economic, Social and Cultural Rights in response to Canada?s Periodic Reports. Joint Submission to the United Na- tions Committee on Economic, Social and Cultural Rights, 36th Session. Inuit Tapiriit Kanatami (2004). Backgrounder on Inuit and housing. Ottawa: Envi- ronment and Health Department, Inuit Tapiriit Kanatami. Mann, M. M. (2005). Aboriginal women: An issues backgrounder (p. 5). Ottawa: Sta- tus of Women Canada. Royal Commission on Aboriginal Peoples (1997). Urban survivors, Aboriginal street youth: Vancouver, Winnipeg & Montreal [Kaptitipis ep imohteyahk: Vancouver, Winnipeg and Montreal]. By L. Gilchrist & R. A. Winchester, RCAP Research Reports, Urban. In For seven generations: An information lega- cy of the Royal Commission on Aboriginal Peoples [CD -ROM]. Ottawa: Librax- us. Trevethan, S., Moore, J.-P., Naqitarvik, L., Watson, A., & Saunders, D. (2004). The needs of Inuit offenders in federal correctional facilities. Ottawa: Correctional Service of Canada. Chapter 6.2 Homeless Aboriginal Men : Effects of Intergenerational Trauma DR. PETER MEN Z IES As an Aboriginal therapist working out of Canada?s largest mental health and addiction treatment facility, I have found the prevailing theo- ries on homelessness fail to provide an adequate explanation for why a growing number of Toronto?s homeless service users are people of Ab- original origin. I work closely with homeless Aboriginal people who struggle daily for survival. Consistently, they report a personal or family history of traumatic events that have left an indelible mark on their lives. In many cases, this has resulted in a severing of ties from both birth fami- ly and community of origin. This scenario repeats itself among a diverse cohort, with those in their early 20s sharing family histories that reflect the experience of those in their 50s and even 60s. While theories related to the cause of homelessness are beginning to recognize broader systemic factors such as poverty and lack of housing, little consideration is given to the cumulative impact government poli- cies have had specifically on Aboriginal peoples. There is increasing evi- dence that more than 140 years of social strategies aimed at the assimila- tion, segregation, and integration of generations of Aboriginal children into mainstream Eurocentric culture have resulted in personal, familial, community, and national trauma. A brief overview of how public policy has reframed the lives of Aboriginal peoples in Canada is set out below. Public Policy and Canada?s Aboriginal Peoples Canadian social policy has been instrumental in creating institutions that have eradicated value systems that existed for thousands of years, re- placing them with doctrines that continue to disrupt life for Aboriginal peoples and creating a legacy of trauma. The Indian Act of 1876 established the federal government as the ?guardian? of Aboriginal peoples. Artificial settlements were created, segregating individuals into groups that were defined by authorities outside existing community networks. It set up authority within these artificial settlements and created hierarchy and decision-making authori- ties that did not reflect traditional values and practices (Royal Commis- sion on Aboriginal Peoples, 1996, Vol. 3). Today, the Act continues to perpetuate unstable and inequitable programming and delivery of support services, especially to Aboriginal peoples living off-reserve and in urban communities. By creating artifi- cial separations and introducing external control over the relations be- tween family members, within communities, and across peoples, the Act has effectually isolated community members from one another. The Canadian government has used other mechanisms, including religious institutions, to transform Aboriginal communities. Between 1840 and 1983, more than 100,000 Aboriginal children were placed in the residential school system (United Church of Canada, 1994) for the pur- pose of assimilation, segregation, and integration into mainstream Cana- dian society. Separation from their family for months, even years, at a time resulted in children losing their language, culture, and spiritual be- liefs, as well as sense of belonging to a family or kinship network. The physical, sexual, and psychological abuse endured by many children, including the more than 100,000 residential school survivors living in Canada today, is chronicled by both the Assembly of First Nations (1994) and the Royal Commission on Aboriginal Peoples (1996). With the integration of Aboriginal children into the public school system by the mid-1900s, child welfare became the new instrument of government assimilation policies. From 1951 until the late 1960s, the fed- eral government and the provinces and territories were locked into fund- ing disputes related to Aboriginal peoples. Johnston (1983) introduced the phrase ?the Sixties scoop? (p. 23) to identify the overwhelming num- ber of Aboriginal children removed from their homes and communities by child welfare authorities during this period. Taking a crisis interven- tion approach to child welfare meant that Aboriginal children were per- manently removed from their homes and placed in foster care or made wards of the Crown (Andres, 1981; Johnston, 1983; Richard, 1989; Timp- son, 1990). Recent child welfare studies have described the long-term effects of removing Aboriginal children from their birth family and placing them in non-Aboriginal homes (Couchi & Nabigon, 1994; Frideres, 1998). Lo- cust (1999) used the term ?split feathers? to describe the long-term psy- chological problems developed by Aboriginal children adopted or placed in foster care outside of their culture. Forced to assume the values of another culture that derided their own belief system, Aboriginal child- ren were left in a cultural vacuum, relating neither to mainstream culture nor to their own community. Warry (1991) reported that as these child- ren matured, they became ?apples?: racially ?red,? or Aboriginal, on the outside, but culturally ?White? on the inside. Aboriginal psychiatrist Clare Brant (1990) identified how social anomie ? a feeling of being dis- connected from any particular cultural group ? has contributed to poor mental health in many Aboriginal communities. Brant (1993) linked this condition to the fact that residential schools created confusion and value conflict for the children, taking them from their homes without provid- ing an alternative home with which they could form a positive identifica- tion. Historical social policies have affected multiple generations of Ab- original peoples. The severing of family and community ties ? that is, creation of a homeless state ? has left a legacy of traumatized individu- als who may be unable to function in mainstream society. Left depen- dent on social institutions, many Aboriginal peoples are unable to ad- dress their individual needs. The trauma of separation from family and community ? the Abori- ginal home ? has affected the ability of individuals to achieve balance in their physical, mental, emotional, and spiritual well-being. When expe- rienced by more than one generation, personal trauma becomes institu- tionalized within a family. Where multiple families within a community experience similar life events, the community is left without the re- sources required to effectively address the resultant social consequences. As adults, former residential school students and child welfare sys- tem survivors have demonstrated symptoms of anxiety disorders, alco- hol and substance abuse, depression, suicide, and low self-esteem that are significantly higher than those of the general population (Beisner & Attneave, 1982; Gagne, 1998; Hodgson, 1990; Mussell et al., 1991). Psy- chologists have developed the phrase ?residential school syndrome? (Brasfield, 2001) to explain the array of behaviours exhibited by Abori- ginal residential school survivors. The physical, sexual, mental, and emo- tional abuse experienced or witnessed by generations of children has left many Aboriginal peoples with a variety of mental health conditions. While post-traumatic stress and associated disorders focus on the individual?s response to such trauma, emerging studies on post- traumatic stress do not acknowledge the systemic conditions that allow these disorders to be sustained or even perpetuated within families and across generations (Duran et al., 1998; Yellow Horse-Brave Heart, 2003). Furthermore, research on trauma does not connect the individual?s expe- rience to broader, systemic conditions that perpetuate and exacerbate the individual?s experience. Kirmayer et al. (2000) concur that the focus on individual trauma does not adequately reflect the Aboriginal experience. The authors suggest: The emphasis on narrating personal trauma in contemporary psycho- therapy is problematic because many forms of violence against Aborigin- al people are structural or implicit and so may remain hidden in individ- ual accounts ... Individual events are part of larger historical formations that have profound effects for both individuals and communities (p. 613). Research Project In 2004, as the basis of my doctoral thesis, I undertook a study to identify whether any link could be established between intergenerational trauma and homelessness. A growing body of research suggests intergenera- tional trauma as an explanation for the social and psychological prob- lems experienced by Aboriginal peoples. However, this theory is not specifically linked to homelessness. The purpose of this study was to determine whether intergenera- tional trauma can serve as an explanation for Aboriginal homelessness. Although the growing body of research describes intergenerational trauma and notes its causes in historical processes, that research does not present a clear set of indicators of intergenerational trauma. This study provided an opportunity to identify the indicators of intergenerational trauma in a sample of Aboriginal homeless men. The study employed a qualitative methodology, involving key Ab- original stakeholders in the collection and evaluation of the data and the- reby contributing to an understanding of the unique factors that explain homelessness among Aboriginal men in Toronto. Arguably, the ap- proach is a modified form of grounded theory (Creswell, 1998; Neuman, 1997; Rubin & Babbie, 1997) in that the intention is to proceed inductive- ly and build an explanatory model rather than test hypotheses. The par- ticipants in this study played an active role in helping us understand the issue, so that any proposed solutions to homelessness reflect their path- way to home. Sample Established in 1986, Na-Me-Res (Native Men?s Residence) is a 60-bed emergency hostel for men in downtown Toronto. The agency was se- lected for this study because it is the only shelter for Aboriginal men where residents are allowed to stay for an extended period of time, and because it also has counselling staff on site to support residents with a range of psychological and emotional needs. The research participants included 16 adult Aboriginal men be- tween 18 and 64 years of age who were using the services of Na-Me-Res during the period in which this study took place. An additional five men participated in a focus group discussion. Quota sampling was used to ensure that the sample included men who were within the age ranges 18 to 24, 5 to 49, and 50 to 64. The sampling also attempted to include men from a variety of places of origin ? urban centres, rural communities, and First Nations communities. The 16 participants ranged in age from 26 to 55 years, with a me- dian age of 40.8 years. Thus, all the participants were born between 1950 and 1977, with a majority born in the early 1960s. Most of the partici- pants indicated that they were born in Northern Ontario. Two had no information about their birth community, and two others indicated that they were born in other provinces. For many, the link to their birth communities had been severed at an early age through placement in child welfare agencies and/or adoption. Only 44 percent of the partici- pants were raised within their biological family. Of the seven partici- pants raised by a biological parent or extended family member, four were raised in blended families with a step-parent strongly influencing their early family life. Of the nine participants who were not raised by their biological families, four were adopted and the remaining five had been placed at a relatively young age in foster care or residential boarding schools. Of the four adopted participants, two had families who had moved outside Canada. Therefore, as compared to the norms for society, the partici- pants in this study had relatively atypical and unstable family arrange- ments. The participants had a range of education and work experience. About two-thirds of the group had either some high school education or less than Grade 8. Two participants had received some vocational train- ing as young adults, two others had some postsecondary training at a college, and one reported having a university degree. Of the 16 participants, only one had a full-time job at the time of the interviews, and that was temporary. Three indicated that they had been previously employed in teaching, providing art instruction at a post- secondary institution, and one worked as a computer programmer. Two others had recently undertaken apprenticeship training. The remaining participants say they had work experience as general or unskilled la- bourers in either the construction or manufacturing industries. Of the 16 participants, 14 had recent experience with the criminal justice system. All had been charged either while intoxicated with alco- hol or under the influence of drugs. Over one-third of the participants indicated a history of mental ill- ness and hospitalization related to depression. One revealed that he had been assessed at an early age as having fetal alcohol syndrome. The issue of being formally diagnosed within the mental health system was not explored at any length, nor did I attempt to substantiate reports of clini- cal diagnoses. Rather, the interviews focused on having the participants describe their lives from their own perspective. The length of time the participants had been staying at the hostel ranged from one night to several months. For many, their current home- lessness was closely linked to deinstitutionalization following a release from a hospital, treatment centre, or incarceration. Aboriginal women were not included in this sample because I be- lieve their homelessness is affected by other systemic factors related to the oppression of women in North American society. A study of home- less women would benefit from a separate analysis that integrates a fe- minist perspective in both the literature review and data analysis. For this reason, the study focused on the factors contributing to homeless- ness among Aboriginal men. Interviews Semi-structured interviews were based on an interview schedule with questions related to specific themes. These themes included personal and family experience with residential schools and child welfare (foster care, adoption); life history as it related to community of birth, connections to birth family, and relationship with extended family; connectivity to Ab- original culture; mental health and substance abuse issues as expe- rienced by the participants and their family members; housing history; and length of time living without permanent shelter. The order of the questions varied to allow each participant to share his story in his own way. These themes were explored with participants to determine if other family members or community members were af- fected by similar events or circumstances. Additional questions naturally emerged from responses to the interview schedule and allowed for fur- ther exploration of themes. Indicators of intergenerational traumatic ex- periences emerged from these themes. The interviews were audiotaped and later transcribed for analysis. Key themes emerged from the data and were recorded in a separate document. Using an inductive approach, the meaningful units for each of the interviews were then recorded in table format to assist in a com- parison of the responses (Tutty et al., 1996). Finally, a cross-case analysis was prepared (Patton, 1990). Initial categories were identified and mea- ningful units of information were organized and coded into the appro- priate theme area on a chart. As Dey (1993) suggests, this initial categori- zation was reviewed several times using the constant comparison method to ensure that themes were easily defined and units coded ap- propriately. Validity was measured by sharing the results of the study with a fo- cus group of five different homeless Aboriginal men who used Na-Me- Res services (Neuman, 1997; Padgett, 1998). The purpose of this exercise was not to modify the findings of the study, but to determine whether the interpretation was valid and to supplement the findings with addi- tional notes (Holt, 1993). Ethical Considerations As I am an Aboriginal male within the age group of the sample frame, I believe that my personal affiliation as a member of the Aboriginal com- munity allowed me to gain the trust of the research participants and con- tributed to their willingness to share their experiences with me. The men who volunteered for this study did not receive any financial remunera- tion for their participation. Instead, the traditional offering of tobacco was made to all the participants as a gesture honouring their willingness to share of themselves. The men were advised that they could withdraw from the study at any time. Details of how their private information would be protected and remain confidential were reviewed. Counselling support was available to each of the participants, after the interview, to ensure their safety and allow for debriefing. Each of the men signed an agreement to participate and was given a copy of that agreement and my contact information. Results: Indicators of Intergenerational Trauma Among Homeless Aboriginal Men The participants were initially asked to reflect on their family history. Many were unable to provide details of their genealogy ? particularly the nine removed from their home at an early age. While these partici- pants had little knowledge of their birth family, others offered poignant details of their family history within the context of residential school ex- perience, child welfare authority, and the impact of these systems on their personal identity. Ned, age 42, was raised by his biological mother and maternal grandmother in an urban centre in Western Canada. His biological father is unknown. Ned struggled to describe his mother?s experience in resi- dential school and identified how it significantly influenced her own be- haviour as a parent: My mother went to residential school and at that time she was, my moth- er was, totally scarred up from residential school ... She did exactly what those people did to her in residential school ... she was abusive. Henry, age 52, was born in Northern Ontario but raised by his mother and paternal grandparents in the northern United States. At an early age, he was made aware of his family?s experience in both Cana- dian residential schools and American boarding schools: She [his mother] don?t like talking about it. Only when she was yel ling at us how rough she had it compared to what we had ... She said it [reside n- tial school] was really strict. The food, the rules, the discipline and no- body cared. She got punished ? whippings and straps ? and they took her away from my grandmother. She was very lonely. She wanted to go home and they wouldn?t let her. Ben attended residential school from age five to 11 in the 1950s. He remembers that most of his extended family attended residential school as well. He illustrated how the resulting isolation from family affected his relationship with his mother upon his return: I used to watch those movies, you know, back then about the kids with their parents ... you know, Leave it to Beaver or something like that, yeah. You know, I saw him hugging his mom, and I tried that once ? tried to hug my mom. And when I hugged her and all that ... actually, I told her I loved her. And she didn?t know how to react. She didn?t know how to take it, you know. So after that, I just shut myself off from her. While only a minority of the participants identified a direct personal experience with the residential school system, nine, or 56 percent, had personal experience with the child welfare system at some point in their lives. Adopted as an infant by a Caucasian family who later had their own biological children, Adam described the emotional disconnect that permeated family life within his adoptive home: The support was lacking in the family. It was a little bit dysfunctional in that aspect ... I don?t know. Being adopted, I think a lot of attention went more to my younger brother and sister who were their natural kids ... I don?t know about, you know, love and being able to talk to somebody, you know, how you?re feeling and whatever. There just wasn?t a lot of that around. Further along in the interview, Adam voiced his frustration with the disconnection from his birth family and the emotional isolation he felt in his adoptive home: You?re adopted as a baby, taken away from your parents, and then you?re in this other setting, and then they disown you, and it?s just like, Christ, it just seems like an ongoing cycle I?m living. Dan, age 40, recalled that before his adoption he had lived with his biological family. He and his younger siblings were removed from their home because of his parents? drinking when Dan was five years old. He was adopted separately into a non-Native family where he was the old- est of three children. His siblings were the biological children of his adoptive parents. Dan described the lack of connection he felt in his adoptive home: I was just there, taking up space?I didn?t love any of these people in this family. A person from age 5 until about 14 or 15 could live with a family and not love anybody in the family. I thought that?s just the way it has got to be. Frank, age 42, commented, without emotion, that he has never felt any connection to the people who raised him. He described himself as particularly independent from an early age: I?m always picking myself up. I?ve never really had no mommy or daddy to run home to ... I got to pick myself up? They were what they were. They were adoptive parents. I?ve never relied on [anyone], whether they?re my adoptive parents or foster parents. I?ve never relied on those people. Pat acknowledged that he has no close connections with the rest of his family or home community. His years in care have left him emotion- ally insecure: I have nobody to really get close to. That?s been a problem for me? When things are really doing good, I feel I really don?t deserve this. Even rela- tionships ? you try to be there for them, but you never could be. More than 40 years later, John is able to recall the poignant details surrounding the removal of a child from the reserve where he stayed with his grandparents each summer: I remember once this kid came running over to [me]? I was fishing on these docks, and he dove in the water and he came up right under where I was standing, because you could breathe a little bit? And the police car comes flying over there? They were looking all over the docks, like u n- der boats and stuff? They asked me and they asked the people hanging around the dock? Well, we said we haven?t seen him . Meanwhile, he was under [the dock]. I could see the bottom of his feet. They were white because he was treading water and just hoping to God they didn?t see him there. And they didn?t. But eventually he ended up going? he got caught? [long silence]. John identified the chronic stress created within the community as a result of witnessing child welfare interventions with other families in the small reserve community: For me, I didn?t see them as being any poorer than me, because I lived in the same conditions as them, and I had no idea why they were ... why they would take them and not [me)? I don?t know how they figured out which kids were ... which ones to pick ... I was angry. I was afraid. Despite acknowledging that the rationale for removing them from their homes at a relatively early age was linked to family violence, alco- holism, or poverty, they felt that this break with their community and subsequent placement in non-Aboriginal foster care, group homes, or adoptive families had detrimental effects on them. In effect, these men were without a home ? or homeless ? from an early age. Over many decades, negative stereotypes of Aboriginal peoples significantly influenced public perception of the value of Aboriginal cul- ture. Several of the men interviewed for this study indicated that they found little support for affirming their cultural heritage from their im- mediate family and from the communities in which they were raised. Issues related to cultural identity were affected by the individual?s phys- ical characteristics, as well as his ?blood heritage,? as demonstrated by having ?status? under the Indian Act. Mike summarized the internal bat- tle waged by many throughout their youth: We didn?t look the same colour as them so we were teased about it. Teased at school? I didn?t want to be an Indian because of the fact that I was getting teased. I wouldn?t hang around them or I?d fight them. I?d tell them off, you know, I?d do everything in my power not to be Indian Many of the men indicated that their current state of living in shel- ters or sleeping outside without permanent shelter is now a fact of life. Ned has lived in Toronto for 16 years, and drug addiction and, on nu- merous occasions, alcohol addiction have led to incarceration. He offered the following description of his own transition into homelessness: Being homeless is actually very easy, you know. It starts, you?re drinking somewhere, the tracks, behind a building, and you pass out. You wake up in the morning and you?ve already spent your first night on the street ... Next thing you know, you?re sleeping there and other people have been there longer before you, and they have their shit together. They have shit stashed everywhere. They know where to go ... I learned how to live on the streets, so there?s no big pressure for me any more to find something different. For the Aboriginal men in this study, their experience with home- lessness appears to be linked to historic social policies. This study?s find- ings suggest that the trauma experienced by generations of Aboriginal peoples has contributed to the experience of homelessness among the men in this study and possibly among other Aboriginal peoples. Much of the research on trauma looks at psychological and social contributors that affect the individual, with an emphasis on family dy- namics (Beisner & Attneave, 1982; Brasfield, 2001). Although there is a level of insight in this type of research, the trauma experienced by the men in this study must be viewed historically. Building on the precepts suggested by Waldram (1997), Kirmayer et al. (2000), and Duran and Duran (1995), the data suggest that indicators of intergenerational trau- ma may exist along four distinct realms: the individual, the family, the community, and the nation. Indicators arising from the data analysis have been isolated for each of these realms and are summarized below. Individual Indicators Individual indicators emerging from the data are: ? lack of a sense of ?belonging,? identification, or affiliation with a specific family, community, culture, or nation; ? feeling of ?abandonment? by caregivers; ? limited or no information about one?s culture of birth, including lan- guage, customs, belief systems, spirituality; ? one or more ?flight? episodes from a caregiver environment as a youth; ? inability to sustain personal or intimate relationships; ? being present-oriented, not future-oriented; ? low self-esteem; ? limited education and employment history; ? history of substance misuse; ? history of involvement with the criminal justice system, precipitated by substance misuse; ? involvement with the mental health system. Rather than pathologizing the individual, as is often done, I would argue that these indicators should be viewed as resulting from a histori- cal process. It is important that these individual indicators of intergene- rational trauma be considered in relation to the indicators for family, community, and nation. Family Indicators Family indicators emerging from the data include: ? chronic or episodic family violence, including physical, sexual, emo- tional, and/or verbal abuse ?children by adults in the household; ? lack of emotional bonding between parents, siblings, and extended family members; ? denial of cultural heritage by older family members; ? perpetuation of negative stereotypes within the family of birth or caregiver environment; ? irregular contact or the absence of contact with caregiver family members; ? unconcealed and rampant alcohol and drug misuse that crosses gen- erations. Community Indicators Additional factors that may be present in the community and that influ- enced the individual?s early life history must also be considered: ? unconcealed alcohol and drug misuse among community members; ? lack of cultural opportunities, including transmission of language skills, history, traditional values, and spirituality; ? unwillingness to ?reclaim? community members; ? low levels of social capital (Putnam, 2000), including trust, reciprocal helping relations, and social engagement. It is important to recognize that within traditional Aboriginal cul- ture, the community?s support is critical for the development of individ- uals and families. Holistic healing is not achievable without the influence and guidance of a balanced and healthy community. These findings sug- gest that for Aboriginal peoples, the definition for ?homelessness? must emphasize the breakdown of community structures. Therefore, I propose a definition of homelessness that is more relevant to Aboriginal peoples: homelessness is a condition that results from individuals being displaced from critical community social structures and lacking stable housing. Nati onal Indicators The data also indicate that a fourth element must be considered. The in- dividual, family, and community are embedded within national struc- tures that both historically and contemporaneously have had a profound impact on these other institutions. Some key national indicators that may contribute to homelessness include: ? popularization of negative stereotypes through mainstream media; ? social policies that perpetuate colonialization of Aboriginal peoples on an individual, family, and community basis; ? lack of support for holistic programs and services targeting Abori- ginal needs; ? lack of support for community self-determination. The impact of trauma on the Aboriginal nation must also be recon- ciled. The need to support the development of community beyond geo- graphic boundaries to include all Aboriginal peoples is critical to the healing process. The data provided have identified how external social policies have cor- roded the links between critical elements within Aboriginal culture. In- dividual, family, community, and nation now exist in isolation of one another. Social policies, including the Indian Act, the residential school system, and child welfare legislation, have systematically negated Ab- original culture and imposed values that are contradictory to our tradi- tional ways of relating to one another (Cross, 1986; Good Tracks, 1973; Proulx & Perrault, 2000). The colonizing impact of these policies has re- sulted in many individuals experiencing ?social anomie? ? a feeling of disconnection from a particular cultural group (Brant, 1990). The perva- siveness of this condition has left the Aboriginal nation in a similar state ? unable to draw upon common bonds to bring individuals, families, and communities together. The Intergenerational Trauma Model The indicators discussed above are integrated within the Intergenera- tional Trauma Model. The model is premised on the main constructs of the traditional teachings of the Aboriginal medicine wheel, a conceptual process that frames our understanding of the world as Canada?s First Peoples: The teachings assume that all humans can exist in balance with them- selves, their families, communities, and their natural surroundings. Where alcoholism, violence, abuse, or any kind of dysfunction exists, there is imbalance: the dark side domination (Nabigon & Mawhiney, 1996, p. 19). The medicine wheel breaks the main constructs of life into four ele- ments, generally referred to as the four directions: east, south, west, and north. There must be harmony between the four elements of life for bal- ance to be achieved (Morrisseau, 1998). Similarly, the Intergenerational Trauma Model is predicated on the assumption that public policies have disrupted relations between the four systems and the resulting trauma has incubated negative social conditions for Aboriginal peoples, making them significantly more vulnerable to a number of threatening condi- tions, including homelessness. This has disrupted the balance of the wheel in which the individual, family, community, and nation exist. The Intergenerational Trauma Model identifies risk factors that may contri- bute to Aboriginal people?s homelessness. Given the limitations of the study, however, it is premature to say how many indicators need to be present to determine the likelihood of an individual?s becoming home- less. Intergenerational Trauma Model Starting outside the larger circle in the figure, the influence of public policy is identified through the Indian Act residential school system, and child welfare authorities. These social policies are external elements that have penetrated traditional Aboriginal culture and caused change to oc- cur within the traditional social systems, as illustrated by the four small- er circles. The large outer circle represents Aboriginal culture, and the four smaller circles represent the four subsystems of individual, family, community, and nation. The four subsystems exist within a permeable boundary signalled by the broken line of the outer circle. The influence of external elements, such as public policy, has weakened the role of cul- ture in supporting the inner circles of individual, family, community, and nation. The proposed indicators of intergenerational trauma are noted in the centre of each of the four inner circles. The circles representing the individual, family, community, and nation exist in isolation from one another. If they operated as an adequate support system, they would intersect, but within the Intergenerational Trauma Model, they do not, symbolizing that they are not able to support one another. The balanced existence between the four systems is thwarted by the pervasive pres- ence of intergenerational trauma, which has prevented the four ?sys- tems? from re-establishing their former balanced and linked existence and in effect is the barrier that thwarts the reunification of the four sys- tems. As indicated by the arrows in the figure, intergenerational trauma pushes the four inner circles apart. Although the impact of the trauma may be most visible in the individual, a holistic approach ? as presented in this model ? suggests that trauma affects all the four spheres. The homeless men in this study operated without a support system from their family, community, or nation ? in fact, as discussed, they were alienated from these entities. The cause of the problems that they experienced as individuals could be linked to these other systems, and once they were homeless, these other systems did not provide the sup- port the men needed. A minimal form of support came from the com- munity through Aboriginal shelters such as NaMeRes, but the problems experienced by the men were so severe that most often it was ?too little, too late.? The men remained alienated from their families, communities of birth, and the Aboriginal nation. The wounds they had suffered through these other systems were deep, and therefore the countervailing influence of a shelter and its services were insufficient to alter their life dynamics. Limitations of the Research This was an exploratory study with a limited number of participants, and a single site was used to obtain study participants. As a result, the study precluded a cross-section of homeless Aboriginal men who, for a variety of reasons, do not use shelter services specific to the Aboriginal population. This study was limited to Aboriginal men using services in the Toronto community, and the data were collected over a six-week pe- riod in the late winter and early spring of 2003. Seasonal use of shelter services may have affected the study sample. I did not use a non-Aboriginal comparison group to identify differ- ences in their experiences. My concern with research on homelessness is the assumption of a generalized ?homeless? population. I wanted this study to focus on the experience of Aboriginal men, as I believe their is- sues should not be compared to those of another group. I believe their experiences warrant singular attention and should not be weighed against the experiences of other ?subpopulations.? The systemic issues identified in this study have affected Aboriginal women differently. In addition, there are broader social issues that com- pound an understanding of homelessness for women and youth. As such, these results cannot be generalized across the Aboriginal popula- tion and, in particular, to homeless Aboriginal women or homeless Ab- original youth. None of the information provided by the participants was verified through other sources, such as interviews with family members, First Nation administration records, or the justice system. I considered this approach, but given the social isolation of the men in the study, it would have been extremely difficult to access their social networks, to the ex- tent that these even existed. However, as noted above, a focus group of other Aboriginal men who used the same hostel services was conducted at a later date and confirmed the interpretation of the study results. Recommendations For more than two decades, significant research has been undertaken in Canada and across North America on the growing crisis of homelessness in our urban centres. For the most part, researchers have focused on the personal or structural factors that have contributed to this phenomenon in our modern society. While this research is valuable, with the exception of a small number of studies it does not capture the uniqueness of home- lessness as experienced by Aboriginal men. In addition to involving homeless Aboriginal men, the current study interprets homelessness from the ?social lens,? or world view, of an Aboriginal researcher. It is meant to expand the discussion of homelessness and promote the value of undertaking culturally congruent social research. The data from the study suggest that there may be significant bene- fits to engaging in further research in the area of intergenerational trau- ma, and specifically to testing the utility of the lntergenerational Trauma Model. A more representative population of Aboriginal peoples across Canada, including women and youth, would provide further insight into how useful the indicators are in determining the impact of intergenera- tional trauma as an explanation for homelessness in our urban centres. Provincial and municipal authorities must consider the need for cultural- ly appropriate support services in urban centres if homelessness is to be effectively addressed. It would also be interesting to assess the usefulness of the proposed model in helping to understand other social issues in which Aboriginal peoples are overrepresented, such as family violence, substance misuse, suicide, and involvement with the justice system. The model can assist us in understanding whether intergenerational trauma is contributing to the incubation of negative social conditions for Aboriginal peoples. The data suggest that focusing on individual causation factors will not effectively address the needs of Aboriginal peoples. A holistic ap- proach, such as in the Intergenerational Trauma Model, is required to assess the impact of intergenerational trauma across the four identified systems. Although research is available on the impact of residential schools, child welfare authorities, and the Indian Act on the individual systems (individual, family, community, nation), little research docu- ments the cumulative impact of public policy on the relationship among all of these four systems. This includes assessing how relationships be- tween individuals and their caregivers, individuals and their birth com- munities, and individuals and their culture (Aboriginal nation) have been affected by public policy. The trauma experienced by the men who participated in this study signals a need for more holistic program responses. While most of the recent literature on counselling for Aboriginal peoples focuses on devel- oping approaches that are culturally relevant and culturally congruent (Cross, 1986; McKenzie & Morrissette, 1993; Morrisseau, 1998; Morris- sette et al., 1993), there is no call for approaches that provide for lifelong treatment modules in a continuum-of-support approach. Long-term support must be incorporated into the current range of responses, in- cluding housing, health, and social programs. These supports must be offered to other family and community members as well. Further, it must be recognized that, given the pervasive influence of trauma in our lives, it may take more than one generation to heal from experiences with a lengthy history. There is no ?quick fix? for these men. Solutions to homelessness that focus only on helping people meet their personal needs are not likely to effectively address the underlying causes of homelessness as experienced by the participants in this study. Some preliminary health data on the rates of youth suicide in Aboriginal communities suggest that there may be a strong correlation between programs that promote positive self-image and community well-being and reduced suicide rates in Aboriginal communities (Chandler & La- londe, 1998). Such programs need to be expanded to include Aboriginals living in urban centres. For Aboriginal peoples, the solution to home- lessness is not necessarily the construction of housing; rather, the re- sponse also requires a holistic approach that reconstructs the links be- tween the individual, family, community, and Aboriginal nation. There have been several attempts to promote the need for public policy and programs that are culturally congruent (Congress of Abori- ginal Peoples, 2001; Maidman, 1981; Morrissette et al., 1993; National Aboriginal Housing Committee, 1993). The need is paramount for pro- grams that support the development of Aboriginal peoples and allow us, as a community, to enhance our knowledge and skills so that we can provide support to community members through health care services, education, public policy, the justice system, and other important ele- ments of our nation?s social infrastructure. Dr. Peter Menzies is Clinic Head for Aboriginal Services at the Centre for Ad- diction and Mental Health, Assistant Professor, Department of Psychiatry, University of Toronto, and Adjunct Professor at the School of Social Work, Lau- rentian University. He also lectures at Ryerson University and the Toronto Hostel Training Centre. He is a member of Sagamok Anishnawbek First Nation and sits on various Aboriginal community boards and committees both at the local and national level. His research interests include Aboriginal homelessness, intergenerational trauma, child welfare, suicide, addiction and mental health. He has published articles regarding Aboriginal health issues, and is a member of the Ontario College of Social Workers and Social Service Workers. References Andres, R. (1981). The apprehension of Native children. Ontario Indian, 46, 32-37. Assembly of First Nations (1994). Breaking the silence. Ottawa: Author. Beisner, M., & Attneave, C. (1982). Mental disorders among Native American children: Rates and risk periods for entering treatment. American Journal of Psychiatry, 139(2), 193-198. Brant, C. (1990). Native ethics and rules of behaviour. Canadian Journal of Psychia- try, 35(3), 534-539. Brant, C. (1993). Suicide in Canadian Aboriginal peoples: Causes and prevention. Dis- cussion paper prepared for the Royal Commission on Aboriginal Peoples. Brasfield, C. R. (2001). Residential school syndrome. BC Medical Journal, 43(2), 78- 81. Braveheart-Jordan, M., & De Bruyn, L. (1995). So she may walk in balance: Inte- grating the impact of historical trauma in the treatment of Native American Indian women. In J. Adelman & G. Enguidanos (Eds.) Racism in the lives of women: Testimony, theory and guides to ethnoracist practice (pp. 345-369). New York: Haworth Press. Cianci, D., & Nadon, S. (1991). Walking the medicine wheel path: An exploration of the medicine wheel teachings on human relationships. Owen Sound, ON: Maplestone Press. City of Toronto. (1999). Report of the mayor?s homelessness action task force: Taking responsibility for homelessness, an action plan. Toronto: Author. Chandler, M., & Lalonde, C. (1998). Cultural continuity? as a hedge against sui- cide in Canada?s First Nations. Transcultural Psychiatry, 35(2), 191-219. Claes, R., & Clifton, D. (1997). Needs and expectations for redress of victims of abuse at Native residential schools. Ottawa: Law Commission of Canada. Congress of Aboriginal Peoples (2001). Programs and issues: Urban Native housing. Retrieved February 3, 2005, from www.abo- peoples.org/programs/housing.html. Couchi, C., & Nabigon, H. (1994). A path towards reclaiming birth culture. In F. Shroff (Ed.) The new midwifery (pp. 41-50). Toronto: LPC Inbook. Creswell, J. (1998). Qualitative inquiry and research design: Choosing among five tradi- tions. Thousand Oaks, CA: Sage Publications. Cross, T. (1986). Drawing on cultural tradition in Indian child welfare practice. Social Casework, 67(5), 283-289. Dey, I. (1993). Qualitative data analysis. New York: Routledge. Duran, E., & Duran, B. (1995). Native American postcolonial psychology. New York: State University of New York Press. Duran, E., Duran, B., Yellow Horse-Nave Heart, M., & Yellow Horse-Davis, S. (1998). Healing the American Indian soul wound. In Y. Danieli (Ed.) Inter- national handbook of multigenerational legacies of trauma (pp. 341-354). New York: Plenum Press. Frideres, J. (1998). Aboriginal peoples in Canada: Contemporary conflicts. Toronto: Prentice Hall, Allyn and Bacon Canada. Gagne, M. (1998). The role of dependency and colonialism in generating trauma in First Nations citizens. In Y. Danieli (Ed.) International handbook of multige- nerational legacies of trauma (pp. 355-372). New York: Plenum Press. Good Tracks, J. G. (1973). Native American non-interference. Social Work, 18(11), 30-34. Graveline, R. (1998). Circle works: Transforming Eurocentric consciousness. Halifax: Fernwood Publishing Co., Ltd. Hodgson, M. (1990). Impact of residential schools and other root causes of poor mental health. Edmonton: Nechi Institute on Alcohol and Drug Education. Holt, J. (1993). How about evaluation: A handbook about project self-evaluation for First Nations and Inuit communities. Ottawa: Department of National Health and Welfare, Medical Services Branch. Johnston, P. (1983). Native children and the child welfare system. Ottawa: Canadian Council on Social Development. Kendrick, M. (1990). Nobody?s children: The foster care crisis in Canada. Toronto: Macmillan Canada. Kirmayer, L., Brass, G., & Tait, C. (2000). The mental health of Aboriginal peoples: Transformations of identity and community. Canadian Journal of Psychiatry, 45(7), 607-616. Lederman, J. (1999). Trauma and healing in Aboriginal families and communi- ties. Native Social Work Journal, 2(1), 59-90. Locust, C. (1999). Split feathers: Adult American Indians who were placed in non-Indian families as children. Pathways, 14(1), 1-5. McKenzie, B., & Morrissette, L. (1993). Cultural empowerment and healing for Aboriginal youth in Winnipeg. In A.-M. Mawhiney (Ed.) Rebirth: political, economic and social development in First Nations (pp. 117-130). Toronto: Dun- durn. Maidman, E. (1981). Native people in urban settings: Problems, needs and services. Toronto: Ontario Task Force on Native People in the Urban Setting. Mawhiney, A.-M. (1994). Towards Aboriginal self-government: Relations between status Indian peoples and the government of Canada, 1969 -1984 . New York: Gar- land. Morrisseau, C. (1998). Into the daylight: A holistic approach to healing. Toronto: Uni- versity of Toronto Press. Morrissette, V., McKenzie, B., & Morrissette, L. (1993). Towards an Aboriginal model of social work practice. Canadian Social Work Review, 10(1), 91-108. Mussell, W., Nicholls, W., & Adler, M. (1991). Making meaning of mental health challenges in First Nations: A Freirian approach. Chilliwack, BC: Sal?i?shan In- stitute Society. Nabigon, H., & Mawhiney, A.-M. (1996). Aboriginal theory: A Cree medicine wheel guide for healing First Nations. In F. Turner (Ed.) Social work treat- ment (pp. 18-38). New York: The Free Press. National Aboriginal Housing Committee. (1993, June 3). First our lands, now our homes: A response to the urban and rural Native housing crisis created by Canada?s federal budget cutbacks. Brief presented to the Royal Commission on Aboriginal Peoples. Neuman, W. L. (1997). Social research methods: Qualitative and quantitative approach- es. Toronto: Allyn and Bacon. Padgett, D. (1998). Qualitative methods in social work research: Challenges and re- wards. Thousand Oaks, CA: Sage Publications. Patton, M. Q. (1990). Qualitative evaluation and research methods (2nd ed.). New- bury Park, CA: Sage Publications. Proulx, J., & Perrault, S. (2000). No place for violence: Canadian Aboriginal alterna- tives. Halifax: Fernwood Publishing Co., Ltd. Putnam, R. (2000). Bowling alone: The collapse and revival of American community. New York: Simon and Schuster. Richard, K. (1989). Kenn Richard fights racism within child welfare. Metropolis, 2(11), 3-4. Royal Commission on Aboriginal Peoples (1996). Volume 1: Looking forward, look- ing back. Ottawa: Supply and Services Canada. Royal Commission on Aboriginal Peoples (1996). Volume 3: Gat hering strength. Ottawa: Supply and Services Canada. Royal Commission on Aboriginal Peoples (1996). Volume 4: Perspectives and real i- ties. Ottawa: Supply and Services Canada. Rubin, A., & Babbie, E. (1997). Research methods for social work. Scarborough, ON: Brooks/Cole. Statistics Canada. (2003, January 21). The Daily. Ottawa: Author. Retrieved Feb- ruary 3, 2005 from http://www.cans.ats.org/readdetail.asplid=473 Timpson, B. (1990). Indian and Native special status in Ontario?s child welfare legislation. Canadian Social Work Review, 7(1), 49-68. Tutty, L., Rothery, M., & Grinnell, R., Jr. (1996). Qualitative research for social work- ers. Toronto: Allyn and Bacon. United Church of Canada (1994). Residential schools booklet. Toronto: Author. Waldram, J. (1997). The Aboriginal peoples of Canada. In I. Al-Issa & M. Tousig- nant (Eds.) Ethnicity, immigration and psychopathology (pp. 169-187). New York: Plenum Press. Warry, W. (1991). Ontario?s first people: Native children. In L. Johnson & D. Barnhort (Eds.) Children, families and pubic policy in the 90s (pp. 207-269). To- ronto: Thomson Educational. Yellow Horse-Brave Heart, M. (2003, January). Historical trauma. Paper presented at the National Aboriginal Health Organization, Ottawa. Chapter 6.3 Hidden Homelessness among Aboriginal Peoples in Prairie Cities JIN O DISTASI O , GINA SYLVESTRE, AND SUSAN MULLIGAN This research examined hidden homelessness among Aboriginal persons in prairie cities. In particular, we gathered data in Winnipeg, Saskatoon, and Regina on the shelter circumstances of persons precariously housed in tenuous situations, including those who lived temporarily with friends or family or those who resided in short-term accommodations such as shelters, rooming houses or hotels. The study was governed by a regional Steering Committee that helped establish the context and purpose of the data collection methods. It was also recognized that researchers needed to be mindful of the rich- ness of Aboriginal culture and diversity in Prairie cities. A respected Elder was invited to provide support. The Steering Committee also sug- gested that the study commence by undertaking informal discussions with hidden homeless persons so as to better understand the core issues facing persons presently finding themselves in housing distress. These sessions helped researchers understand and set the course for the formal survey process. Where possible, community-based researchers were used to assist in the gathering of knowledge from participants and sitting on the Steering Committee and in establishing relationships within each of the study cities. Following the more structured surveys that were undertaken, a tra- ditional Talking Circle was held in Winnipeg. Led by a respected Elder, this session was used to connect participants together and to respect In- digenous methodological approaches. A second set of interviews were then held with service providers in each of the three cities to better understand their perspectives in dealing with housing distress. The results of the study continue to reinforce the need for additional housing and supports to those currently struggling to find adequate and affordable shelter. Perhaps a key finding was also the nearly 20 percent of participants who indicated they had a seasonal connection to their home communities. This group represented what the Australian litera- ture refers to as ?Spiritual Homeless.? It is our contention that this group remains unique among those in various levels of housing distress. However, given that just under 45 percent of the participants indi- cated they moved more than three times in the last six months is also an indication of the high level of residential instability among participants. For their part, service providers recognize the enormity of the situa- tion and do their best to cope with a system bursting at the seams with respect to being able to deal with the crushing need to provide shelter and supports to a range of persons who move for a variety of reasons. With respect to mobility, it remained clear that the movement of people is due to a range of factors that this study can only begin to understand. Perhaps it is critical to reiterate that the ?homeless? are by far a he- terogeneous group and they remain equally complex and diverse, rang- ing from the absolute and visible homeless to the more invisible ?hidden homeless.? But these are merely broad descriptors that do not capture the uniqueness of the urban Aboriginal communities in Winnipeg, Regi- na and Saskatoon, and in particular, those in housing distress. The simple but powerful words of one participant sum up much of the research in saying, ?home is where the heart is? and right now that is nowhere.? Analysis of Surveys Social support helps motivate me to work, to pick myself up when people put me down, to keep going on with life, even when I don?t want to any- more. This section evaluates the results of the survey in relation to the socio- demographic characteristics of the sample, to their housing circums- tances and sources of social support. The discussion includes a consider- ation of participants? experiences of discrimination, their sources of so- cial support, their participation in organizations, and their connection to the reserve. Given the lack of understanding regarding the issue of hid- den homelessness, the intent of this section is to describe the general housing circumstances and characteristics of First Nations peoples in housing distress so as to better understand the factors that have contri- buted to their present shelter situation. Residential instability Just over 40 percent of individuals in the sample had lived in more than three locations in the six-month period prior to the survey. This high de- gree of movement further substantiates the internal residential churn among Aboriginal persons that was introduced in the literature review. Almost one-half (47.2 percent) of the respondents expressed some level of apprehension about remaining in their respective city on a per- manent basis. The residential instability experienced by participants in this study raises significant concerns with respect to both overall affor- dability and availability of housing in Canadian Prairie cities. As home- lessness is characterized by high mobility, an important element to eva- luate is the multiplicity of temporary accommodations in which individuals reside for varying periods of time (Springer, 2000). Over half (55.8 percent) of sample members were in relatively stable residential situations as they reported staying in only one or two temporary resi- dences during the period. In contrast, the remainder of the sample mem- bers was considerably more mobile, as they reported residence in three or more accommodations. The high frequency of movement suggests that this sub-sample experienced considerably more residential instabili- ty. Thus, the following overview compares sample members experienc- ing relative residential stability (one or two temporary accommodations) and those subjected to greater instability (three or more temporary ac- commodations). Demographic and Economic Characteristics The highest proportion of our sample was male (55.8 percent) and under the age of 30 years old (47.5 percent). In relation to residential stability, a higher proportion of males and older sample members reported living temporarily in only one or two accommodations in the six months prior to the survey. In relation to Aboriginal identity, almost three-quarters of the sam- ple (71.7 percent) reported being part of the First Nations of Canada, while 28.3 percent reported M?tis ancestry. While one-half of First Na- tions respondents indicated residence in three or more temporary ac- commodations during the previous six months, a greater proportion of M?tis respondents reported relative residential stability. The overwhelming majority (68.6 percent) of those experiencing hidden homelessness were unemployed, while 17.7 percent were em- ployed in some capacity and 13.7 percent were students. Approximately three quarters of those who were unemployed indicated that they had received some form of social assistance. Furthermore, in order to better understand the circumstances of these individuals, it is important to con- sider other strategies they utilized to earn money. For example, over 20 percent of the sample indicated that they were involved in activities in the informal sector.6 ?I clean up every day and watch the children while the mom goes out all night. I also get food requisitions and I?m like the main supplier for food. I would also sell myself for sexual favours just to get money for food, smokes and whatever else we needed.? (28-old fe- male respondent). Generally speaking, the informal sector includes such activities as the drug trade, etc. One-quarter (24.8 percent) of the sample reported an education lev- el below grade 9, while over one-half (57.1 percent) of the respondents had obtained some level of high school education, although they had not obtained a high school certificate. As for income, three-quarters of the sample received less than $10,000 annually. Specifically, 55.2 percent of the sample reported an income of less than $10,000 annually and 19.8 percent reported no income at all. These indicators of socio-economic well-being emphasize the mar- ginalization of Aboriginal persons experiencing housing distress. None- theless, 63.4 percent of the sample indicated that they were optimistic that their economic future would at least get slightly better. It is interest- ing to note that respondents who reported being employed and had little or no income but were optimistic about their economic future were more likely to have resided in three or more temporary accommodations in the past six months. Conversely, greater residential stability was expe- rienced by those respondents who were unemployed. Housing circumstances A key piece of the housing analysis of the study probed the use of emer- gency shelters or places where people went when there were no other options available. Thirty percent of the sample indicated that they had used an emergency shelter in the last year. Respondents who expe- rienced greater residential instability were more likely to use shelters. Similarly, those who had resided in three or more accommodations in the past six months reported using shelters more frequently than indi- viduals who were in more stable housing circumstances. The over- whelming majority of the sample indicated that they were treated fairly and with respect by both agency staff (56.9 percent) and other individu- als staying at the shelter (61.5 percent) when accessing services. The comments of one person summed up the experience by stating ?yes, we were in the same boat and were there for each other.? The use of emergency shelters by 30 percent of the sample points to the ongoing need to provide supportive shelter services, especially in cold climate cities. In fact, one respondent noted that there was a need for both additional units and a general public awareness of existing re- sources and stated that government should ?make more places into homeless shelters and make the public know more about shelters.? Other respondents were positive about the treatment they received with one persons stating that staff were ?kind, gentle, caring people who help people in time of trouble.? The Aboriginal community is complex, characterized by a high de- gree of differentiation among various groups (e.g., First Nations, M?tis and Inuit) and geographic location (e.g., rural, northern, remote and on and off reserve). To explore the complexity of connections to home communities, respondents were asked whether they had ever lived on reserve. We found that 62.5 percent of the sample had lived on reserve, with a slightly higher proportion of those respondents in greater residen- tial instability having lived on reserve previously. Reasons given most often for leaving the reserve included the desire to access educational and employment opportunities, as well as better housing. We asked whether respondents lived in their home community on a seasonal basis. The overwhelming majority (81.7 percent) of the sample replied that they did not move seasonally. Nevertheless, 18.2 percent of respondents indicated that they indeed had a connection to their home community. Furthermore, although similar proportions of both the residentially stable and unstable sub-groups did live in the city seasonally, those res- pondents who reported residing in three or more temporary accommo- dations in the past six months visited their reserve community with greater frequency. It is important to recognize that those individuals who move on a seasonal basis undoubtedly represent a key segment of the population that will move periodically between urban and home com- munity. When asked about the seasonal aspect of living in the city and home community, those who offered comments noted that friends and family in the home community contributed to their decision to move between places. With respect to the current housing situation, 69.3 percent of the sample indicated that they presently lived in an apartment, row or single detached home. The remainder of the sample was housed in rooming houses, single room occupancy hotels or other transitional housing units. Those respondents who reported a greater number of temporary resi- dences were more likely to be housed in accommodations consisting of a single room. Data related to length of time in a temporary accommodation is an indicator of the degree of residential instability. Almost all respondents indicated that they had resided at the current address for less than one year. Furthermore, almost three-quarters (73.8 percent) of the sample had lived at the current location for less than six months. Those residing in fewer temporary accommodations were more likely to report a longer average length of time at each residence, as well as a longer period of time at the current temporary residence. In contrast, those respondents experiencing greater residential in- stability reported shorter periods of residency in each temporary ac- commodation. Almost 80 percent of the sample had lived in the city for over a year at the time of the survey. In addition, those respondents reporting fewer temporary accommodations also reported longer residency in the city. And while well over half of the sub-sample of those in relative stable residential accommodations indicated that they planned to remain in the city permanently, more respondents experiencing greater residential in- stability were unsure of their future plans. In relation to residential quality, a series of questions was posed to examine the general condition of the present shelter, perceptions of crowding and overall satisfaction with the shelter. With respect to the general condition of the current residence, 40.5 percent of the sample indicated that their shelter required some repairs, while 23 percent of respondents felt their current shelter was in poor condition and needed significant attention to improve the unit. In relation to perceptions of crowding, almost half (47.9 percent) of the sample did not consider their current residence to be crowded. In contrast, 32.5 percent of respondents were in somewhat crowded condi- tions, while 19.7 percent indicated that they were living in very crowded conditions. When we asked about the overall satisfaction of respondents in their current shelter, only a small proportion (10.9 percent) of the sample was very satisfied with current accommodations. The high proportion of respondents who were only somewhat satis- fied or were unsatisfied with their shelter emphasizes the poor condi- tions in which Aboriginal persons experiencing housing distress must endure. It is also notable that a slightly higher proportion of those res- pondents who had changed residences at least three times in the past six months were in very crowded conditions and were unsatisfied with their shelter. From the brief review above, it becomes clear that there is some lev- el of uncertainty about shelter and that satisfaction and condition varied among participants. To examine these issues in more detail a series of questions were posed to examine the level of use of subsidy programs among participants. These questions included asking whether persons had accessed subsidized housing, whether they were on a waiting list and whether they were aware of existing programs. A second and re- lated set of questions then asked whether there were any problems when seeking subsidized housing with respect to their treatment and general experiences. Overall, the use of subsidized housing was low among participants with only 22.7 percent indicating that they had previously accessed sup- portive housing. (When examining the income distribution of the sam- ple, it was apparent that while few ?accessed? subsidized housing, many would benefit because of their low income levels.) Furthermore, only 18 percent of the sample was currently on a wait list, while just over 15 per- cent stated that they had been denied subsidized housing when applying in the past. With respect to those who had applied but were placed on a wait list, over one third (35.7 percent) indicated that they had been wait- ing for more than a year. When asked to explain their experience in applying and trying to secure subsidized housing, comments ranged from persons feeling that they were mistreated to those who were completely satisfied with the experience. One person stated they were unsure of the process, writing ?I don?t know how to go about it, and I just never hear back from the housing company,? while another said, ?Winnipeg regional housing takes forever to answer back?six month to have the first interview.? Another person felt that being single posed a barrier to accessing subsi- dized housing, ?I have never applied because they give priority to people with children and families?never applied too worried about not getting in and getting my hopes up.? Interestingly, many suggested that subsidized housing was only given to persons who were employed or had families and children under their care. When asked about the barriers they experienced, most com- mented on not having proper references or the necessary deposits to se- cure a place. Others also raised concerns about discrimination and mi- streatment by landlords who profile perspective tenants. In an effort to probe the living situations of respondents in more de- tail, we asked a series of questions examined the temporary nature of each person?s housing situation. At present, 75 percent of the sample in- dicated that they were currently living temporarily with friends and or family. Most important, was that 81.5 percent of persons staying in a temporary accommodation with either friends or family indicated that they contributed to the household in a variety of ways and that if they were to leave this accommodation, 35 percent felt this would pose a hardship for the household. When asked how persons contributed to the household, many included both financial and non financial elements such as doing chores, providing childcare and basically helping out around the home. Others noted that they contributed money on a fre- quent (monthly) or infrequent basis (when they had the ability). From this brief overview, it is clear that the housing circumstances of those presently considered part of the hidden homeless population poses a challenge. On the one hand, many who are currently living with friends or family contribute to that household through a variety of means, with many feeling that their departure would put undue stress should they leave. However, it also became clear that the living arrangements are te- nuous, with many moving quite frequently over a short period of time. Ultimately, supportive housing is needed and must recognize the hard- ship of providing deposits and potential reference checks, especially for those just moving into an urban centre. With a high number on waiting listings, there is a need to continue to provide not only access to units and housing but also to get the necessary information out to persons moving into urban centres about the programs and supports currently available to assist (both on a short term emergency basis and a long term permanent solution). Support and Hidden Homelessness We evaluated support with respect to respondents? experiences of dis- crimination, social support, participation with organizations and support from the reserve. First, the survey included a question regarding the respondents? ex- periences with discrimination and unfair treatment. It is important to note that while this question was not central to the study, the partici- pants expressed that various forms of unfair treatment were pervasive across all sectors. Well over half of the participants (60 percent) articulated the view that acts of racism and discrimination affected their daily existence in the urban setting. Similar proportions of males (61 percent) and females (59 percent) replied that they experienced some form of discrimination and unfair treatment. It is notable that compared to those in relative residential sta- bility, a higher proportion of respondents who moved frequently expe- rienced discrimination. Examples were provided of the systematic or institutional discrimi- nation experienced by Aboriginal persons in their encounters with hous- ing organizations, government agencies, and potential employers. One male stated that he preferred Aboriginal run housing organizations be- cause he ?would not feel so discriminated against.? In terms of possible employment opportunities one person claimed that ?I feel discrimination by not getting jobs because of being Aboriginal? while another person felt that their appearance was a factor, ? Yes, there?s a lot of discrimina- tion in this city, most won?t hire you because of your appearance.? However, not all comments were related to racial discrimination as some participants felt discriminated against by housing organizations because they received social assistance, yet others felt employers did not treat them fairly due to a lack of a high school certificate or a ?poor work history.? For the purpose of this study, social support was defined to include emotional guidance and encouragement, mentoring and net- working (being told about a job opportunity etc.) and consisted of rela- tionships with friends, relatives, neighbours, professionals, as well as community organizations. Social support could mean providing child- care or support for an elder in the home. When asked what social sup- port meant, the comments of survey respondents ranged from ?co- dependency? to ?accessibility to resources within the community wheth- er it be friends, family or food banks.? Over three-quarters (77 percent) of the sample had some form of social support. It is interesting that while 75.6 percent of the sample indicated that they had the support of family, only 66.1 percent of respondents had the support of friends. It is also significant that a lower proportion of respondents who moved three or more times in the six months prior to the survey had some form of social support. Moreover, while similar proportions of the residentially stable and unstable had the support of family, a greater proportion of those moving more frequently had the support of friends. I?d like to see more Aboriginal organizations and/or persons providing advocacy services on behalf of other Aboriginals, especially for single moms like myself. I?d like to have more support from a native worker to help me adjust to the city and the areas that I find myself in, someone to do home visits, and possibly a small group of women like myself, I can hang around with, who have the same concerns and feelings like myself. (24 year old female respondent). Many participants spoke of having social support from family through offering a temporary place to stay or simply congratulating them on successes and informing them of different opportunities that may interest them or help them to further succeed in their endeavors. One person expressed a lack of support due to the size of his family, ?no family, except one brother, and he has a family, I am a grown man and pride gets in the way.? He further stated that, ?men have pride, don?t want to take help when they know they need it, agencies should be dis- creet and compensate for this.? Despite this viewpoint, this individual has social support through friends and calls them, ?if I need a place to stay.? The overwhelming majority (81.5 percent) of those persons staying in a temporary accommodation with either friends or family indicated that they contributed to the household in a variety of ways. If they were to leave this accommodation, 35 percent felt this would pose a hardship for the household. When asked how persons contributed to the house- hold, many included both financial and non financial elements such as doing chores, providing childcare and basically helping out around the home. Others noted that they contributed money on a frequent (monthly) or infrequent basis (when they had the ability). In addition to the availability of social supports, we considered the involvement of survey respondents in community organizations, as this may also reflect a form of support for individuals experiencing hidden homelessness. Over half (55.1 percent) of the sample did participate in organizations. Those respondents experiencing relative residential stabil- ity were more likely to be involved in two or more organizations. In con- trast, a greater proportion of the sub-sample that moved more frequently were involved in only one organization and participated in the activities of the organization less often. In relation to the experience of support for Aboriginal persons in housing distress, it is also important to consider whether support was received from the reserve or home community. Only 10 percent of the sample received financial support from their home community. Almost 90 percent of the sample did not receive financial support with little dif- ference between the residentially stable and unstable groups. In addition to financial support, respondents were asked whether they were in- volved with their band or community. Approximately 20 percent of the sample did indicate that they were involved with their home community with a slightly higher proportion of involvement reported by those in more unstable residential situations. The purpose of this section was to examine the support system for those who experience the hidden homeless condition. Often people with no fixed address or living in unstable situations have lost contact with family, friends and have little or no support networks. The challenges that relate to their uncertain circumstances of day-to-day life includes the exclusion of established community networks. Unfamiliarity with exist- ing networks and agencies can result in frustration, inability to find the right supports and isolation from the rest of the community. Many people do not want to be part of a ?culture of dependency? and want to find ways to contribute and engage in their communities. Personal connections with friends and family can be critical when seek- ing some basic needs such as food and clothing while reaching stability in housing, seeking employment and education. It was revealed during this study that a majority of the participants have social supports, whether it is from service providers, home communities or family and friends. Service Providers The two most important issues facing Aboriginal persons in finding ade- quate shelter were the lack of financial resources and availability of shel- ter in general. Data were collected by interviewing 60 key personnel involved in the provision of services in Winnipeg, Regina and Saskatoon. The selec- tion of service providers interviewed ranged from a diverse set of organ- izations in each of the three prairie cities. Examples of organizations that generously offered their time for this research included, but was not li- mited to, housing, education, health and employment services. Housing organizations ranged from shelters, transitional housing, supportive housing, affordable rental housing agencies and safe houses. There were many community and ?grassroots? organizations such as drop-in, crisis, learning, sport, family and support, community centres and schools that offered their knowledge and experiences. Government agencies were also very supportive and included Police Services, Provin- cial Housing Authorities and Municipal Governments. Finally, the non- profit sector such as the Manitoba Assembly of Chiefs, the Indian and M?tis Friendship Centres and Tribal offices in each of the three prairie cities were instrumental in providing a balanced insight into the chal- lenges and opportunities in providing services to those part of the hid- den homelessness population. Not all participating agencies provided programs and services sole- ly to the Aboriginal community; in fact more than half held a mandate that covered their community as a whole. The individuals that gave their time was also diverse in that some held positions of leadership while others held the equally important position of frontline worker. During interviews a portion of the agencies preferred to have more than one staff member participate which provided a rich and thorough insight into how their organization delivers services to the community. While they were busy fulfilling their mandates, they participated wholeheartedly in the interview process. This willingness to participate is thought to be partly a result of extensive relationship-building by the researchers throughout the study and the dedication of these agencies have to assisting the community they serve. For example, one drop-in centre worker in Saskatoon told us, ?In 1995 we had 6,000 visits per year, the past year we exceeded 20,000 visits per year.? Participants told re- searchers that the two most important issues facing Aboriginal persons in finding adequate shelter were the lack of financial resources and availability of shelter in general. Both service providers and persons seeking adequate housing main- tained that by not being able to supply a damage deposit or supply suffi- cient references were a significant barrier in accessing housing. A non- profit housing manager in Winnipeg maintained, ?You need a co-signer for renting a place; it is ridiculous, [it?s] not fair to have guarantee, [it?s] too hard for some.? The lack of finances often pushes people into resid- ing in accommodations located in neighbourhoods that are considered ?unsafe? by some. One service provider in Winnipeg reported, ?The on- ly place a resident can find, on the amount allocated by social assistance, is to reside in an area that is economically disadvantaged. In the long run, this causes problems for some families due to community issues such as gangs, prostitution, bad role models for children, etc.? In Regina a participant stated, ?Adequate shelter isn?t available, that is the issue. Some of the higher ups should be given a welfare cheque and see what type of living situation they can rent with the money.? Lack of employ- ment and education and were other reasons commonly cited. Affordable housing, lack of urban knowledge and accessing support services were the most commonly cited reasons for the experiences for those who are new to the urban setting. A staff member, who works in a transitional housing organization in Saskatoon reported, ?People lack knowledge of who they are, where they are, and what they can do. Wel- fare has programs but do not tell the people, clearly what they are, if you don?t ask, you don?t get it.? Another key issue were language barriers, especially for migrants from northern communities ?Not knowing the system, the language etc. it does not take much for a person to believe that they can?t get help.? Aboriginal Philosophy or Approach to Service Delivery The service providers were asked if their organization has a specific Ab- original philosophy or approach. A majority of these agencies reported that they have developed and delivered culturally appropriate approaches to service delivery over the number of years they have been in operation. For example, one housing organization that serves the Aboriginal community in Winnipeg, deliv- ers their services in a holistic manner by being adaptable to the changing requirements of their clients. This organization works with tenants, some of whom may experience personal challenges that lead to difficulties in paying their monthly rent. Rather than being threatened with eviction, they are ?listened and cared for, understanding that budgets are limited.? Other organizations reported that they offered sensitivity training for their staff or had dedi- cated staff members, such as Elders who were trained to be supportive and sensitive to the healing of Aboriginal peoples. Service providers were asked if they could identify any gaps in the current system or what other programs were needed that would be of benefit to persons accessing their programs. One frontline worker at an emergency shelter in Winnipeg stated, ?I am a gap worker, I fill the gaps.? Agencies share the common concerns in long-term funding ar- rangements. Many stated the procedures for obtaining funding for vari- ous grants and projects was too time consuming and interfered with ful- filling their organization?s mandate. In Winnipeg, a provider of adult education stated, ?We are funded short term for a long term problem.? The following list highlights suggestions that service providers per- ceived as beneficial to the current system in relation to access. ? More funding for social service workers to conduct home visits; ? Education programs related to house maintenance; ? Increased focus on youth programs; ? After care support for people after treatment programs (substance use); ? Quicker response time in arranging appointments with Social Ser- vices; ? More sport programs for youth; ? Increase in shelters; ? Increase in shelters for victims of domestic abuse; ? Increase in drop-in shelters for over-night accommodations; ? Central source of information and referrals; ? Reduction in government bureaucracy; ? Increase in funding for childcare; ? Increase in advocacy in general; ? Increase in communication. In terms of outreach and advertising their services, 90 percent of the participants reported they did not publicize outside the city limits. In particular, non-profit housing organizations expressed that they saw no value in creating awareness of their organization, regardless of geo- graphical location, due to their extremely long and lengthy waiting lists of tenant applicants. Forms of service delivery promotion were primarily through part- nerships with other organizations, attending conferences and serving on committees where a diverse range of agencies participated. Web sites, pamphlets and various forms of literature (posters, reports, newsletters, etc.) were also cited as methods of creating awareness of their services offered. Most importantly, a majority of organizations acknowledged that their clients primarily learn about their services through ?word of mouth.? One agency in Saskatoon reported that their clients became aware of their services through multiple ways, ?The majority of our clients come to us through word of mouth, client to client, agency work- er to agency worker to client. We aim to keep agencies informed.? Or- ganizations continuously adapt to meet the needs of the ever-changing community that they serve. For example, one Winnipeg organization that primarily serves the Aboriginal population has, over the past few years, added parenting, literacy, solvent abuse prevention and nutrition components to their programs due to reports concerning observations the population that they serve. Others have rightfully recognized the impacts of the residential school syndrome and the ?sixties scoop? and have adapted programming sensitive to those specific needs required to contribute to the healing process. When asked if they knew of any trends or best practices relating to the provision of services, surprisingly many did not. This may be reflec- tive of the pressures related to the challenging nature of their work with- in the public realm. One person spoke highly of the recent initiatives and funding from NHI and believes that this program has greatly assisted persons on the verge of homelessness. Another person simply stated that a differing viewpoint is needed regarding the hidden homeless popula- tion, ?Don?t blame the poor, the victims. Deal with solutions versus who is to blame. Poverty is the key, then the rest follows.? What is ?home?? At the broadest level, this research has confirmed that hidden homeless- ness among Aboriginal persons in Prairie cities remains an area of con- cern. This concern stems from the pressing need to address the chronic shortage of housing and related supportive services that has contributed to the high numbers of persons living in temporary accommodations. In the initial forty informal discussions that were held in Winnipeg, participants raised four areas of concern: (1) lack of affordable housing; (2) lack of support networks; (3) lack of information; and (4) institutional discrimination. These four areas were explored in greater detail as the research progressed, and to a large extent, each area was confirmed as being problematic for the wider sample. In particular, it is clear that there is a shortage of quality and affordable housing and mechanisms are needed to better connect persons to the supports they require to bet- ter their present situation. Furthermore, the frequency with which respondents raised con- cerns of discriminatory practices needs to be further addressed to more precisely understand why this was echoed so strongly in each of the three cities. Perhaps to understand the results of this work is best represented in the first survey question that simply asked respondents to define what they meant by ?home.? There is no doubt that defining home was a complex as it resulted in multiple interpretations. In this research home applied to many aspects of life that sometimes included the ?physical house? but for most it was more of an intangible feeling such as ?home where you were born and raised? or ?where my family grew up.? To others home was a place where they sought safety and refuge: ?Some- where you can go anytime. Somewhere you can feel safe and not have to worry about violence. A place where you won?t get kicked out on the street. Home is supposed to be a safe place where you can go. Home is supposed to be yours and it is supposed to be a special place where you can to have some privacy. It?s supposed to be your temple.? Home also evoked an emotional attachment to family and friends: ?a safe place where friends and family can come to see me and be able to enjoy the basic comforts (hot water, food plumbing, heat, security and laundry).? Another person offered home is ?a place that is safe, comfort- able. Somewhere you can raise a child.? Perhaps the words of one per- son, to whom we titled the report after, sums up the meaning and power of home in saying ?home for me is where the heart is at and right now that is nowhere so I am homeless.? These words clearly echoed the chal- lenges that surfaced in the comments. Many contended that home was a place that they could exert con- trol over and independence in thinking and acting, it was about having a space such as one person who observed: ?a home for me would be someplace where I can rest and forget about the world and my problems, it would be a place where I don?t have to listen to other people?s prob- lems for a while.? It was clear that home was a house and that meant having a clean place that was free of maintenance problems and was affordable. How- ever, too often, home appeared to as a distant thought, something that was just out of reach. Participants often spoke with emotion and a desire for a better life: ?a place where you have people around that love you? or ?a house where you live everyday with your family.? It was this type of sentiment that emphasized the fact that so many lacked home, but felt that it was out there even if was momentarily unattainable. The words of respondents also confirmed that for many, finding a place to call home remains a challenge that is hard to overcome. It is hoped that this report was able to articulate the complexity of this issues and that all too often many are left without ?someplace that you can call yours, a place where your stuff is. A place you feel comfortable .? Within this research we also learned a valuable lesson; that was that the ?process of conducting the research? was a vital as the final outcome. This meant that the time spent in each community, collecting information and build- ing the necessary relationships allowed the research team to connect bet- ter with those who participated in the study. This thought is important in understanding the value of including the Talking Circle, which was more of an expression of the research process rather than a means from which to collect more data. The researchers wanted to embrace indigen- ous approaches and perhaps in a few small ways, we took some steps in moving in this direction. Conclusions This research was guided by three research questions that were posed in the introduction of this report. The following discussion provides in- sights from the study that address these research questions: 1. What are the general characteristics of the hidden homeless popula- tion among Aboriginal persons, and has the pattern of migration in- to large urban centres played a contributing factor in exacerbating the extent of hidden homelessness? 2. To what extent does the condition and availability of the housing stock, and housing services, exacerbate the hidden homelessness situation in prairie cities? 3. How are governments, community -based organizations and sup- port agencies addressing the needs of Aboriginal people who expe- rience hidden homelessness in Winnipeg Saskatoon and Regina? The focus of the first research question relates to the characteristics of Aboriginal persons experiencing hidden homelessness, and the effect of migration to urban centres on this phenomenon. Overall, the results of the study suggest that: 1. Aboriginal persons experiencing hidden homelessness a diverse group represented by males and females, youth, single parent fami- lies, elders, and, increasingly, families. 2. The reasons for housing distress amongst this group are wide- ranging, however, all suffer from overwhelming poverty and the lack of adequate shelter. 3. The primary concern for the majority of respondents in the sample was the inability to access a permanent residence. 4. For many Aboriginal peoples, migration from rural communities to ur- ban areas creates a complex dynamic between their inability to find ap- propriate accommodation in the city and their connections to home. The second research question refers to the condition and availability of housing stock and the provision of housing services in Prairie cities. Specifically, the question poses whether the inadequacies of housing provision exacerbate hidden homelessness amongst Aboriginal peoples. Two main responses were generated from this question: 1. There is a significant shortage of affordable shelter accommodations for the urban Aboriginal population in Canadian Prairie cities to address both short- and long-term needs. 2. Despite the lack of housing provision, most respondents indicated that they had social supports that assisted them in maintaining a roof over their heads. This social support distinguishes absolute homelessness from hidden homelessness. Moreover, this social support network ?hides? the problem of Aboriginal hidden home- lessness from mainstream society. The third question asks how government, organizations and agen- cies are attempting to address the needs of Aboriginal people who expe- rience hidden homelessness in Prairie cities. The study found: 1. Most program responses to hidden homelessness are reactive rather than proactive. In order to eliminate hidden homelessness, pro- gramming must establish long-term goals that will lead to perma- nent housing. 2. In each Prairie city, supportive networks do exist for the hidden homeless Aboriginal population. These supports span a continuum that ranges from formal to informal supports. In addition, individu- als staying temporarily also contribute support to the household through contributions to the rent as well as in-kind support such as childcare. 3. A paradigm shift is occurring with the downloading of services to the community level. With increasing demand on community agen- cies, their resources are being strained. Nonetheless, the grassroots foundation of these agencies has allowed the development of sup- portive networks that would not be possible in government pro- gramming. Recommendations The following recommendations are outlined in relation to the core themes of the study (mobility, shelter and services). Mobility Further investigation is required to gain a better understanding of the complex dynamic between home communities and urban centres for Aboriginal peoples. In particular, focus should be on the hidden home- less experiences of those in the sample who indicated a connection with their reserve resulting in movement between reserve and urban centres. Programming must be established to address the hyper-mobility of Aboriginal peoples in urban areas. It is only with substantive increases in housing provision, both on and off reserve, that the ?churn? of Abori- ginal peoples will be recognized. Shelter Increased funding is required for the construction of transitional and permanent housing units to accommodate both short-term needs, such as migrating to the city, as well long-term needs, such as those wishing to reside in urban centres. Choices in housing design must be expanded and diversified to in- corporate Aboriginal culturally appropriate housing for the Aboriginal population. For example, such housing could accommodate the tradition of maintaining three- and four-generation households through multi- generational housing units and guest accommodation. Recognition must be accorded to emerging literature that promotes a holistic approach to the provision of housing. Based on this approach, housing represents far more than shelter and incorporates a range of ser- vices that enables Aboriginal peoples to sustain an independent lifestyle in a metropolitan centre. The overwhelming message of participants in the study was that access to shelter is significantly hindered by systemic barriers that in- clude perceived discrimination by landlords, as well as requirements for references and damage deposits, especially for those new arrivals that lack local connections and financial means. These barriers must be addressed to facilitate access to housing for Aboriginal peoples experiencing housing distress. Services and Support The significance of informal support networks (such as family members providing shelter or assistance) for the hidden homeless in the Aborigin- al population must be acknowledged. Moreover, the critical nature of this support must be formally solidified so that financial resources will be available to those households that are providing shelter to the hidden homeless. This might take the form of an innovative program that recog- nizes the unique circumstances of those in need of shelter and the role of friends and family in providing care . Increases in shelter assistance programs are required to allow great- er access to housing through increases in shelter dollars. For example, in Manitoba, the shelter assistance rates, which have not increased substan- tially since the early 1990s, must be addressed to match the current mar- ket conditions, which have increased dramatically. In addition, access to shelter assistance programs needs to be better communicated to those in housing distress to ensure that they are all aware of all of their options to address their situation. For example, in Winnipeg, there are numerous organizations and agencies that provide programs ranging from temporary or emergency accommodation to long-term, affordable options. In addition, govern- ment subsidized housing programs also exist that provide shelter on a rent to income ratio or provide shelter assistance payments to those in need. While many of these programs have extensive wait-lists, it is sug- gested by the outcome of this research that many in critical need of shel- ter are sometimes unaware of the programs and options available to ad- dress their shelter needs. Therefore, continuing to disseminate information about existing programs and supports is one small piece of the solution. Jino Distasio is Associate Professor of Geography and Director of the Institute of Urban Studies at the University of Winnipeg. Gina Sylvestre and Susan Mulligan are research associates at the Institute of Urban Studies. Chapter 7 .1 A Revolving Door? Homeless People and the Justice System in Toronto SYLVIA N OVAC , J OE HERMER, EMILY PARADIS, AND AMBER KELLEN The homelessness capital of Canada Toronto has been named the homelessness capital of Canada; more than 30,000 people a year are admitted to a shelter. Many shelters provide only a bed for the night, leaving homeless individuals little choice but to inhabit public spaces. As in other cities across the country, politicians have reacted to visible homelessness by considering or enacting legal responses to curtail the behaviour and very presence of homeless people in their midst. Which leads researchers to question how many homeless people end up in correctional facilities for reasons relating to their lack of housing. And how many of the approximately 50,000 ex-prisoners re- leased each year from provincial correctional facilities in the Toronto area end up on the streets? This chapter draws on research prepared for the Housing and Homelessness Branch that received funding from the National Research Program of the National Homelessness Initiative. Researchers from the Centre for Urban and Community Studies and the John Howard Society of Toronto conducted a literature review, an analysis of administrative data; a review of client files; a survey of 57 homeless individuals; in- depth interviews with 22 homeless individuals; focus groups with home- less individuals and service providers; and interviews with 23 key in- formants. Incarceration among homeless populations In 1998, the Mental Health Policy Research Group conducted a survey of unaccompanied homeless adults in Toronto. Among a sample of 300 adult shelter users, 73 percent of the men and 27 percent of the women had been arrested since age 18; 49 percent of the men and 12 percent of the women had served jail time at least once. Within the previous year, 30 percent had spent some time in jail; this was 6 percent more people than had spent nights in psychiatric facilities. Most jail stays were short ? only 5 percent of the homeless were incarcerated for more than six months. This suggests that the charges were for relatively minor of- fences. Of the 5,052 people counted in the City Toronto?s 2006 Street Needs Assessment survey, 18 percent had had ?an interaction with corrections? and 17 percent had ?had an interaction with probation or parole? in the previous six months. The likelihood of having been incarcerated is even higher among those who sleep in places considered unfit for habitation. In its 1998 study, the Mental Health Policy Research Group found that so-called ?rough sleepers? in Toronto were more likely than shelter users to have been arrested, held overnight, convicted, to have served a sentence, and done so more than once. Studies in Toronto, Vancouver and Edmonton have also found that a high proportion of homeless youth have been involved with the crimi- nal justice system. Three-quarters of the homeless youth interviewed in Toronto in 1995 were involved in delinquent activities such as stealing and burglary, and had been incarcerated. Youth whose peers engaged in criminal activity, such as drug selling and theft, were more likely to commit such acts themselves, regardless of their home and school ex- periences and previous criminal experience. Finally, several studies have shown that individuals of Aboriginal descent are overrepresented among both those who are homeless and those who are incarcerated. The City of Toronto?s Street Needs Assess- ment showed that homeless individuals of Aboriginal descent were overrepresented in jails, shelters, and public spaces. In total, 16 percent of those surveyed identified themselves as Aboriginal?7 percent among those incarcerated, and 26 percent among rough sleepers (compared to 2 percent of the Toronto population). These studies can be contrasted with those that indicate that people living in poverty do not commit more crimes than those with higher in- comes, but are more often arrested and held on remand because they cannot obtain bail funding. At least two studies have substantiated the claim made by advocates that being held on remand is an almost guaran- teed outcome for those without a fixed address. Being homeless in- creases the odds of being detained or remanded into custody and the No Fixed Address Admissions to Adult Correctional Facilities in Toronto Area, 200 1 to 2005 decision to plead guilty. Moreover, persons without a permanent ad- dress are more likely to be denied bail than those with a permanent home, indicating that homeless individuals are more likely to be re- manded to custody. Homeless people entering the justice system The number of people who are homeless when they are arrested, jailed, or released from jail in Canada is unknown. To begin to fill in this blank, the research team analyzed data for the Toronto area on admissions and releases from the Ontario Ministry of Community Safety and Correc- tional Services (OMCSC), and on shelter use from the City of Toronto. Between 2001 and 2004, the number of adults with no fixed address (NFA) admitted to five correctional facilities in the Toronto area per year increased steadily, along with the number of admissions (which includes multiple admissions by the same persons within the specified 12-month period). The number of individuals admitted with no fixed address in- creased by 64 percent from 2001 to 2004 (from 174 to 286). There was a slightly higher increase of 68 percent (from 296 to 496) in the number of admissions with no fixed address. During 2004-2005, a total of 286 people were admitted 496 times (i.e., 42 percent of the admissions were individuals returning to jail within a 12-month period). This indicates a great deal of cycling of homeless people in and out of jails. These are substantial annual increases in the number of homeless individuals jailed and in the number of admissions, which includes the repeat incarcerations of some people. In other words, more homeless people were jailed each year. And four of ten admissions of a homeless person between April 2004 and March 2005 were returnees. This is a marked pattern of recidivism. According to the OMCSC data, the annual proportion of NFA to all admissions is quite small ? about 1 percent. However, there is little rea- son for individuals to disclose their homeless status, and we believe this to be a severe undercount. Reportedly, a snapshot count conducted on April 9, 2005, determined that 18 percent of the prisoners (79 out of 441) admitted to the Toronto (Don) Jail had no fixed address. Just as males constitute the vast majority of those incarcerated, al- most all of the NFA individuals admitted to a correctional facility were male. The proportion of all NFA admissions that were women ranged from about 1 percent to 3 percent. OMCSC records also show that the proportion of individuals admit- ted to youth correctional facilities in the Toronto area who reported hav- ing no fixed address is small ? almost 1 percent of all youth admissions per year, with negligible variance from year to year. The number of ad- missions slightly exceeded the number of individuals in each year, indi- cating repeat admissions of some homeless youth within a 12-month pe- riod. Homeless people leaving the justice system According to OMCSC figures, a total of 414 adults with no fixed address were released from Ontario provincial jails (excluding those released from court) between April 1, 2004, and March 31, 2005. The level of miss- ing data is high ? almost one out of eight adults released from jail did not provide information on his or her address. Imminent homelessness can- not be assumed in all cases in which an ex-offender was uncertain (for example, the ex-offender may not yet know whether he or she can stay with a family member). Some of the non-respondents were probably homeless. And, of course, those who did report an address may not have stable or adequate housing. The Re-integration Program managed by the Ontario Multi-faith Council on Spiritual and Religious Care is the only province-wide pro- gram that offers help to adult prisoners who have no home on release and collects data on them. The Program?s needs assessment kit includes the question: ?Do you have a place to stay when you get out?? This is a more pertinent question than asking for one?s address on release. From April 2004 to March 2005, the Reintegration Program documented 656 individuals in Ontario who had no place to live on release. The program statistics distinguish persons with no plans for a place to live from those who anticipate staying in a shelter and those who plan to stay temporar- ily with family or friends. If the 30 percent of respondents who plan to be temporarily housed by family and friends are excluded, 460 individuals remain ? still more than the 414 identified by the Ministry. A possible explanation for the discrepancy is that prisoners are more comfortable disclosing their homeless situation to a non-profit agency that offers assistance than to corrections system authorities. Nevertheless, data from both sources con- firm that, at a minimum, well over 400 persons left a provincial jail with- out a place to live in 2004-2005. Moving from jail to a shelter Some men enter a shelter in Toronto dressed in standard-issue prison orange jumpsuits and blue shoes. In some cases prisoners are released from court without the opportunity to retrieve their clothes from jail. People admitted to a shelter on discharge from a correctional facility are not always so easily identified, but information from the shelter system indicates the number of people in this position. Since 1988, the City of Toronto has maintained a database on shelter users. All those admitted are asked the main reason for their use of the shelter, and this information is recorded according to a set of categories that include release from a correctional facility. According to the City data, an average of more than 800 shelter admissions a year in Toronto are individuals discharged from a correctional facility who have no place to live. Based on evidence of widespread under-reporting by shelter staff and limited disclosure by shelter users, this number is an under-count. For this reason, the characteristics of individuals for whom data was available, as outlined below, may be partial. Who are the individuals entering shelters from corrections? The following profile of individuals from corrections admitted to a shel- ter between 1988 and 2003 describes a steady, if not worsening, problem, predominantly affecting single men. The number of individuals varied little over the selected years?about 630 persons per year. Virtually all were single. The total number of admissions per year?825?is higher, due to multiple admissions of the same person within a year. This shows that some individuals are repeatedly and quickly being shuttled from jail to shelter and back. Most were male, although the proportion of adult females increased three-fold, from 7 percent to 21 percent; the proportion of middle-aged women more than doubled, from 9 percent to 23 percent. The reason for this change is unclear, but it may be associated with the exclusion from the mid-1990s onward of data from abused women?s shelters (whose residents are unlikely to have come from corrections). Most were adults. The average age increased slightly, from 29 to 33 years, between 1988 and 2003, possibly reflecting little more than a grad- ual aging of the population. However, youth were overrepresented. In 2003, 33 percent of them were aged 16 to 24. This is higher than the pro- portion all shelter users who are youth ? 21 percent in 1999; and higher again than the general population of youth in Toronto ? 12 percent. The overrepresentation of youth in this profile may be due in part to closer questioning and more diligent recording in youth shelters. Shel- ters that provide more services, such as case management and counsel- ling, may be more likely to probe for this type of information; this ap- plies primarily to shelters for youth and abused women. Moreover, youth and those with a severe mental illness are more likely to receive assistance (from social workers, lawyers, etc.) when released from court or corrections. They are more likely to be directed to shelters rather than end up on the street. Where do people go when they leave a shelter? During 1988 and 1993, the vast majority of people left one shelter only to enter another. But during 1998 and 2003, very few people moved from one shelter to another; instead, the whereabouts of most individuals was unknown. Also, the number of individuals who went to a hospital or treatment program decreased. How long and how often do people stay in shelters? Most individuals from corrections stayed in a shelter less than one month. A small proportion stayed longer, but the average length of stay was less than three months (within a particular year). Compared to ear- lier years, 2003 showed a slight increase in the number of stays of more than one month, a pattern consistent across all age groups. Those who came from corrections tended to spend longer in the shelter system than other shelter users. Our data covers each 12-month period independently and does not track individuals. For these reasons, it may not be a wholly accurate reflection of individuals? patterns of re- admission. The proportion of repeat shelter admissions (multiple admissions by the same person) within a year increased from 22 percent in 1988 to 28 percent in 2003. Among youth, the increase was from 29 percent in 1988 to 50 percent in 2003. This mounting frequency of cycling between jails and shelters within 12-month time periods suggests that it has become harder for individuals to break the trans-institutional cycle. Discharge experiences Prior to going to jail, I lived on the street for most of my life. I have been in jail on several occasions. Nothing was ever put in place for housing upon my release.? This comment was made by one of the 22 homeless in- dividuals we interviewed. Seventeen of the 22 respondents were home- less when they were last jailed, and five had lost their housing while in jail, in a few cases because they were not allowed to return to live with family members. More than half of the respondents (13) were unaware that they could have requested assistance with discharge planning. A few requested help, but did not receive it. Two people released from court received no help, one because his worker was not in touch at that point. Only six re- spondents were assisted by jail or agency staff to find a place to live when they were released. Two refused the assistance offered, due to mis- trust of the worker or a mistaken belief that they had already secured a place to stay. Four received limited assistance from jail staff or another agency. Jail staff assistance consisted of being given a list of shelters and transportation costs. Some respondents expected little help from jail staff. Several respondents interpreted ?being housed? as simply getting a shelter bed. As one commented, ?I was given a list of phone numbers for shelters. The jail staff didn?t make any phone calls to make referrals. Shelters will not hold a bed in advance. I have to call the shelter once I am released to see if there is a space.? Ten respondents said they had a place to live on release, but this usually meant staying with family or friends, at least temporarily, or go- ing to a shelter. These were usually unstable arrangements of a few days, weeks, or months. Only two respondents maintained their post- discharge residency for more than a year. Seven respondents received some assistance after their release, such as help with applications for welfare and subsidized housing; having shelter beds held until their arrival; or replacing lost documents. One respondent outlined the difficulties of recovering possessions, including identification documents, and the effects of discharge with no assistance. ?In order to retrieve belongings taken at the point of arrest, we have to go through a lengthy process to get them back. The property room at the cop shop is not helpful. Usually it takes a lawyer to get be- longings back. When a person gets out, you find yourself in a worse po- sition then before: no housing, no money, and being forced onto welfare and into shelters. Every time I have been in jail, I lose my I.D.? These comments suggest that discharge planning in remand and short-term facilities within the provincial jail system is limited. Respondents mentioned some basic ways to improve discharge practices and reduce recidivism. ?Considering that people come out of prison with no money, no clean clothes, and no housing, it?s not hard to understand why some people quickly re-offend. Immediately on my re- lease, in order to survive, I starting thinking in a criminal way and soon began to re-offend. I think that Corrections ought to be able to get some support in place, like welfare and subsidized housing, before a prisoner is released. For example, a bus pass and first and last month?s rent in the amount of $1,500, pro perly set up through Ontario Works, could help to stop an ex-offender?s return to crime.? The key informants we interviewed had mixed opinions about how homeless youth and adults were treated within the criminal justice sys- tem. One respondent commented on the way in which some people cy- cled in and out of shelters and jails: ?For some homeless youth, jail is a respite for 30 days ? meals, a roof over their head, everything becomes predictable for 30 days. After a while, this conditioning leads to institu- tionalization, and another set of problems develop.? Some stated that homeless people, particularly youth, were treated with disdain. Others referred to positive outcomes from jail terms: ?Sometimes when someone has committed a crime that results in incar- ceration, the time spent in jail can help to stabilize an individual. They get access to dental work, nursing staff, a staff psychiatrist. The individ- ual may not have access to these services on the streets. The individual has the opportunity to withdraw [from toxic substances] in a safe mon i- tored environment. Access to other supports, such as a chaplain, health services, etc. may provide a window to make some changes. That?s not what the criminal justice system is for, but it?s relevant to the homeless population.? Proportion of Repeat Admissions to Shelters from Corrections, Selected Years (bas ed on admissions) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1988 1993 1998 2003 Year P ro po rt io n of A dm is si on s th at w er e P re vi ou sl y A dm itt ed In di vi du al s Proportion of admissions that were repeated individuals Proportion of admissions that were new individuals One informant within the criminal justice system outlined various ways in which being homeless may affect treatment. ?Biases toward homeless people are built into the system. If an individual is well known as being homeless, with a history of failing to appear before the court or skipping bail, then they are more likely to be remanded. Show cause [a hearing to determine whether bail is an option] is more likely. In some cases, an individual may feel compelled to plead guilty, between having no address or home, together with the difficulties of accessing a Legal Aid lawyer. Homeless offenders are more likely to be held in pre-trial custody, plead guilty more, and have higher conviction rates.? Another informant noted that Aboriginal and racial minority youth are overrepresented within the criminal justice system. Youth of Abo- riginal descent have the option to go to an Aboriginal court and can benefit from this option. But youth access to mental health services, in- cluding the mental health court diversion program, is limited. ?Young people are not always clearly identified or diagnosed [with mental health issues] at that stage in life, so they do not get that option as an alternative to incarceration.? The costs of incarceration vs. housing For the destitute, the corrections system has become a provider of basic services and more of a relief from a precarious existence than a deterrent to crime. It is, however, a very costly option. Consideration of the costs of homelessness alone should prompt a re-assessment of allowing people to cycle through prisons, hospital, and shelters rather than providing them with affordable and supportive housing. The cost of a new unit of non-profit housing per person is lower than various institutional alternatives or the provision of support ser- vices to the homeless. The figures in Table 2 are from 1999. Supportive and transitional housing such as that provided by the John Howard Society and Elizabeth Fry, as well as group homes for in- dividuals at risk of homelessness, are less than one-tenth (6%) the cost of incarceration in provincial corrections facilities. Table 2. Cost by type of housing per month, 1999 Correctional facility: adult $3,720 Correctional facility: young offender $7,917 General hospitalization $4,500 Psychiatric facility $10,800 Shelter/hostel $900?$2,100 General homeless support (police, health, and other support services) $4,583 New non-profit unit $1,080 Conclusions and recommendations Although the statistics we compiled under-estimate the number of peo- ple who are homeless before and after their involvement with the crimi- nal justice system, their pattern reveals a worsening problem, with a sizeable sub-group of homeless people who are stuck in a cycle of stay- ing in shelters, jails, and hospitals, and becoming increasingly alienated from community life. The long-term homeless adults and youth we interviewed had had extensive involvement with the criminal justice system. Most who are incarcerated for short periods of time or held on remand in provincial correctional facilities received little or no assistance to prepare them for community re-entry. The results of this study support a recommendation that the Gov- ernment of Ontario adopt a policy that no person being released from jails or courts in Ontario be released to the streets without access to ade- quate housing. Also, the provincial government should provide dis- charge planning services to all those who are serving sentences and those who are held on remand. Such planning should be available upon ad- mission to correctional institutions and at the time of release from these institutions, and include appropriate action to locate and provide access to adequate housing. Finally, at the time of release from court or jail, all persons who are identified through the discharge planning process as homeless, or likely to be homeless, should be offered and provided with appropriate civilian clothing, a current list of community resources, and public transit fares, at no personal cost. These, and several other recom- mendations, are described in more detail in the full report. Sylvia Novac is a research associate at the Centre for Urban and Community Studies, University of Toronto. Joe Hermer is a Professor in Criminology at the University of Toronto. E mily P aradis is a doctoral candidate at the Ontario Institute for Studies in Education, University of Toronto. A mber Kellen is Su- pervisor of Advocacy/Community Programs at the John Howard Society of To- ronto. Chapter 7.2 More Sinned Against than Sinning? Homeless People as Victims of Crime and Harassment SYLVIA N OVAC , J OE HERMER, EMILY PARADIS, AND AMBER KELLEN A dangerous world Media reports and academic research tend to focus on the real or per- ceived criminal involvement of the homeless. What is less well known is that homeless people are more often victims of crime than housed peo- ple. Numerous studies have established that homeless individuals have experienced high levels of violence and victimization both before and after becoming homeless. In 1992, the Toronto Street Health Report included the results of a survey of homeless single adults that asked about their experiences of victimization during the previous year. Among the sample of 106 women and 352 men, 46% of the women and 39% of the men said they had been physically assaulted. One-tenth of the respondents reported that they had been assaulted by police, some more than once. Sexual assault and violence were common experiences for women ? 43% of women and 14% of men said they had been sexually harassed, usually multiple times. Even more disturbing, 21% of the women said they had been raped. The survey respondents described instances of be- ing assaulted by security guards in shopping centres, beaten with night- sticks by police officers, and sexually harassed on public sidewalks. It is unclear how many of these assaults caused injury, but one-tenth of the respondents said they had gone to a hospital emergency room for as- sault-related injuries, and about half of them were admitted to hospital for treatment. Another study in 2001 that drew on 316 clinical records from a To- ronto sexual assault care centre found that sexual assaults against home- less women differed from those against housed women?they were more violent and more often perpetrated by strangers in public places. Homeless youth have reported rates of victimization higher than those of homeless adults or housed youth. A study that compared the experiences of homeless youth and high school students in Toronto found that 69% of homeless youth said they had been physically a s- saulted in the previous year, compared to 39% of high school students. Similarly, 29% of homeless youth said they had been sexually assaul ted in the previous year, compared to 6% of high school students. Studies on relations between homeless people and the police sug- gest that the homeless see police and the criminal justice system as agents of control, not protection. They have accused police officers of attempting to control their behaviour by the overuse of tickets for of- fences related to their lifestyle: consuming alcohol in a public place, uri- nating in a public place, or falling asleep on a public bench, and frequent arrests for minor offences related to survival such as shoplifting food. Street youth complain that when they are standing or sitting on the sidewalk, or sitting on a park bench, the police will approach them, ask for identification and attempt to move them on. Often this involves searches, verbal harassment, confiscation of goods and in some cases the use of force. The quality of treatment people receive from police and courts helps determine the public?s views of criminal justice institutions. Perceived fairness influences one?s sense of obligation to obey the law and reduces re-offending rates. Even police use of discourteous language can deter- mine the way they are seen by ordinary citizens. Perceptions of policing vary by socio-demographic and ethno-racial group, related to how police treat group members. Although few studies have explicitly examined the state of relations between police and home- less people, there is evidence that it is strained and fraught with suspi- cion if not hostility. A survey of perceptions and experiences To probe these issues, we conducted a survey to learn about the experi- ences and views of homeless individuals who have been involved with the criminal justice system or been victimized. The sample was not ran- dom, so the results cannot be generalized; nevertheless they reveal the perceptions and experiences of some long-term homeless youth and adults in Toronto. A sample of 57 homeless youth and adults was recruited from sev- eral drop-in centres and other services for homeless people to participate in a face-to-face survey. Of this group, 22 respondents who had been ar- rested or incarcerated or multiply victimized within the previous year were interviewed more intensively to explore their experiences. The sample included 17 youth; 18 women and one male-to-female transgender person; 17 Aboriginal people; and 12 racial minority indi- viduals. The respondents ranged in age from 16 to 59 years. Most (88 percent) were born in Canada; 31 percent in Toronto. Of those born in another country, most were not recent immigrants, but had lived in To- ronto for many years. Most (63 percent) of the respondents had not com- pleted high school. Only four had completed a postsecondary school program. Each of the respondents had experienced lengthy periods of abso- lute and relative homelessness. Almost three-quarters had slept rough for some period of time during the year before the survey. At the time they were interviewed: ? 28 (49 percent) were staying in a shelter; ? 11 (19 percent) were sleeping rough; ? 8 (14 percent) were staying with friends or family; ? 4 (7 percent) had their own apartment; ? 6 (11 percent) were staying at a halfway house, had just been re- leased from jail, or did not answer the question. Two-thirds of the respondents had been incarcerated in the previ- ous year. Three-quarters had spent at least one month in jail or prison at some time. Policing and fairness Respondents shared with most Canadians an appreciation of the neces- sity of policing to maintain law and order in society. However, they also believed that the police act unfairly in their treatment of racialized groups, the young, and the poor. The respondents, most of whom were white and male, generally agreed with the statements that the police unfairly stop and question racial minorities or treat poor people worse than wealthy people. And they tended to disagree that the police were criticized too much or that it was rare for an innocent person to be jailed. Respondents? gender or age did not significantly affect their re- sponses. There were some differences by racial group: more Aboriginal people felt that innocent people are sometimes wrongly jailed; that the police treat males worse than females; and the poor worse than the rich. And fewer whites agreed that police treated people from some racial groups worse than others. Table 1. Statements on policing and fairness Agree / strongly agree (%) The police treat wealthy people better than poor people 90 The police treat people from some racial groups worse than people from other racial groups 90 We need police in this country to keep law and order 90 People from my racial group are more likely to be unfairly stopped and questioned by the police than people from other racial groups [minority] 92 [whites] 39 [total] 61 The police treat young people worse than older people 60 The police treat males worse than females 53 Everyone has an equal chance of getting ahead in Canada 46 People criticize the police too much 35 It is rare for an innocent person to be wrongly sent to jail 25 Contact with police We asked respondents about their interactions with the police. Most had been stopped at least once by the police during the previous year, 24 had been stopped more than five times. During the previous month alone, half of the respondents had been stopped by police, and five had been stopped more than five times. Table 2: Police Action Experience by Respondents During previous year (%) During previous month (%) Stopped at least once 86 49 Stopped more than 5 times 42 9 Warned 81 39 Photographed 56 11 Ticketed 58 18 Searched 79 32 Detained 68 20 Arrested 67 14 More than half of the respondents had not only been stopped by the police during the previous year, but also warned, photographed, tick- eted, searched, detained, and arrested. These are not mutually exclusive events ? it is common to be stopped, detained, and warned in a single incident. There were no statistically significant differences in terms of sex, age, or racial status. When asked to describe their personal contact with the police, 33 of the 57 respondents (58 percent) characterized it in predominantly nega- tive terms, frequently suggesting that the police abused their power. A minority of respondents (11 out of 57), including some who had been arrested, said they were treated fairly by police (now, if not in the past). Nearly half of respondents (24) said they had been assaulted by po- lice officers. The reported assaults ranged from being harshly pushed to violence causing serious physical harm. A few respondents appeared equivocal in their characterizations of treatment by the police, referring to mostly positive or neutral contacts with police, yet also recounting abusive treatment. Several respondents said their own demeanour affected how the po- lice behaved, usually stating that police officers became (more) confron- tational or aggressive when a homeless person asserted his or her rights. One respondent said this was especially the case with people of colour. A few respondents referred to Aboriginal or racial status as a factor in the likelihood of mistreatment by police. Although the respondents were not asked to describe in detail any abuse by the police, a few Abo- riginal respondents disclosed severe physical assaults. Of those who re- ported severe physical abuse causing significant injury, all five were Aboriginal adult males. Some respondents qualified their assessments by distinguishing ?good? and ?bad? officers. They usually credited seasoned, older, and female police officers with more professional behaviour than young male ?rookie? officers who needed more training and life experience and were less likely to explain procedures regarding bail, court, and night security in police stations. Not surprisingly, given their predominantly negative experiences with the police, some respondents said they could not rely on the police for protection, because they were known to be homeless, had a record of offences, or anticipated being treated badly. Informants? views Those who work with homeless people generally shared the view that police sometimes treat homeless individuals unfairly, to the point of ?hassling? or abusing them. They have heard many such stories and complaints from their clients and sometimes have witnessed what they perceive as unnecessarily rough handling of homeless individuals by the police. On occasion, service providers themselves have been treated harshly by the police when they intervened on behalf of their clients. Service providers confirmed that some homeless youth and adults do not trust the police, in part because of adherence to street culture norms, and in part because they fear abuse and reprisal from the police. The marginalization of homeless people leads them to believe they have no protection under the law, whether or not that perception is accurate, and to act in accordance with that belief. This behaviour is a factor in their relations with the police, the community, and each other. It con- tributes to their victimization by placing them in a lawless context in which raw power holds sway. Contact with security guards To many people, the distinction between a police officer and security guard can be confusing, since both wear uniforms and engage in policing activities with some degree of institutional authority. For visibly home- less people who spend a lot of time in semi-public spaces, however, deal- ing with security guards is commonplace. Virtually all respondents were quite knowledgeable about the main distinguishing markers ? type and colour of uniform and presence or absence of badge numbers and guns ? and the territorial and legal limits of guards? authority (for example, they would say, ?Security only have legal rights to warn and only on the property they are guarding?). Of those who commented on guards? behaviour and demeanour, most considered them to be more aggressive than police officers, several referring to them as ?wannabe cops? and ?thugs? who overstep their authority, knowingly or not. Police officers were credited with having more education and knowledge of the law and being more likely to act professionally, that is, following a protocol of informing a person of their rights and using legal terminology. Respondents were asked about their experience with security guards. Overall, they had less direct contact with guards than with the police, but more than half of them had been intercepted by a security guard at least once during the previous year. Eleven respondents had been detained by a guard and released to be arrested by the police in the previous year; in this way, security guards can occasion an entry into the criminal justice system. Being stopped and warned by a security guard was the respon- dents? most frequent experience. More than half of the respondents had been stopped and warned by a guard at least once in the previous year. During the previous year, three out of ten had been detained by a guard, and two out of ten had been arrested and released to the police. There were no statistically significant differences between sub-groups in terms of sex, age, or racial status. Treatment in the criminal justice system We interviewed 22 of the survey respondents more intensively about their experience within the criminal justice system. Seventeen of the 22 respondents had been detained or arrested by the police within the pre- vious year. A few were detained and released on the spot or after being taken to a police station. Their charges included: failure to appear in court, failure to comply with a court order, theft, communicating for the purpose of prostitution, possession of drugs or alcohol, possession of a pocket knife, soliciting (i.e., begging, squeegeeing), indecent exposure, prowling at night, and drug trafficking. They were detained or jailed for periods of time that varied from one hour to several months. Thirteen respondents had appeared before a court, sometimes more than once. Six said they understood what was happening when they were brought to court and why they were being treated as they were, and six said they did not (one person did not respond). Some respondents felt they had not been treated fairly. For example, one woman said she?d been abused and degraded by a female officer during a strip search. Others said the court had not paid attention to their statements or failed to consider important evidence. One person said his sentence was unfair. One man was released from court in an orange jumpsuit with no money, no identification, or change of clothes. Unlike prison or jail dis- charges, court releases are unplanned ? a conditional outcome ? and an inmate?s clothes and belongings are left behind, sometimes in another city. Retrieving clothes and other possessions can be difficult and time- consuming. Such releases are frequently a precursor to homelessness. More than half of the respondents (13) were unaware that they could have requested assistance with discharge planning. A few re- quested help, but did not receive it. Two people released from court re- ceived no help at all. Only six respondents were assisted by jail or agency staff to find a place to live when they were released. Two refused the assistance offered, due to mistrust of the worker or a mistaken belief that they had already secured a place to stay. Four received limited assis- tance from jail staff or another agency. Jail staff assistance consisted of being given a list of shelters and transportation costs. Some respondents expected little help from jail staff. Several respondents interpreted ?being housed? as simply getting a shelter bed. These responses suggest that homeless people are frequently con- fused about their court proceedings; and discharge planning in remand and short-term facilities within the provincial jail system is minimal. Victimization Most of the homeless youth and adults we surveyed (72 percent) had been victims of crime within the previous year (N = 52). Only eight of the 41 people victimized had reported the crime to police; and five of the eight were very dissatisfied with the police response. There were no sta- tistically significant differences in responses by gender, age, or racial status. Slightly more women than men said they had been a victim of crime, but they were no more likely to have reported it to police. A simi- lar proportion of youth and adults said they were victims of a crime, but fewer youth reported it to the police (only one). All the women and ra- cial minority respondents who reported a crime to the police said they were very dissatisfied with the police response. Why were so few crimes reported to the police? The respondents said they did not trust that the police would be fair or would protect them. Two-thirds believed the police would be biased against them (would insult them or cause more trouble for them). More than half (53 percent) believed that the police would be uninterested in their report. About one-third (32 percent) believed the police would be ineffective (would arrive late or not show up at all; would not do a good job of deal- ing with their report). Most disturbingly, about one-quarter of respon- dents (26 percent) said the offender was a police officer. Seven respondents referred to the ?code of the street? as a reason for not contacting police. That is, street people may protect other street people (for example, by not ?squealing? on a homeless person). If loyalty is not a sufficient reason to keep quiet, fear of retaliation is generally ef- fective. Violations of street culture can result in reprisal from other homeless people, drug dealers, or pimps. Some respondents said they would take care of any problems in their own way. Others just kept problems to themselves. Those with criminal records or a current war- rant did not want to attract police attention. And previous unpleasant or degrading experiences with the police convinced several respondents that they could anticipate more of the same kind of mistreatment. Respondents were asked their experiences of different forms of crime. Most respondents (85 percent) had had some belongings stolen during the previous year, usually at a shelter or other location, rather than directly from their person. Money or identification documents were most often stolen (60 percent and 56 percent, respectively). Other items included clothing, shoes, blankets, toiletries, drugs, medication, food, and electronic devices, such as a cell phone or CD player. If identification was stolen, respondents were usually unaware of who had done it, but three people claimed it had been the police. Among the respondents, 28 (17 men and 11 women, including a male-to-female transgender person), had been physically assaulted dur- ing the previous year. The assaults usually occurred on the street. Sixteen of them had been assaulted more than once during that time period. De- scriptions of the most recent assault revealed that, overall, the perpetra- tor was equally likely to be a stranger, acquaintance, or police officer; more men had been assaulted by the police than women. Seven men and one woman said they had been assaulted by the police (in once case at a police station). Comments indicated that members of the general public are included among those who have targeted homeless individuals. A weapon was used in nine cases (including a baseball bat, gun, mace, pepper spray, police baton, and knife). Men were more often as- saulted with a weapon than women. Fourteen respondents said they should have seen a doctor as a result of the assault, but only eight of them had actually done so. Twelve respondents (23 percent) reported being sexually assaulted during the previous year ? nine women, one (male-to-female) trans- gender individual, and two men. Five adult women, two of them Abo- riginal, and one white adult male reported multiple sexual assaults. De- scriptions of the most recent (if more than one) sexual assault revealed that the perpetrator was equally likely to be a stranger or known person (boyfriend, spouse, acquaintance, or john). Respondents were asked about verbal abuse, threats, or attacks dur- ing the previous year, that they attributed to their homeless state, their ethnicity or race, or a perception that they were gay, lesbian, or trans- gender. Overall, 22 respondents (42 percent) said they had been verbally abused and 17 (33 percent) had been threatened or attacked because of their race or ethnicity. Of the twelve racial minority (excluding Aborigi- nal) respondents, nine had been verbally abused and six had been at- tacked or threatened with harm. Of the 16 Aboriginals who answered the question, nine had been verbally abused, and five had been attacked or threatened. More of the racial minority and Aboriginal respondents re- ported abuse and attacks or threats. Being homeless was the most common reason given for the abuse, except among racialized respondents, for whom it was more often their ethnicity or race. More than half the respondents (56 percent) said they had been verbally abused due to prejudice against homeless people. Al- most one-third of them (31 percent) had been attacked or threatened on this basis. More women than men said they had been verbally abused, attacked, or threatened because they were homeless. Ten respondents, more of the youth than the adults, said they had been verbally abused because they were perceived as gay, lesbian, or transgender. Four had been attacked or threatened on this basis. These responses indicate a high degree of vulnerability to mistreatment. Vulnerability and self-protection Respondents outlined some of the dynamics that contribute to their vul- nerability to mistreatment. Meeting basic needs becomes a challenge when a person is homeless ? access to a bathroom, a restaurant, hospital care. This contributes to a poor appearance, poor health, low self-esteem, and the stigma of ?not belonging? ? being outside what is considered a regular community of people with homes and jobs. The outcome is a feeling of isolation and alienation from (and by) the mainstream. High visibility combined with social exclusion put respondents in a weaker position relative to authority figures such as the police. Respondents used various tactics to protect themselves. Most indi- cated they maintain a general wariness and watchfulness. Some carry a potential weapon, such as a box cutter, knife, razor, spray perfume, or mace, for which they may be charged by the police. One man kept a large dog. They avoid certain shelters, areas, houses, or people; they stay near the few people they trust. They use storage lockers, carry their valuables and belongings everywhere, or store them with a friend or agency (especially their identification). They sleep face down, sleep in a group taking turns as guard, or walk around and stay awake all night. Several respondents said they were avoiding ?old haunts? where they had ?gotten into trouble,? usually involving drugs. They stay in familiar areas and try to avoid trouble. Conclusions Our survey found that homeless individuals appreciate the need for law and order, but are highly critical of perceived unfair policing practices, especially differential treatment of racialized persons. Also, although homeless individuals experience a high level of victimization, they are quite reluctant to report crimes to the police and feel alienated from po- lice protection. This is a preliminary study, and more research is needed. For in- stance, an understanding of police officers? views of homelessness would contribute to a productive dialogue of how they deal with homeless in- dividuals and the role of policing in dealing with a major social and po- litical problem such as homelessness. Further research would also draw out ideas on reducing some of the problems we identified. Better security may be needed in shelters, where bullying, intimidation, and theft commonly occur. There is also a role for mental health and addictions treatment for homeless people, advocacy services (including liaison workers at courts and detention centres), bet- ter discharge planning, special training for police, and affordable, sup- portive housing. Sylvia Novac is a research associate at the Centre for Urban and Community Studies, University of Toronto. Joe Hermer is a Professor in Criminology at the University of Toronto. E mily Paradis is a doctoral candidate at the Ontario Institute for Studies in Education, University of Toronto. A mber Kellen is Su- pervisor of Advocacy/Community Programs at the John Howard Society of To- ronto. Chapter 7.3 Homelessness, Incarceration, and the Challenge of Effective Discharge Planning: A Canadian Case STEPHEN GAET Z AND BILL O ?G RADY Canada has the fifth-greatest incarceration rate in the world (Griffiths, 2007).1 While a ?get tough on crime? approach may put more people in jail, the vast majority of individuals who wind up in jail will eventually be released into the community. Whether one sees the goal of prison to be Over the past several decades, we have seen calls for reforms to the criminal justice system that emphasize harsher prison sentences. This has included calls to reform the Young O ffenders Act (now the Youth Criminal Justice Act), to place greater restrictions on bail, to eliminate house arrest (conditional sentences), and impose minimum sentence prison terms. The federal government has announced that it has plans to build more prisons in Canada. Critics of ?law-and-order? approaches to crime contend that such measures can increase the prison population, both by incarcerating more people, and also by lengthening the amount of time inmates remain in prison. 1 Yet Canada?s incarceration rate in 2003-2004 was at its lowest since 1981-1982 (The Daily, Statistics Canada, December 16, 2005). punishment or rehabilitation, from a public safety perspective, all agree that it is desirable that those who are released from prison successfully integrate into communities, and that they not reoffend. Since almost all prisoners ? whether sentenced or in remand2 While the nature of homelessness may differ in many regions, it is becoming apparent that across Canada, many inmates discharged from prison are winding up homeless, and conversely, that many homeless people end up in prison (Novac et al., 2006; Vitelli, 1993). Without access to housing and employment, without support for problems such as ad- dictions and mental health, inmates released from prison are at risk of becoming homeless. At the same time, people who become homeless are at increased risk of becoming involved in the criminal justice system. This ?revolving door? syndrome is fuelled as local jurisdictions continue to adopt and enforce policies and ordinances such as the Safe Streets Act in Ontario (2000) and Vancouver (2004) that criminalize activities such as sleeping and panhandling in public spaces. Once into the criminal justice system, incarcerated or detained, many people leave jails and prisons without a destination, and bereft of the resources necessary to secure housing or social supports. ? will one day be released, how we prepare prisoners for release has important implications for communities across the country. Inadequate transitional supports may increase the risk of recidivism for inmates, undermining a key goal of corrections, and jeopardizing not only the health and safety of inmates, but of all Canadians as well. This is where the relationship between incarceration, prisoner re-entry must be considered. If a goal of corrections in Canada is to make Canadians safer through reducing recidivism, then we need a deeper understanding of the relationship between incarceration, prisoner re-entry, and homeless- ness. While a considerable body of research exists on the nature and ex- tent of homelessness, specific research on the relationship between homelessness and incarceration in Canada is limited. 2 Remand refers to accused people who are placed into custody while await- ing a further court appearance. The study We conducted research in 2005?06 in Ontario and British Columbia, us- ing both quantitative and qualitative methodologies. Our investigation was guided by the following research questions: ? What is the nature of discharge planning in provincial correctional facilities (in Ontario)? What are the policies that frame discharge planning practice? How do provincial employees within the correc- tions system manage discharge planning? ? What are the key similarities and differences between the prison population and the homeless in Canada? Are the backgrounds of homeless and domiciled prisoners and releasees similar? To what degree do releasees ? whether housed, underhoused or homeless ? share risks associated with housing, employment, health, mental health and substance use? ? How do inmates and releasees experience discharges? Is discharge planning seen to have an impact on the lives of releasees? How do releasees envision a more effective discharge planning process? Data and Methods We collected information from male informants who are or have spent time in provincial corrections facilities in Ontario or British Columbia, including inmates and releasees. All interviewees had some experience with discharge planning (or the lack of it) in provincial facilities. In addi- tion, we interviewed personnel involved in the discharge planning process in both British Columbia and Ontario. Inmates refers to persons currently in provincial corrections facili- ties who have been convicted of a crime, and who are awaiting release. One provincial institution in western British Columbia, and one provin- cial institution located in central Ontario were selected for the study. Releasees refers to persons who are no longer in prison, but who were recently held in provincial corrections facilities either as convicted offenders or while on remand. The population of releasees interviewed includes: ? Housed releasees ? individuals who have stable, permanent housing. ? Underhoused releasees ? individuals with a precarious housing situ- ation, temporarily staying with friends or relatives and paying rent, or in treatment at the time of the interview. ? Homeless releasees ? individuals without stable or permanent hous- ing who are sleeping outdoors or in temporary shelters. Planners refers to persons who are responsible for ?discharge plan- ning? in provincial corrections facilities, either as corrections employees or as volunteers. Fourteen planners were interviewed for the study. In exploring the experiences of discharge planning from the pers- pectives of inmates and releasees (both domiciled and homeless), we found a lack of congruence between policy and practice, and between the perceptions of discharge planning by those who provide it, and those who receive it. Correctional programs for provincial inmates In Canada, provincial governments are responsible for persons convicted of crimes and sentenced to less than two years, as well as the care and incarceration of people awaiting trial (on remand). Programs and servic- es available for convicted prisoners are not the same as those for people on remand. We interviewed those responsible for discharge planning in both provinces, but the focus of analysis here will be on the policies and practices of correctional programs and discharge planning in Ontario. The policy directives of the Ontario Ministry of Community Safety and Correctional Services (Commissioner?s Directive 726) include the mandate: ?to ensure that correctional programs meet the identified needs of offenders and promote successful reintegration.? This includes ?in-prison? supports as well as discharge planning. Prisoners who are convicted of a crime are provided with a ?correctional plan? that identi- fies needs and makes suggestions regarding programming. Psychologi- cal assessments are mandatory only for prisoners convicted of violent crimes and high-risk sex offenders, but not those deemed to have other mental health problems. Prisoners with identified substance use prob- lems are referred to counselling. Correctional programs provided within prisons include work/skills training, education, counselling, and programs for substance use, vi- olence, and sex offenders. Participation is voluntary and requires in- formed consent, but prisoners are encouraged to take part in such pro- grams and early release may be tied to such participation. For sex of- fenders, post-treatment assessments are done to determine the effective- ness of treatment and to assess risk. The Province of Ontario has procedures for releasing prisoners. All prisoners convicted of a crime receive information on the conditions of their discharge, a ?certificate of release,? and identification. They are is- sued any savings they may have accrued while in prison, and released with a minimum of $50 in hand. They are released into the community in which they were arrested in or to their last place of residence, and if ne- cessary, police inform specific community services of the impending re- lease. Those who receive conditional releases (parole, work release) may receive additional community supports. The Ontario Government has no mandate or responsibility to provide programming and services in the community once a prisoner has completed a sentence, nor for those who were on remand but who have been released. Discharge Planning for Provincial Inmates While discharge planning is not officially mandated for all prisoners in Ontario (it is available only for those convicted of a crime), the Ministry of Community Safety and Corrections apparently intends that prisoners should receive some form of support and planning by corrections staff prior to discharge. However, because different institutions are responsi- ble for the development of program plans and delivery, the nature of discharge planning varies from institution to institution.3 Discharge planning is provided for inmates in Provincial Correc- tions Facilities by corrections staff and by community-based groups such as the John Howard Society and Ontario Multi-Faith Council on Spiritual and Religious Care. The community-based groups provide both in- 3 While the provincial government has been criticized for a lack of a standard policy throughout the province, best practices guidelines for discharge plan- ning, including areas that focus on offenders with special needs, have been de- veloped (Ministry of Community Safety and Corrections, 2006). For this project, we were not able to evaluate the implementation of this new policy di- rection. prison counselling regarding discharge, as well as some transitional supports in the community. Changes in criminal justice policy relating to corrections in Ontario that have occurred since the mid-1990s have had an impact on discharge planning in provincial prisons and supports for prisoner re-entry. Many of these changes were the result of reforms by the Conservative govern- ment to reduce costs and move towards ?no-frills? prisons. The result has been that fewer inmates in Ontario have access to discharge planning and transitional supports. Some of these changes include: ? Reduced program support for discharge planning. ?Funding for many community-based programs and services for offenders have been eliminated or reduced. These include programs geared to: employ- ment, family counselling, men who are abusive, community youth support, diversion and discharge planning for those being released? (John Howard Society, 2000:1). It is not clear how many of these pro- grams have been replaced or enhanced since 2000. Reduced funding means that non-governmental organizations have less capacity to en- ter prisons and engage in thorough discharge planning with high- risk prisoners, and that there are fewer resources in the community to which discharge planners can refer releasees. Cutbacks for parole and probation officers means higher case loads which compromises their ability to offer effective supports (Government of Canada, Na- tional Homelessness Initiative, 2005). ? Reduction in parole and conditional release programs. Since the mid - 1990s, the use of conditional release programs for provincial inmates has declined dramatically. For instance, in 1993-94, the average count of convicts on provincial parole in Ontario was around 1,800, and in 2004-05, this number dropped to fewer than 200. ? Elimination of halfway houses. The Ontario government has eliminated halfway house and transitional housing programs for provincial prisoners. According to the John Howard Society of Ontario, this change limits released inmates? opportunities to ?maintain commu- nity and family responsibilities, such as employment, school, or child care? (John Howard Society, 2000: 1). Instead, the government ex- panded electronic monitoring. While ?house arrest? gives restricted offenders some access to the community, electronic surveillance does not normally involve programs and services designed to reduce of- fending, which were a major component of the halfway house sys- tem. In addition, people who are homeless are clearly not eligible for house arrest. ? Increase in remand population. Persons on remand are held in custody awaiting trial, and are therefore presumed innocent. Since the early 1990s, fewer persons charged with offences have been granted bail, leading to a steady increase in the size of the remand population; at the same time, the number of persons convicted of offences has de- clined. In addition to the increase in the percentage of persons held in Provincial Corrections Facilities being held on remand (at present, they make up 60 percent of the prison population in Ontario) the av- erage length of time on remand has increased from 22 days to 34. Those on remand for even a short period of time whether convicted of a crime or not, experience disruptions to their day-to-day lives, in- cluding the potential loss of housing and employment. People who are homeless are more likely to be held on remand, as bail conditions often require a home and employment. In the 1990s, Ontario?s Minis- try of Community Safety and Correctional Services deemed that per- sons being held on remand are not eligible for programs available to sentenced prisoners, including discharge planning supports. How- ever, in some prisons, inmates on remand have access to supports. Staff Perspectives on Discharge Planning in Ontario Provincial Corrections Facilities We interviewed five discharge planners in Ontario provincial facilities, all of whom were provincial employees, including four social workers and counsellors responsible for substance use and sex offenders, as well as a ?transition coordinator.? Because different facilities organize their correctional programs (including discharge planning) in different ways, we recommend caution in generalizing from these findings. As a group, they spend between 15 minutes and an hour with each releasee. The intensity and depth of the discharge planning varies from planner to planner, based on their job description and the resources to which they have access. This means that there is no consistent approach to discharge planning, and different inmates receive different levels of support. The transition coordinator, presumably because of her job de- scription, does the most intensive work with releasees. Discharge planning may involve the identification of needs, sharing of information, establishing community linkages where needed, ensur- ing accommodation support and providing other supports within the community to assist in the transition. However, given large caseloads and the lack of resources in prisons and in the community, the bulk of discharge planning involves merely the sharing of information, rather than more intensive planning involving contact with and establishment of appropriate and necessary supports in the community. Planners may try to make referrals to major institutions in the community, such as Al- coholics Anonymous, the Centre for Addictions and Mental Health, the Canadian Mental Health Association, Ontario Works, or the reintegra- tion programs of the Multifaith Council and John Howard Society. The areas covered in discharge planning may include employment, mental health and substance use issues, family contacts, and accommodation. Employment supports arranged by discharge planners vary consi- derably, ranging from nothing at all to providing the person job ads from the Internet. Arrangements with Ontario Works are sometimes set up prior to release. Some planners strive to provide more intensive support: Yes, I will make telephone calls to employers for them or allow them to make telephone calls to employers. I?ll counsel them about the appro- priateness of employment, i.e., if a guy has a substance abuse problem and wants to go back to an environment that might trigger him. (Social Worker 4, Ontario Provincial Facility) In other cases, participation in in-prison programming may help. (Our facility) has a bricklaying pre-apprenticeship program. ? If they complete this program, inmates leave with a job, hard hat, work boots, housing secured, and Ontario Works. They are then monitored by Educa- tion Department while out for next 8 weeks to ensure they are coping. (Social Worker 5, Ontario Provincial Facility) Convicted sex offenders or prisoners who have been identified pre- viously as requiring support for substance use problems or anger man- agement may receive additional discharge planning support from the counsellors they have been working with while incarcerated, in order to establish supports in the community. With sex offenders, we notify them [inmates] of the sex offender registry, review conditions, and remind inmates of them. (Social Worker 2, Ontario Provincial Facility) Although inmates may get some support for substance use while in prison (one prison had three substance abuse counsellors), referrals to the community (which seem to focus only on treatment rather than harm reduction) are often difficult to arrange, given not only the lack of staff resources within prisons, but also the challenges of finding and obtain- ing timely and appropriate sources within the community. More staff are required ... at my institution we are doing what we can, but resources are limited. Linkages to counselling prior to release are non- existent. Bed availability and timing to get inmates into treatment is not effective. (Social Worker 4, Ontario Provincial Facility) Help in contacting family is available if the inmate requests it. However, in some cases this can lead to problems with accommodation, if spouses, partners, or family members do not wish the inmate to move back with them. The range of supports for obtaining post-release accommodations is limited. While one planner attempted to assist all her clients in obtaining housing, she was unusual. Several of the planners simply provided re- leasees with a list of shelters and hostels in the area they were to be dis- charged to. If prisoners do not have pre-arranged accommodation or an expectation of finding housing through family or friends, the main dis- charge planning strategy seems to be to direct releasees to the shelter system for homeless people. In spite of the importance of accommoda- tion as a transitional support, several staff indicated they do not have the resources to do an adequate job in this area. For me, it?s a time thing and I have never been directed to do that kind of work. I don?t think here it has been deemed a priority. (Social Worker 2, Ontario Provincial Facility) Yet all the planners we spoke with see the value of discharge plan- ning and transitional supports, and believe that if done effectively, can have a positive effect on recidivism. If an effective plan is not set up that gives the inmate an opportunity to use that for an excuse to offend again. Inmates are overwhelmed with ad- justment to community. They need an effective discharge plan. (Social Worker 5, Ontario Provincial Facility) At the same time, the planners recognize that under the current conditions, it is extremely difficult to meet the needs of releasees. More staff are required. At my institution we are doing what we can, but resources are limited. Linkages to counselling prior to release are non- existent. Bed availability and timing to get inmates into treatment is not effective. (Social Worker 4, Ontario Provincial Facility) From the remand side of it, we technically do not have a discharge process. This is very frustrating. For the guys, they can?t get their proper- ty, court may have not even been in the town they come from, they are stuck with no property, no money, no transportation. (Social Worker 1, Ontario Provincial Facility) Inmates, Releasees and the Discharge Planning Process How do inmates experience the discharge planning process? To address this question, we interviewed 38 inmates and asked about their custodial discharge experiences. All told, 35 percent of the inmates reported that they had met with a staff person in the institution who discussed a dis- charge plan with them. Inmates in this sample were all entitled to dis- charge planning, since all were serving time under sentence, not under remand. Moreover, most were approaching the date of their release. There were few differences in the percentage of British Columbia and Ontario inmates who reported receiving discharge planning from institutional staff?about one-third in each sample. There is some evi- dence to suggest, however, that inmate drug use may be associated with receiving some kind of discharge planning in both provinces. In British Columbia, 83 percent of those inmates who had been in contact with a discharge planner also reported having been heroin users before their incarceration. While there were relatively few heroin users in the Ontario sample, 67 percent of those who received discharge planning had used cocaine or crack before being incarcerated. On the other hand, inmates who had not reported being drug users in the past were less likely to have a discharge plan. It appears that institutional drug counsellors are involved to some extent in the discharge planning process. There were significant ethno-racial differences in the reported expe- rience of the discharge planning process as well. Aboriginal inmates re- ceived less contact with discharge planners than non-Aboriginals. While 11 of the 27 non-Aboriginal inmates indicated to our interviewers that they had contact with a staff person to discuss a discharge plan, only 1 of 7 Aboriginal inmates reported this type of pre-release contact. These dif- ferences, however, can be partly accounted for by the fact that fewer Ab- original people in either the Ontario or B.C. samples were cocaine or he- roin users, compared to the non-Aboriginal population. All Aboriginal inmates reported being heavy drinkers prior to their incarceration, but most did not receive any assistance with their release. Inmates with drug problems seem to attract more attention from authorities than men with drinking problems. Age, sentence length, and number of past convictions appeared to be unrelated to whether or not an inmate received a discharge plan, al- though the sample is too small to make generalizations. But the connec- tion between illegal drug use and discharge planning should be a focus of further research. We interviewed only inmates who were serving court-imposed sen- tences. But relatively few inmates in Ontario and British Columbia pro- vincial correctional institutions receive systematic release planning, be- cause so many inmates are in prison on remand. Also, inmates who have been serving time on remand and then are given short additional sen- tences by the courts may also miss out on a formal discharge plan due to problems of timing and organization. As one inmate told us: I don?t see a lot of remand inmates who have some sort of access to health and abuse programs, even though they are not sentenced. A lot of people come here and do remand, and get their time served. In my situation, I got a 90-day sentence, since my remand time, I only have 2 weeks left. It?s a disadvantage. It takes longer time to set up discharge plans, but it?s not the inmates? fault. (B.C. inmate) Inmates who received some discharge assistance reported a number of areas that were addressed. In Ontario, assistance with finding housing was an area where inmates reported the most attention. In B.C., consid- eration was most likely to be directed at substance abuse issues and sup- ports upon release. Employment, family reunification, mental health, and physical health counselling were also considered in the discharge process, according to our interviewees. Releasees and the Discharge Planning Process We also explored the experience of discharge planning from the perspec- tives of those who are now released from prison. This included those who are housed, those who are underhoused, and those who are home- less. All interviewees had been incarcerated in provincial facilities in the previous six months. We asked currently housed recent releasees from Ontario and Brit- ish Columbia if anybody in the institution from where they were last released had been responsible for their discharge planning. Less than half (40 percent) in both provinces reported that they had seen a staff discharge planner before the completion of their sentence. Of the un- derhoused releasees, 45 percent had received discharge planning (the higher percentage of underhoused receiving discharge planning may reflect the fact that several are in treatment). Among the 52 homeless releasees living in shelters or on the streets who had recently been incarcerated in an Ontario or British Columbia correctional facility, only 15 (29 percent) reported that they had met with a correctional staff person who had talked to them about a release plan. Releasees who did receive discharge planning ? whether housed, underhoused, or homeless ? were asked to state how useful the planning had been in terms of employment, housing, family reunification, health, and dealing with substance abuse. Some found discharge planning to be quite helpful: The John Howard Society [representative] helped with getting access to my vehicle and my personal property that had been seized by police. She made arrangements with a friend of mine and contacted three detach- ments of the RCMP and acted as a liaison and arranged from my friend to be able to go down and pick everything up. I didn?t know about JH be- fore, so it opened the door for getting help for things that were sort of beyond my control at that time. (British Columbia releasee ? currently housed) For releasees who received planning, the area that was typically found to be most the effective was that of substance abuse. The only discharge planning that I had which had success was at the Don Jail ... pretty good, too, because a lady came to see you a week before you get out. And had everything for you before you got released, like welfare, you gotta pick your cheque for your housing, and you go to get your money, and get your money for the housing for the next week or two... a social worker came in ... also someone from the John Howard. (Ontario releasee ? currently homeless) However, most releasees ? whether housed, homeless or underh- oused ? had not received a discharge plan prior to release. No discharge planning ? that?s it, you?re free to go, gave me a token and told me to go. (Ontario releasee ? currently homeless) No [discharge plan]. I?ve been to jail when I was younger, 15-16 times ? never once has anyone ever asked me where I lived. This is how I ended up on the streets several times. (Ontario releasee ? currently homeless) Regardless of government policy for discharge planning for provin- cial inmates, and regardless of existing resources within correctional fa- cilities and provided by non-profit groups, the vast majority of releasees in this study either did not receive discharge planning, or did not expe- rience whatever help they did receive as a transitional support. Challenges of Post-Prison Release Many releasees face challenges in making the transition from prison to the community, and people with a history of homelessness are particu- larly vulnerable. We collected data on the experiences of housed, un- derhoused and homeless releasees in the areas of housing, income, health (including mental health and addictions), and relationships. Housing Never in my life have I ever been in a prison in Toronto where someone has offered to assist me in finding things like housing before I get out of jail. (Ontario releasee ? currently homeless) Obtaining adequate housing is a challenge for many releasees, and the inability to do so is a defining feature of homelessness. Individuals who are poor, who have a previous record of homelessness, and who have weak (or nonexistent) attachments to family and friends are particularly at risk. We selected individuals for whom obtaining and maintaining housing is a challenge: 45 percent of our sample of releasees were home- less at the time of the interview, 20 percent were underhoused, and 34 percent were housed. The category we have defined as underhoused is important. These are people who currently have shelter (staying with friends or family, temporarily in treatment, or in rental accommodations they are having difficulty maintaining), but for whom their tenancy is very uncertain. When I was released in November 2005, I stayed with family members in London for a few months. I then came to Barrie in January and stayed with a friend ? no rent ? for one week, and another friend?s for a few days. We went to a shelter for about 3 weeks and I?m now staying at a friend?s house again. (Ontario releasee ? currently underhoused) Of those defined as housed or underhoused, 80 percent reported be- ing homeless at some time in the previous six months or since being re- leased from prison; 43 percent reported that since being released from jail they had found it difficult finding a place to stay at least one a week, and 34 percent said that it was a problem they faced every day. Of those who were currently housed, 65 percent had stayed in a shelter on at least one occasion during the last six months. Few releasees overall, then, re- ported a high degree of stable housing since being released. One consequence of being released from prison with inadequate housing supports and minimal income is that people often wind up hav- ing to live in unsafe neighbourhoods, where they are at risk of being vic- timized or being drawn into criminal situations. In our survey, most homeless releasees (71 percent) described their neighbourhoods in unfa- vourable terms, compared with those who are underhoused (29 percent) and those who were housed (22 percent). No matter what the weather is like, there?s always some kind of commo- tion outside. So it?s a nonstop, crazy-paced, stab-you-for-2-cents-in-your- pocket kind of neighbourhood... (Ontario releasee ? currently homeless) Many releasees argued that when releasing inmates, consideration should be given to where they are released. Releasing an inmate into the community and placing him in a ?bad? environment was considered inappropriate. They should give people a choice, like if you want to go into a place where there?s not a lot of drugs around? I don?t want to go into a house where people are smoking crack and drinking Listerine and doing all kinds of crazy behaviour. That doesn?t do anything good for you. It brings you down even more. (Ontario releasee ? currently housed) Access to housing was viewed as a fundamental first step for suc- cessful prisoner reintegration. Housing planning should be mandatory. If they keep wondering why people re-offend, they shouldn?t be allowed to release anybody who has no address to go to. ? Every person in jail [should have] a case worker and they should be on getting somewhere to live, even if its just a halfway house. The crime rate would drop significantly, just from re-offenders. (British Columbia releasee ? currently underhoused) Finally, releasees were aware of the importance of arranging ac- commodation before release from remand. That is, the process of being release in such situations often means a direct discharge from the court- house, without giving the inmate the opportunity to recover belongings or the necessary resources to make arrangements for transition into the community: Having somewhere to go instead of being released onto the street is im- portant. People that get released from jail don?t really have nothing, and they have no place to go. And they?re right back into the crime. I know when I got released before, I got released from the courthouse, and, you?re kind of out there, and you do whatever to survive, right? (British Columbia releasee ? currently housed) Income: Jobs, benefits, and making money The releasees we spoke to are underemployed, regardless of their current housing situation, with only 27 percent currently having a job, and 57 percent reporting any kind of employment since being released. Only two-thirds of the sample who had jobs reported full-time employment. The unemployment rate for the releasees (73 percent) can be compared to the national unemployment rate which for the past six years has ho- vered around 7 percent. Those who were housed were much more likely to be employed (40 percent) than those who were homeless (22 percent) or underhoused (17 percent). Those who have some kind of shelter are also more likely to access government benefits such as welfare and disability support than are releasees who are homeless. For inmates at risk of becoming home- lessness, helping to facilitate obtaining necessary government benefits should be seen as a priority transitional support. Releasees are well aware of the challenges of generating income once released. They spoke of the importance of correctional programs relating to employment, and of the necessary role that income plays in a successful transition to the community. Several said how the lack of in- prison supports and discharge planning reduces their ability to generate income and avoid crime. A lot of people end up going back to the institution 10 days after they are released, because they got no choice but to do crime to get money in their pocket to survive. Takes three weeks to get an appointment with social services alone, and they won?t even accept you, you can?t book an ap- pointment when you?re in jail, you know. (British Columbia Releasee ? currently underhoused) Given the high rate of unemployment and the fact that only one- fifth of the sample is receiving government benefits of any kind, it should not be surprising that a percentage of releasees ? whether housed or homeless ? engage in non-conventional ways of making money, rang- ing from quasi-legal (mostly panhandling or squeegeeing) to criminal activities. (In the interviews, 83 percent of the inmates and those inter- viewed in the community reported that on at least one occasion they had broken the law in order to survive.) These strategies are also used by some releasees who have housing, suggesting the degree of financial stress they also experience. Finally, releasees spoke of the importance of also providing material supports at the time an inmate is discharged. Many felt that discharging inmates without proper funds, transportation, or proper clothing put them in a difficult situation: [When I was released] some money would have been nice. Some $200 ? 300, because I was released on a Saturday on a long weekend, the bus ticket was all paid for ... but [nothing] to buy personal things, deodorant, toothbrush, all that... basic things. (Ontario releasee ? currently underh- oused) One of my friends was in the clothes he came in with. And he was re- leased in the winter, and went in the summer... (Ontario releasee ? cur- rently housed) Health, Mental Health and Addictions The homeless population is characterized by poor health, which is both a cause of and a product of being homeless. When asked to evaluate their health compared to other people their age, 37 percent of homeless relea- sees rated their health status as ?poor to fair,? which was worse than either those who were underhoused (25 percent) and those who were housed (22 percent) releasees. Moreover, over half of all releasees (51 percent) report not getting enough food at least once a week. Overall, one-third of all releasees identified themselves as having a disability, with domiciled releasees being the least likely to report this. Many inmates may be eligible for disability benefits and unable to work. But only 20 percent of releasees we interviewed were on government benefits, and in most cases this was welfare, not disability benefits. Compromised mental health is another factor associated with the homeless population, and also a risk factor for incarcerated individuals. Of our total sample of releasees, 29 percent indicated a past diagnosis of a serious mental illness, including schizophrenia (3.4 percent), bipolar disorder (15.5 percent), personality disorder (10.3 percent), and fetal al- cohol syndrome (4 percent). Relationships Relationships are important for successful transitions to the community. Released inmates who have strong relations with family, a partner or friends are more likely to obtain shelter (even if it is temporary and tran- sitional) by moving in with family members (parents, relatives, siblings). Family relationships ? whatever their form ? not only mean that relea- sees have a home to return to, but also may form a basis for support in obtaining employment and in dealing with other post-release challenges. The absence of relationships ? especially quality relationships ? can place releasees at greater risk of homelessness: If they don?t have family out there, or they don?t have money or their not established out there, they need for the first month or at least, some kind of assistance with respect to shelter ... Some people get out and they?re lost, you know, they got to do all this stuff and they?re overwhelmed, es- pecially if they have been in for a while. (British Columbia releasee ? cur- rently housed) Several releasees in interviews discussed the importance of relation- ships in helping with transitions back to the community, and even the necessity of family for obtaining bail: I?m living with my family again. My wife works part time, and we have a joint bank account. I work driving a cab, doing odd jobs. I?ve been at the present address for two years. (Ontario releasee ? currently housed) Their bail program makes no sense to me... if you?re homeless, and have no family here, then you?re stuck in jail. (Ontario releasee ? currently homeless) Some interviewees reported supportive relations with family mem- bers, but others reported strained relations. My family doesn?t want anything to do with me since I?ve been released ? I?m the ?black sheep? of the family. (British Columbia releasee ? currently underhoused) Friendships are another important type of relationship. Friends can provide support, including temporary shelter. However, many releasees said they don?t have close friends or that friends are a problem, rather than a potential source of support, because their friends are involved in crime or are substance users, and continuing a relationship with them once released jeopardizes their ability to avoid jail. I?m kind of distant [ from my friends]. I don?t associate with them an y- more. They were a bad influence. (British Columbia releasee ? currently underhoused) The Case for Discharge Planning: in their own words. We asked participants if they had any thoughts on what could be done to improve the discharge process. They are aware that the absence of strong transitional planning places them at risk for homelessness, addic- tions, recidivism, and reincarceration. They make a strong case for the necessity of discharge planning and transitional supports. They should have more people going around in the jails and helping people. People?s heads aren?t in the right place at the time of release. No- where to go, no assistance, so you just spend the money ? at the bar. (Ontario releasee ? currently homeless) Make sure inmates have a place to go or live, give inmates counselling to not get back in trouble and to look at why you were in there. Have wel- fare appointments ready for them. (British Columbia releasee ? currently housed) While there, you are given clothes, food, shelter, and you become sober. Upon release, you are given just the sobriety and a piece of paper? we didn?t have follow-through with the drug counselling, mental health counselling, life management skills, how you manage your time and money, and, possibly teach us how to make resumes. (Ontario releasee ? currently homeless) They should contact social services tell them/verify his identity in order for a release to get a welfare check upon release. They should not release you on Friday or a weekend. (Ontario releasee ? currently homeless) When asked who should provide these services, many respondents said that the correctional institutions should be required to ensure that inmates are released with the proper supports. Moreover, several partic- ipants mentioned that there should be more coordination between cor- rectional institutions and social services. Consider these words: Where do we put the responsibility? Do we leave it up to corrections? Corrections has a mandate that once your sentence is up, they are no longer responsible for you. And just because I don?t have a welfare file set up or whatever, it doesn?t mean that they are not going to let me go. They gotta let me out. So, based on their budgets and their policy, their hands are tied. They can?t keep us after about release date. Again, the support workers up at the jail, there aren?t enough of them to go around. If you are in a remand situation they won?t even see you because they?re not under contract to counsel people who have been detained without being convicted and then go to court and then are exonerated or released or given a conditional sentencing to the community where the community is not ready for them, don?t even know they?re coming. Why? Because there?s not enough support staff. It?s madness, I?m telling you. (Ontario releasee ? currently underhoused) Conclusions A shelter is an external jail with more yard time. (Ontario releasee ? cur- rently homeless) Our research suggests a need for reinvestment in resources both within prisons and the community to support prisoner re-entry. There is a bidirectional relationship between homelessness and in- carceration; people who are homeless are more likely to wind up in pris- on and the prison experience itself places releasees at risk of becoming homeless. Allowing releasees to slide into homelessness puts them in a situation where intersecting challenges (lack of housing, lack of income, health problems and nutritional inadequacy, mental health problems and addictions) may make it more difficult to reintegrate and move on with their lives, and as a result may make it more likely that they reof- fend. Different inmates require different kinds of supports once they are released from prison. Some will be considered to be at higher risk than others. High risk clients have more extensive and complex needs, either because of the degree to which their health problems (mental health, substance use, for instance) are acute, or because of the multiplicity of risks that they face. Moreover, the risks that offenders face are linked. For instance, lack of housing affects employability and vice-versa. Per- sons with mental health challenges or concurrent disorders may face ad- ditional barriers to obtaining housing. Thus the potential risks ex- convicts face should not be treated in isolation from each other. One could argue that one of the consequences of cuts to social spending and prison reform over the past ten years has been that the task of managing prisoner re-entry has been downloaded from correc- tions services to the homelessness sector in many jurisdictions. Yet homeless shelters, drop-ins, and soup kitchens are not mandated to do the work of prisoner re-entry and reintegration. Staff are not trained in discharge planning. Such agencies are not funded to do this work. Perhaps the most important barrier that needs to be overcome be- fore any meaningful change can occur in social and criminal justice poli- cy is the politics that surround corrections and homelessness in Canada today. Politicians vow to ?get tough on crime,? and during the 2006 and 2008 federal election campaigns the Conservative Party made promises to build more prisons, eliminate conditional sentences, and impose man- datory minimum sentences. At the same time, the media and general public are normally silent when community supports for inmates, such as halfway houses, are reduced or eliminated. When the harm reduction programs such as safe injections sites or needle exchange programs are promoted by local public health officials, the response from the public and some media outlets is often that of hostility and contempt. Moreover, the surge in the number of inmates who await trial in custody has not been the subject of attention in the mainstream media. The public is much more likely to hear reports of inmates who are free on bail committing heinous crimes, than they are to read about the costs involved in incarcerating inmates while in remand. The issue of homelessness in Canada has not been adequately ad- dressed by governments. Even though thousands of people are without safe and healthy shelter in Canada, the issue has not risen to the top of political agendas at either the federal or provincial levels. As such, it is our contention that unless the issues of prisoner reintegration and home- less receive the level of political attention that they deserve, urgent calls for action on discharge planning will fall upon deaf ears. Stephen Gaetz is the Associate Dean of Research and Field Development in the Faculty of Education, York University. Bill O?Grady works in the Department of Sociology and Anthropology at the University of Guelph. References Griffiths, C. (2007). Canadian Criminal Justice (3rd edition). Toronto: Thompson. John Howard Society of Ontario. (2000). The Changing Face of Ontario Corrections: An Assessment. Fact Sheet # 15 . Novac, S., Hermer, J., Paradis, E., & Kellen, A. (2006). Justice and Injustice: Home- lessness, Crime, Victimization, and the Criminal Justice System. Toronto: Centre for Urban and Community Studies with the John Howard Society, Research Paper 207. Toronto: John Howard Society of Toronto. Ministry of Community Safety and Correctional Services. (2006). Accessibility Plan 2005 ? 200 6 . Toronto: Queen?s Printer of Ontario. Vitelli, R. (1993). The homeless inmate in a maximum-security prison setting. Canadian Journal of Criminology 35(3):323-31. Chapter 8 .1 A Longitudinal Approach to Research on Homelessness F RAN KL ODAWSKY , TIM AUBRY, REBECCA NEMIR O F F , CRISTINA B ONETTA , AND ALETTE WILLIS It is ... time to rethink directions, and to become more ambitious in our ef- forts ... We need to harness a variety of theories and methods to enable us to understand nuanced ?place -specific happenings as well as more struc- turally-determined ?space -compressing? processes... (Wolch & Philo, 2000, p. 149). Wolch and Philo?s rallying cry to geographers who study mental health might equally apply to Canadian homelessness researchers. Their observations about a disjuncture between broad quantitative analyses of incidence on the one hand, and small, in-depth analyses of particular populations or problems on the other, might well describe extant Cana- dian research. It is our assertion that conducting carefully constructed, systematic studies of what happens to homeless people over time is one way in which to overcome this disjuncture and contribute new insights to homelessness research in Canada. While the Panel Study on Homelessness in Ottawa is limited insofar as it is a single site study, it does illustrate a number of the features pro- moted by Wolch and Philo (2000). This article discusses some of these features and reports some preliminary findings. This article is divided into four sections. We begin by reviewing longitudinal studies that ad- dress homelessness in the North American context, with an examination of recent academic health and social science literature in this area. Fol- lowing this review, we introduce the Panel Study on Homelessness and highlight some of its key theoretical and methodological features. We then examine the distinctive housing pathways of five groups of indi- viduals who were homeless when first interviewed in 2002-2003 and then interviewed again in 2004-2005, as well as their reflections on what helped and hindered their efforts to find stable housing. The concluding section reflects on the lessons learned from this longitudinal analysis and suggests there may be value in planning and executing a multi-site longi- tudinal study of homeless people in Canada as a next phase of research. Longitudinal Studies of Homelessness Longitudinal studies that collect data on the same people over time are scarce in the current literature on homelessness, and there are few Cana- dian studies of this type. Extant Canadian research is dominated by in- ferences drawn from the American context, without fully accounting for differences in the social safety net, or in health and social service provi- sion and other institutional factors unique to this country. Most profile the population of homeless individuals in general, or describe one par- ticular group, such as youth, women, aboriginal persons, or families. Some studies discuss and describe policies, programs, and services for persons who are homeless. The scarcity of Canadian longitudinal re- search restricts our understanding of why and how people become homeless in this country, what factors help individuals escape homeless- ness, and how to evaluate the effectiveness of programs, services, and supports developed to address and end homelessness. Our review of the Canadian research literature found only two studies that followed a cohort of homeless persons over time. In Toronto, Goering et al. (1997) conducted a longitudinal study to evaluate the ef- fectiveness of outreach services by examining their impacts on homeless single male and female adults with psychiatric disabilities. In Montreal, a group of researchers is following a cohort of street youth to determine the incidence and risk factors for HIV infection and mortality rate (Roy et al., 2000, 2001, 2003, 2004). Among unpublished research reports, one recent longitudinal study in Toronto is also noteworthy (Anucha, 2003). This study had both cross- sectional and longitudinal components, and examined housing stability in participants of housing programs for ?hard-to-house? individuals. The longitudinal component focused on the experiences of a small sam- ple of participants who had received eviction notices and whose housing was at risk. Although these participants had successfully maintained their housing at follow-up, in-depth interviews revealed a number of challenges. Unemployment and underemployment were cited as major barriers to maintaining housing. Unfavourable living situations, such as sharing accommodations with individuals who sabotaged participants? efforts to battle substance abuse or attend employment training pro- grams were also cited. A review of U.S. research identified several city or county-based studies where data about individuals were collected over a period of time. Some of these studies evaluated the impact of psychosocial inter- ventions for homeless persons with psychiatric disabilities (e.g., Bebout et al., 1997; Humphreys & Rossenheck, 1998; Morse et al., 1992; Toro et al., 1997). Others used a panel study design to examine the use of health and social services (Kreider & Nicholson, 1997; Wong, 1999), the patterns and course of homelessness (Sosin et al., 1990), and the prevalence of HIV and HIV-risk behaviour (Clatts et al., 1998; Sobo et al., 1997). Other American longitudinal research documented the experiences and consequences of homelessness among specific subgroups in the homeless population such as women (Browne & Bassuk, 1997), families (Stretch & Krueger, 1992), children (Clatts et al., 1998; Zima et al., 1999), and men (Concover et al., 1997; Lam et al., 1995). Among these studies, results reporting interactions between such variables as gender, disabil- ity, length of time homeless, and number of times homeless have been inconclusive and inconsistent. Such longitudinal research has drawn on data collected in several locations. Piliavin and his colleagues examined pathways into and out of homelessness for homeless men, women and families in Minneapolis and Alameda County, California (Piliavin et al., 1996; Wong & Piliavin, 1997; Wong et al., 1998; Zlotnick et al., 1999). In New York City, Shinn and colleagues investigated predictors of homelessness and of exiting homelessness for families (Shinn et al., 1998; Stojanovic et al., 1999). More recently, also in New York City, Caton et al. (2005) explored risk factors for long-term homelessness. Many of these longitudinal studies concluded that the provision of economic resources was the best predictor for: (1) avoiding homeless- ness, (2) becoming re-housed, and, most importantly, (3) retaining stable housing after an episode of homelessness. For example, Wong and Pilia- vin (1997) noted that while personal disability, including drug and alco- hol abuse and mental health problems, was associated with a decreased chance of exiting homelessness, institutional resources, such as being employed, or receiving benefits or housing subsidies, were associated with a reduced risk of repeat homelessness. In their study of families using a homeless shelter, Stojanovic et al. (1999) found that the vast ma- jority of those who obtained subsidized housing remained housed dur- ing the study?s entire five-year follow-up period, usually in the same dwelling. In contrast, families without subsidized housing were unlikely to be at the same address at follow-up; nearly half experienced subse- quent emergency shelter stays. Those who initially left the shelter to en- ter unsubsidized housing often reported doubling up with friends or family, resulting in unstable, overcrowded housing situations. Shinn et al. (1998) confirmed the obvious conclusion: provision of subsidized housing was the main predictor of housing stability for such families. Similarly, Zlotnick et al. (1999) found that economic variables, such as receiving consistent benefits and subsidized housing, were the strongest predictors of exits from homelessness to stable housing in a sample of homeless men, women and families. In this study, women with children became rehoused more quickly than the single adults, and were more likely than the other groups to report contact with case man- agers and to receive benefits and subsidized housing. Cohen, Ramirez, Teresi, Gallagher, and Sokolovsky (1997) found that only social support and the number of community services used predicted exits from homelessness in a sample of older women. Piliavin and his colleagues (Piliavin et al., 1996; Wong & Piliavin, 1997) found that having an inconsistent work history was associated with repeated episodes of homelessness, while recent employment and job training were associated with exiting homelessness. The longitudinal literature raises many questions about causes of homelessness and the factors that predict a successful exit to regular housing. There appears to be a growing consensus about the significance of subsidized housing for keeping formerly homeless individuals and families stably housed, and more generally about the efficacy of institu- tional or systemic factors to entrench or mitigate people?s vulnerability to homelessness. In contrast, the individual factors usually associated with homelessness in popular opinion, such as substance abuse, unem- ployment, and mental illness, do not unequivocally predict longer or more entrenched homelessness in these studies and moreover, their ex- planatory power appears to differ among demographic groups. Study Objective Our study examined the range of factors that facilitated or impeded peo- ple who were homeless from becoming housed again. The Panel Study on Homelessness in Ottawa addressed some of the gaps in previous re- search by examining pathways out of homelessness for a relatively large sample, reflecting the diverse subgroups within the homeless population in a Canadian context. To extend understanding of the housing trajecto- ries of people experiencing homelessness, we used mixed methods that included testing quantitative measures as predictors of housing status, and analyzing retrospective qualitative accounts of people?s housing ex- periences subsequent to their episode(s) of homelessness. The present paper focuses on some of the qualitative findings in our study. Method The Panel Study of Homelessness (PSH) emerged out of a history of col- laboration between university researchers, the City of Ottawa, and the Alliance to End Homelessness in Ottawa, a community-based organiza- tion established in mid 1990s to address the increasing problem of home- lessness in the Ottawa area. In line with previous research in the area (Caton et al., 2005; Pilavin et al., 1996; Shinn et al., 1998), ?homelessness? was defined as ?a situation in which an individual or an adult with one or more children under 16 has no housing at all, or is staying in a tempo- rary form of shelter.? Study sample The sample consisted of 412 individuals who were homeless, in five ap- proximately equal subgroups based on age, sex, and family status were interviewed at baseline. These subgroups were: ? single male adults (n=88); ? single female adults (n=85); ? male youth between 16 and 19 years of age (n-78); ? female youth between 16 and 19 years of age (n=78); ? adults within families (n=83) . Overall, the sample was made up of more women (55 percent) than men (44 percent), because of the larger proportion of women who were inter- viewed in the family subgroup. Two respondents identified themselves as transgendered. Further details about the sampling strategy used in the study are outlined in Aubry, Klodawsky, Hay, and Birnie (2003). Of the original 412 people, 225 (62 percent) were re-interviewed at follow-up. In the Panel Study, the success rate varied between sub- groups: ? 49 percent of single men (n=43); ? 65 percent of single women (n=55); ? 63 percent of male youth (n=49); ? 65 percent of female youths (n=50); ? 71 percent of adults in families (n= 58). The response rate was comparable to other longitudinal studies of homeless persons (Aubry et al., 2004), ranging from 49 percent for a gen- eral homeless sample followed for a period of 18 months (Toro et al., 1997) to 90 percent for a sample of homeless individuals with substance abuse problems followed for 12 months (Wright et al., 1995). Follow-up respondents were compared to those who had partici- pated only in the first phase of the study (hereafter called non- respondents). Overall, respondents and non-respondents were similar in terms of sex, age, marital status, education, employment and country of origin. One point of difference though, was that non-respondents were more likely to have reported living in Ottawa for less than 6 months and less likely to have lived in the city for more than 10 years. Measures The Panel Study?s conceptual framework and instruments were strongly influenced by the insights and approaches of the American studies iden- tified above. A combination of quantitative and qualitative interview methods was used in both phases. The measures used at baseline served as the foundation for the follow-up instrument. Interviews for the first phase of the Panel Study included closed and open-ended questions that asked individuals about their demographic characteristics, housing his- tory, health status, and health and social service use and needs (Aubry, et al., 2003). Questions to facilitate tracking individuals over time were also incorporated (Aubry et al., 2004) to address the objective of conduct- ing a second phase of interviews about two years after the first set had taken place. Open-ended questions were incorporated throughout the interview, to encourage respondents to elaborate upon or explain a specific close- ended response or scaled selection. The respondents? answers to the fol- lowing open-ended questions form the basis of the current paper. ? ?Why did you leave the last place you were living?? ? ?What would have been helpful to keep you housed?? ? ?What would help you achieve the best/most desirable place for you to live?? ? ?What prevents you from getting regular housing?? ? ?What helped you find this housing?? ? ?What has been the most helpful in keeping you in this housing?? Procedures Initial interviews were carried out in English (356), French (30), or Somali (14). The services of cultural interpreters were used in 16 other inter- views; they helped participants understand interview questions by ex- plaining them in a language other than English or French. Initial inter- views were conducted in a private area in emergency shelters and drop- in centres. The length of initial interviews ranged from 50 minutes to 150 minutes with an average of 75 minutes. The follow-up interviews were conducted over a seven-month pe- riod, in a similar sequence to the initial interviews. For example, every effort was made to interview those we first met in October 2003, two years later in October 2005. The interviews explored the extent to which individual, interpersonal, and community-level resources contributed to a successful exit from homelessness, and to assess the relationship be- tween housing stability and health. The relationship of respondents? sex, age, family status and citizenship to successful exits, was also explored. Follow-up interviews were conducted in a private location, such as a private room in a community centre or at the University of Ottawa, near the individual?s place of residence. A small number of phone inter- views were conducted to accommodate people living outside the Ottawa region. Follow-up interviews were carried out in English (221), French (13), or Somali (13) and eight others were conducted with the help of cul- tural interpretation services. Interviews ranged from 50 minutes to 150 minutes and averaged 75 minutes overall, with phone interviews taking an additional 10-20 minutes. Results As Table 1 indicates, among respondents, 76 percent were considered stably housed at the time of the second interview by virtue of having been in their own housing for 90 or more days. However, significant variability in housing stability emerged among the subgroups. Almost all the families (97 percent) were housed and had been housed for longer periods (average = 646 days) than those in other groups. Conversely, less than half of the single men (47 percent) had exited homelessness and, on average, had been housed for a shorter period (average = 265 days) than others. Adults in families at 77 percent and single women at 47 percent were the two subgroups most likely to report living in subsidized hous- ing. No men reported living in subsidized housing. We calculated the average number of moves per year for respon- dents, in the period between the initial and follow-up interviews (see Table 1). Adults in families were the most stable group: 53 percent re- ported moving only once within the two-year period, typically from an emergency shelter into permanent housing. Single women were less sta- ble than families but more stable than the other subgroups, with an aver- age of approximately 1.5 moves per year. Single men, and male and fe- male youth were more frequent movers, reporting an average of 2 or more moves per year. Table 1: Comparison of Different Subgroups on Housing History Variables Between Initial and Follow -up Interviews Percent Stably Housed1 Percent Self- Reported Living in Subsidized Housing** * Average # of Moves Per Year Over Study Period*** M (SD) Average # of Homeless Episodes Over Study Period**** M (SD) Adults in Families (N = 58) 97 77 0.87 (0.55) 0.10 (.31) Single Women (N = 55) 73 47 1.45 (0.98) 0.80 (.73) Single Men (N = 43) 47 0 1.97 (1.19) 1.14 (.99) Female Youth (N = 49) 90 16 2.45 (1.35) 0.92 (.90) Male Youth (N =50) 67 14 2.32 (1.36) 1.15 (1.07) Total Sam- ple (N = 255) 76 36 1.77 (1.25) 0.79 (.91) 1 * Chi Square, p < .001; Adults in Families > Single Women, Male Youth ,Single Men; Fe- male Youth > Male Youth, Single Men Respondents considered stably housed if in their housing for 90 days or more. ** Chi Square, p < .001; Adults in Families > Single Women, Single Men, Female Youth, Male Youth; Single Women > Female Youth, Male Youth, Single Men; Female Youth, Male Youth > Single Men *** F=18.67, p < .001; Adult in Families < Single Women, Single Men, Female Youth, Male Youth; Single Women < Single Men, Female Youth, Male Youth; **** F=14.56, p < .001; Adult in Families < Single Women, Single Men, Female Youth, Male Youth There were similar results when the average number of new home- less episodes per year between interviews was examined. Adults in fami- lies reported fewer new episodes of homelessness than did any of the other subgroups. Only six of the 58 adults in families (10 percent) experi- enced homeless episodes after being housed, and each of them experi- enced only one episode of homelessness. In contrast, 33 of the 48 male youth (71 percent) experienced homelessness during the study period after being housed, and 15 of these individuals (31 percent) experienced two or more homeless episodes. These results are in keeping with the overall findings of the study, which indicate that adults in families tend to be the most stably housed of the subgroups (Aubry et al., 2007). Finding and Keeping Housing: Insights from the Respondents A series of overall themes that addressed barriers and facilitators to housing stability emerged from the respondents? comments. These themes were: economic factors; interpersonal supports and conflicts; substance use problems; health status; community services and supports; and housing and neighbourhood quality. Economic factors Many respondents described their housing problems as being closely tied to the high cost of housing and not having enough money for rent. In response to a question about what services were helpful, one man ex- plained: ?Even for those that are employed with minimum wages, the rents are too high. Even when I worked, I came here to eat, because I was making $800 a month and had to pay $600 for rent.? The gap between housing costs and income was particularly pronounced for individuals receiving social assistance but not supplementary Ontario Disability Support Program (ODSP) benefits. Many single adults highlighted the significance of the extra funds they received through this benefit pro- gram. When asked about helpful services, one man replied: ?ODSP, they help pay for this place.? Those in low-wage jobs also expressed frustration at their inability to bridge the gap between income and expenses, or to live so close to the edge that any negative change would mean the loss of housing. For ex- ample, one man explained that the reason he lost his housing was that ?the rent cheque did not come in on time.? Economic problems were ap- parent across all subgroups. For adults with children though, access to subsidized housing was certainly a more realistic possibility than it was for other respondents. As one woman explained, ?It has made all the difference for our family. We are no longer worried about how we are going to pay $1,000 in private housing.? For single youth or adults, the inability to access rent-geared-to- income housing, or other affordable housing, seemed to result in a series of tenuous arrangements that often had negative outcomes. Living from cheque to cheque while dependent on an unsympathetic landlord was one scenario reported by respondents. For example, one female youth explained that she was particularly bothered by a landlord in private housing: ?[The] last landlord lived right above and was always coming over. With community housing, landlords are more understanding.? Another common problem was sharing the cost of housing with others, either by living with roommates or by returning to the family home. While these strategies sometimes were successful, they more often seemed to result in negative outcomes. When asked why he moved out of his last residence, one man explained: ?Everybody was splitting up and going their way. I could not afford to stay with the one unreliable roommate that was staying. If he did not pay his share, I would not have been able to foot the whole rent.? Interpersonal supports and conflicts Many single respondents tried to find and keep regular housing by shar- ing the cost of rent with others, or by returning to their family home. For some, living with others was a reflection of other changes that substan- tially increased the possibility of success. This certainly was the case for one male youth who explained what had helped him stay housed: [I] live with grandparents and girlfriend and her kid, do not feel hom e- less... Wanted to move in because my grandparents are getting old and they need help getting around the house, so I asked them if I could move in. I am done with street life. My girlfriend changed my life. I don?t want her to have a loser boyfriend and she and her kid are relying on me. Also, I see my friends and my brother who are junkies and older and they have kids of their own and I don?t want to be like that. It?s a good place, my grandparents? place, to save some money because they pay for much of the expenses. You don?t have to worry about take-out, more regular meals and regular habits, you can shop for good food to eat, it?s your base, you have a place to relax and set up things, and make it your own. A female youth, reflecting on the reasons she has been able to stay housed, echoed some of these sentiments: ?[I] stopped d oing drugs, so I don't fight with my mom about it anymore; [I] stopped being cocky .? For others, though, continuing differences heightened the chances of becom- ing homeless again. One female youth explained that she wasn't able to do what her parents expected of her because she was depressed and anx- ious, and her parents? way of coping was to kick her out. For those using sharing as a simple cost-saving measure, there were many challenges and a significant chance of failure. One adult male ex- plained the reason he lost his housing as follows: I got sick and tired of my roommate. I would buy food and he would eat it all. (What would have been helpful to keep you housed?) My roommate being more considerate. Some respondents identified a conflict between being housed and retaining ?street friendships,? including one male youth who reflected on the negative aspects of being in regular housing: ?Lot more responsi- bility. Draws me away from my friends who are still on the streets. They show up late at night, but I have to get up at 5 a.m. to work.? Unresolved interpersonal conflicts and experiences of abuse further confounded efforts to stay housed. Among single females, both adults and youth, conflicts with abusive partners was often a reason for leaving regular housing, as captured in the statements of three respondents: I was not getting along with my roommate. Her boyfriend threatened me, so I left. My boyfriend was abusive. He would not help with the rent, he beat me up. He went to jail. I could not afford to stay there. My boyfriend was violent emotionally and sexually, but I couldn't leave because the shelters were full and I was scared of him. Plus I had a dog, which is a problem in shelters. Health status For some respondents, physical and/or mental health issues created cir- cumstances that affected their search for and ability to maintain housing. One adult female, when asked why she left the place she was living, ex- plained the interactions between health status and housing as follows: ?Medical reason ? couldn't live by myself; the Y [YWCA] wanted to move me into a smaller apartment without my own bathroom or kitchen, which increased my panic attacks.? Another male respondent explained his situation as follows: ?The reason that I am homeless is because my wife has chronic schizophrenia. For seven months, I was giving her money for rent, but she didn't pay the landlord. As a resul t, we were evicted from our home.? In other cases, being identified as having physical or mental health difficulties was a route to stable housing. As one male explained in re- sponse to a question about what had been most helpful in keeping him in his housing: ?My disease. I can stay because of that.? The same woman who had to leave the Y located a suitable apartment with the help of her Canadian Mental Health Association worker. Substance abuse problems For some respondents, substance abuse problems were barriers to achieving stable housing. One adult female described the reason she was evicted as ?drinking.? Some individuals tried to address these problems. One single man blamed crack for having to leave the last place he was living: ?People would use my room as a ? dart board.? Other homeless men would always come to my place to hang around and do drugs.? Having greater security in the building would have been helpful to keep him housed, ?if the place was more fixed up so that less people could climb through the window.? One single woman echoed this sentiment: saying that having ?less drugs around; less partying? would have been helpful in keeping her housed. Community supports and services Many respondents identified a significant role for community workers and organizations in helping them find housing. Respondents differed widely in identifying the services and supports that they found helpful. For some, it was a matter of accessing one key targeted service at the right time and place, as in the case of one man and his family: ?I have been getting a monthly cheque from social services.? A female youth told us: ?I turned myself into the CAS [Children?s Aid Society] . I was sick of dealing with being homeless, hungry and cold.? For many others though, the need was for an appropriate, on-going ?basket? of comple- mentary services and supports. For example, one young woman identi- fied a series of steps and support services that she would require. She was living in a young women's emergency shel ter at the time, but she wanted to go home. To do so, she needed to get better grades and con- tinue to go to counselling for substance abuse, but personal problems were preventing this from happening. Some respondents highlighted the value of supportive housing in helping to stay housed. One adult woman told us about the staff on site who had helped her with practical and emotional support. Others em- phasized the complex nature of their challenges and the consequent need for a range of services. One man told us: ?Most shelter residents suffer from substance abuse and mental health problems. There needs to be more resources for these problems as opposed to putting you in jail. More help, more housing, more counseling, more second-stage housing are needed. People like me are sick and we need help.? Housing and neighbourhood quality Some respondents, primarily women, emphasized the significance of certain housing and neighbourhood characteristics for helping them to stay housed. Safety and location were commonly mentioned as a reason for leaving an unsafe area or preferring a safer one. Other respondents identified negative environmental features as a reason for leaving their housing and becoming homeless. One man who lived with his family told us: ?The hydro didn't work. It was cost ing us $400 per month per unit. Prostitutes were doing tricks in front of the house. The neighbour- hood was too far from [my] west -end connections.? Conclusions Even at this early stage, the longitudinal approach has yielded some im- portant findings. One was the differing levels of success among people who have been homeless in achieving housing stability after experienc- ing an episode of homelessness. Specifically, adults in families had more success than other individuals. They also were far more likely to access subsidized housing than the other subgroups in the study. In contrast, single men had the least success in exiting homelessness and none of those who were housed were living in subsidized housing. These differ- ences raise questions about the impact of differential institutional rules on the experience of being homeless. The self-reports also suggest that less support was available to help single people find stable housing than for adults in families, even though respondents in all of the subgroups highlighted economic problems as barriers to finding and keeping regular housing. These results add Ca- nadian evidence as well as generalizability to the conclusions of several American longitudinal studies concerning the critical role of subsidized housing in helping people who were homeless achieve housing stability (Shinn et al., 1998; Stojanovic et al., 1999; Wong & Pilavin, 1997; Zlotnick et al., 1999). The need for further analysis on the significance of health and social services such as supportive housing, case management, drug rehabilita- tion, mediation, and counselling, and the governance rules that deter- mine their availability, is also raised by these findings. It is evident, for example, that in Ottawa, access to subsidized housing is much easier if you are homeless with a child than if you are alone, or if you are a single woman with a mental illness or someone who is fleeing domestic abuse. Much of Ottawa?s supportive and supported housing is specifically for individuals with severe and persistent mental illness and thus provides a route for accessing subsidized housing. In addition, priority status for social housing is available to individuals, usually women, escaping spousal abuse. These matters draw attention to the work of anthropolo- gists (Knowles, 2000; Passaro, 1996), and geographers (DeVerteuil, 2003; Wekerle, 1997; Wolch & Dear, 1993) who have explored the extent to which governance rules shape the differential experiences and ultimate duration of being homeless for various groups of people. The role of institutions and institutional rules about support ser- vices, especially those dealing with financial matters, interpersonal con- flict, and substance abuse, are significant, according to our participants. What is less clear though, is the extent to which they affect individuals? ability to secure and maintain stable housing. For example, respondents? remarks about the positive experience of being on ODSP are likely ex- plained by the higher funds that this category of social support yields. Yet, it also highlights the fact that standard social assistance payments are too low to allow people to find or stay in regular housing in the pri- vate market (Alliance to End Homelessness in Ottawa, 2005, 2006, 2007, 2008). It raises the question: what impact would a increase in social assis- tance rates have on the extent to which formerly homeless individuals are able to maintain stable housing, in spite of health vulnerabilities or interpersonal difficulties that they might have or encounter? Similarly, respondents? remarks about problems that accrue from the availability of illicit drugs and drug users, and about interpersonal conflicts more generally, raise questions about the value that increased access to psychosocial treatments such as mental health case manage- ment and drug rehabilitation might yield in conjunction with access to stable, affordable housing. These questions highlight the value of encouraging further Cana- dian longitudinal research on homelessness that would include several cities and allow a comparative investigation of issues raised in this analysis. Given the diversity of provincial and municipal rules, regula- tions, and procedures with regard to the delivery of specialized health and social services, including social housing, case management, and drug rehabilitation, this (admittedly ambitious) venture has the potential of contributing significant insights about the relations between govern- ance rules and homelessness. Fran Klodawsky is an Associate Professor at Carleton University in the De- partment of Geography and Environmental Studies. Tim A ubry is a Full Pro- fessor in the School of Psychology and Senior Researcher at the Centre for Re- search on Educational and Community Services at the University of Ottawa. Rebecca Nemiroff is a Ph.D. student at the School of Psychology, University of Ottawa. Cristina Bonetta is a researcher at Saint Paul University, Ottawa. Alette Willis is a post-doctoral student in the School of Health in Social Sci- ence at the University of Edinburgh. References Alliance to End Homelessness in Ottawa (2005). Experiencing homelessness: The first report card on homelessness in Ottawa, 200 5. Ottawa: Author. Alliance to End Homelessness in Ottawa (2006). Experiencing homelessness: Second report card on homelessness in Ottawa, January-December 200 6 . Ottawa: Au- thor. Alliance to End Homelessness in Ottawa (2007). Experiencing homelessness: Third report card on ending homelessness in Ottawa, January-December 200 7 . Ottawa: Author. Alliance to End Homelessness in Ottawa (2007). Experiencing homelessness: Third report card on ending homelessness in Ottawa, January-December 2008 . Ottawa: Author. Anucha, U. (2003). A study of tenant exits from housing for homeless people: Where do they come from? Why do they leave? Where do they go?. Ottawa: Canada Mort- gage and Housing Corporation. Aubry, T., Klodawsky, F., Nemiroff, R., Birnie, S., & Bonetta, C. (2007). Panel study on persons who are homeless in Ottawa: Phase 2 r esults. Ottawa: Centre for Research on Community Services. Aubry, T., Klodawsky, F., Hay, E., & Birnie, S. (2003). Panel study on persons who are homeless in Ottawa: Phase 1 results. Ottawa: Centre for Research on Community Services. Aubry, T., Klodawsky, F., Hay, E., Nemiroff, R., & Hyman, S. (2004). Developing a methodology for tracking persons who are homeless over Time: Final report. Ot- tawa: Canada Mortgage and Housing Corporation. Bebout, R., Drake, R. E., Xie, H., McHugo, G., & Harris, M. (1997). Housing status among formerly homeless dually diagnosed adults. Psychiatric Services, 48(7), 936-941. Browne, A., & Bassuk, S. (1997). Intimate violence in the lives of homeless and poor housed women: Prevalence and patterns in an ethnically diverse sam- ple. American Journal of Orthopsychiatry, 67(2), 261-278. Caton, C. L. M., Domingues, B., Schanzer, B., Hasin, D. S., Shrout, P. E., Felix, A., McQuistion, H., Opler, L. A., & Hau, E. (2005). Risk factors for long-term homelessness: Findings from a longitudinal study of first-time homeless single adults. American Journal of Public Health, 95(10), 1753-1759. Clatts, M. C., Davis, W. R., Sotheran, J. L., & Atillsoy, A. (1998). Correlates and distribution of HIV risk behaviors among homeless youths in New York City: Implications for prevention and policy. Child Welfare, 77(2), 195-207. Cohen, C. I., Ramirez, M., Teresi, J., Gallagher, M., & Sokolovsky, J. (1997). Pre- dictors of becoming redomiciled among older homeless women. The Geron- tologist, 37(1), 67-74. Concover, S., Berkman, A., Gheith, A., Jahiel, R., Stanley, D., Geller, P. A., Valen- cia, E., & Susser, E. (1997). Methods for successful follow-up of elusive ur- ban populations: An ethnographic approach with homeless men. Bulletin of the New York Academy of Medicine, 74(1), 90-108. DeVerteuil, G. (2003) Homeless mobility, institutional settings, and the new pov- erty management. Environment and Planning A, 35(2), 361-379. Farrell, S. J., Aubry, T., Klodawsky, F., Jewett, P., & Pettey, D. (2000). Describing the homeless population of Ottawa-Carleton: Fact sheets of selected findings. http://www.uottawa.ca/academic/socsci/crcs/homeless. Goering, P., Wasylenki, D., Lindsay, S., Lemire, D., & Rhodes, A. (1997). Process and outcome in a hostel outreach program for homeless clients with severe mental illness. American Journal of Orthopsychiatry, 67(4), 607-617. Humphreys, K., & Rosenheck, R. (1998). Treatment involvement and outcomes for four subtypes of homeless veterans. American Journal of Orthopsychiatry, 68(2), 285-294. Klodawsky, F., Aubry, T., Behnia, B., Nicholson, C., & Young, M. (2005, Febru- ary). The panel study on homelessness: Secondary data analysis of respondents whose country of origin is not Canada. Ottawa, ON: National Homelessness Secretariat. Klodawsky, F., Aubry, T., & Nemiroff, R. (2006, March 21-23). Comparing aborigi- nal and non-aboriginal Canadian born respondents to the panel study on home- lessness. Paper presented at the Aboriginal Policy Research Conference, Ot- tawa. Knowles, C. (2000) Burger King, Dunkin Donuts and community mental health care. Health and Place, 6(3), 213-224. Kreider, B., & Nicholson, S. (1997). Health insurance and the homeless. Health Economics, 6(1), 31-41. Lam, J. A., Jekel, J. F., Thompson, K. S., & Leaf, P. J. (1995). Assessing the value of a short-term residential drug treatment program for homeless men. Journal of Addictive Diseases, 14(4), 21-39. Morse, G. A., Caslyn, R. J., Allen, G., Tempelhoff, B., & Smith, R. (1992). Experi- mental comparison of the effects of three treatment programs for homeless mentally ill people. Hospital and Community Psychiatry, 43(10), 1005-1010. Passaro, J. (1996). The unequal homeless: Men on the streets, women in their place. New York: Routledge. Piliavin, I., Wright, B. R. E., Mare, R. D., & Westerfield, A.H. (1996). Exits from and returns to homelessness. Social Service Review, 70(1), 33-57. Roy, ? ., Haley, N., Leclerc, P., Lemire, N., Boivin, J.-P., Frappier, J.-Y., & Claes- sens, C. (2000). Prevalence of HIV infection and risk behaviours among Montreal street youth. International Journal of STD and AIDS, 11(4), 241-247. Roy, ? ., Haley, N., Leclerc, P., Boivin, J.-F., C?dras, L., & Vincelette, J. (2001). Risk factors for hepatitis C virus infection among street youths. Canadian Medical Association Journal, 165(50), 557-560. Roy, ? ., Haley, N., Leclerc, P., Cedras, L., Blais, L., & Boivin. J.-F. (2003). Drug injection among street youths in Montreal: Predictors of initiation. Journal of Urban Health, 80(1), 92-105. Roy, ? ., Haley, N., Leclerc, P., Sochanski, B., Boudreau, J.-F., & Boivin, J.-F. (2004). Mortality in a cohort of street youth in Montreal. Journal of the Ameri- can Medical Association, 292(5), 569-574. Shinn, M., Weitzman, B. C., Stojanovic, D., Knickman, J. R., Jim? nez, L., Duchon, L., James, S., & Kranz, D. (1998). Predictors of homelessness among families in New York City: From shelter request to housing stability. American Jour- nal of Public Health, 88(11), 1651-1657. Sobo, E., Zimet, J., Zimmerman, T., & Cecil, H. (1997). Doubting the experts: AIDS misconceptions among runaway adolescents. Human Organization, 56(3), 311-320. Sosin, M., Piliavin, I., & Westerfelt, H. (1990). Toward a longitudinal analysis of homelessness. Journal of Social Issues, 46(4), 157-174. Stojanovic, D., Weitzman, B. C., Shinn, M., Labay, L., & Williams, N. (1999). Trac- ing the path out of homelessness: The housing patterns of families after ex- iting shelter. Journal of Community Psychology, 27(2), 199-208. Stretch, J. J., & Krueger, L. W. (1992). A social-epidemiological five year cohort study of homeless families: A public/private joint venture policy analysis utilizing applied computer technology. Computers in Human Services, 9(3-4), 209-230. Tessler, R., Rosenheck, R., & Gamache, G. (2001). Gender differences in self- reported reasons for homelessness. Journal of Social Distress and the Homeless, 10(3), 243-254. Toro, P. A., Rabideau, J. M., Bellavia, C. W., Daeschler, C. V., Wall, D. D., & Thomas, D. M. (1997). Evaluating an intervention for homelesss persons: Results of a field experiment. Journal of Consulting and Clinical Psychology, 65(3), 476-484. Wekerle, G. (1997). The shift to the market: Gender and housing disadvantage. In P. M. Evans & G. Wekerle (Eds.) Women and the Canadian Welfare State (pp. 170-196). Toronto: University of Toronto Press. Wolch, J., & Dear, M. (1993). Malign neglect: Homelessness in an American city. San Francisco: Jossey-Bass. Wolch, J., & Philo, C. (2000). From distributions of deviance to definitions of dif- ference: Past and future mental health geographies. Health and Place, 6(3), 137-157. Wong, Y. I. (1999). Utilization of community-based services among homeless persons: An application of a theoretical model. Journal of Community Psy- chology, 27(3), 327-345. Wong, Y. I., & Piliavin, I. (1997). A dynamic analysis of homeless-domicile transi- tions. Social Problems, 44(3), 408-423. Wong, Y. I., Piliavin, I., & Wright, B. R. E. (1998). Residential transitions among homeless families and homeless single individuals: A comparison study. Journal of Social Service Research, 24(1-2), 1-27. Wright, J. T., Allen, T. L., & Devine, J. A. (1995). Tracking non-traditional popula- tions in longitudinal studies. Evaluation and Program Planning, 18(3), 267- 277. Zima, B. T., Bussing, R., Bystritsky, M., Widawski, M. H., Berlin, T. R., & Benja- min, B. (1999). Psychosocial stressors among sheltered homeless children: Relationship to behavior problems and depressive symptoms. American Journal of Orthopsychiatry, 69(1), 127-133. Zlotnick, C., Robertson, M. J., & Lahiff, M. (1999). Getting off the streets: Eco- nomic resources and residential exits from homelessness. Journal of Commu- nity Psychology, 27(2), 209-224. Chapter 8.2 Pathways into Homelessness: Testing the Heterogeneity Hypothesis TRACY PERESSINI One of the most common findings to emerge from research on the home- less over the last two decades is that they are not homogeneous (Rosen- heck et al., 1999). The homeless are a remarkably diverse group of people who come from all walks of life. They include men and women, single persons and families, young and old people, those with and without mental and physical health problems, rural and urban dwellers, the rich and poor, people with high and low educational and occupational sta- tuses, people from all racial, ethnic, and visible minorities, illegal immi- grants, former criminal offenders, runaway youth, prostitutes, and people with drug and alcohol addictions (Burt, 1993; Chamberlain & MacKenzie, 2006; Rossi, 1989; Varney & Vliet, 2008; Zald, 2004). People in each of these sub-groups go through different hardships and stressful life events on their way to becoming homeless (Burt, 1993; Munoz et al., 2005; Rossi, 1989), and these experiences in turn have been identified by researchers as causes of homelessness and then used to ex- plain why some people may be at greater risk for homelessness than oth- ers (Blau, 1992, p. 17; Burt et al., 2001, p. 55; Hirschl, 1990, p. 448; Wright et al., 1998, pp. 5-6). Yet, at the same time, researchers note that all home- less persons share three societal factors: extreme poverty, an inability to find affordable housing, and interpersonal conflict or violence (Burt et al., 2001; Ji, 2006; Rosenheck et al., 1999; Tessler et al., 2001). Other com- mon risk factors are related to social structural causes, such as chronic unemployment, reductions in welfare support, deinstitutionalization, declining personal incomes, economic restructuring, gentrification, in addition to the individual pathologies of mental illness, problem drink- ing, substance abuse, criminal behaviour and deviance, and deficiencies in human capital (Burt et al., 2001; Clapham, 2003; De Venanzi, 2008; Ji, 2006; Main, 1998; Ropers, 1991; Rossi, 1989). Many researchers have equated the heterogeneity of demographic characteristics in the homeless population with different risk factors and argued that each sub-group of the population therefore has unique social service needs. For example, Tessler et al. (2001) argue that because wom- en report different reasons for homelessness, a different approach to building therapeutic relationships with them is required. Similarly, other studies have demonstrated that shelter options, programs, and services need to be developed to meet the distinctive needs of older adults and younger homeless adults (Crane, 1996; Garibaldi et al., 2005; Keigher & Greenblatt, 1992; Shinn et al., 2007). The same arguments have been made for homeless families and children; single persons; persons with mental or physical health prob- lems; persons of different ethnic, racial and sexual orientations; individ- uals with addictions (alcohol, drugs, gambling); people of different social status (criminal, refugee, immigrant) or levels of education; and those who are newly homeless versus those who are chronically homeless (Roll et al., 1999; Toro, 2007). Much of the current research literature con- sistently argues that attending to the distinctive risk factors associated with and reasons cited by the different sub-groups of the homeless is critical in developing effective interventions and service delivery models, and in program planning (Rosenheck et al., 1999). In other words, the heterogeneity hypothesis suggests that, in order to effectively address the problem of homelessness in North America, we must develop separate and distinctive services, programs, and policies for each of the sociodemographic sub-groups of the population. This study examines the self-perceived reasons for homelessness reported by the population of homeless persons in a midsize Canadian city in the context of the pathways into homeless that have been hy- pothesized in the research literature (e.g. poverty, interpersonal conflict, health, housing loss and affordability, addictions, deinstituuonalization and social safety net failure). I then test the heterogeneity hypothesis by examining the association between sociodemographic risk factors and the self-reported reasons for homelessness reported by the sample. In summary, this study addresses the following research questions: 1. Is the likelihood of becoming homeless a function of the sociodemo- graphic traits of the population? 2. Which group of Canadian homeless persons is at greater risk for homelessness caused by: (a) poverty; (b) interpersonal conflict; (c) health problems; (d) housing loss or housing unaffordability; (e) ad- dictions; (f) deinstitutionalization; (g) lack of public support (i.e., so- cial safety net failure)? 3. To what degree are the reasons or risk factors for homelessness cor- related with the individual traits of the population? 4. What are the service, program, and policy implications of the soci- odemographic variations across risk factors? Surveying the homeless in Peel Region The data for this study were collected from a longitudinal survey of the population of homeless persons using food and drop-in services and shelters in the Region of Peel (composed of the municipalities of Missis- sauga, Brampton, and Caledon), Ontario, between April 2000 and March 2001. Interviews were conducted by trained interviewers every three months at eight sites: two drop-in centres (one adult, one youth); one food program; four shelters (two male, one female, and one mixed), and on the street. In total 268 unique homeless persons were interviewed, with an overall response rate of 93.2 percent. Voluntary informed consent was obtained in writing from all study participants each time. Study participants answered a 10-page question- naire designed to collect detailed information on their sociodemographic background (e.g., age, sex, educational attainments, etc.), work history (occupation, skills, training, current and previous employment, reasons for unemployment), levels of income (previous and current sources and amounts of income), health problems (both mental and physical, using standardized instruments), family background (history of family vi- olence, disruption, or breakdown), social networks (previous and current associations, day-to-day activities), and rates of service utilization (types of services used, the frequency of utilization, the duration and periods of utilization, types of services needed but not available), as well as the in- dividual?s history and background of homelessness. Participants were selected for inclusion in the study if they did not have their own place to stay (e.g., room, apartment or house), or if they did have their own place, but they had not stayed there for a period of 30 days or more. This definition included anyone who slept in an outdoor encampment, in a shelter, on the street, in an abandoned building, in a vehicle, and in a welfare motel. Participants were asked to state the number of times they had been without a home of their own or a place to stay and how long they had been in this situation. Factor analysis The dependent variables were created by factor analyzing the following reasons for homelessness identified by the study participants: ? no money or income; ? no job (unemployed); ? lost entitlement or became ineligible for social assistance; ? didn't qualify for so cial assistance; ? lost entitlement or became ineligible for unemployment insurance benefits; ? didn't qualify for unemployment insur ances benefits; ? lost entitlement or became ineligible for worker?s compensation ben- efits; ? couldn't get worker? s compensation benefits (even though injured on the job); ? experienced a family crisis (e.g. relationship breakdown, divorce, family conflict); ? family asked individual to leave or threw individual out; ? mental health problems; ? physical health problems; ? couldn?t find work; ? couldn?t find an apartment or place that is affordable; ? evicted or asked to leave last place of residence; ? recently released from prison or jail; ? recently released from a mental health facility or hospital; ? recently released from foster care (only youth asked); ? ran away from home (only youth asked); ? couldn?t find a room-mate; ? trouble or problems with a room-mate. Factor analysis is a data reduction technique used to identify the underlying structure of highly correlated survey items, as well as to identify what the factors represent conceptually (Norusis, 2008). The above list of reasons were factor analyzed using principal axis factoring extraction with a varimax rotation in order to simplify the number of factors extracted (Norusis, 2008). The outcome of this analysis was the extraction of seven primary factors representing the following aggregate reasons for or pathways into homelessness identified by the study participants: 1. poverty; 2. interpersonal conflict/'violence; 5. health (mental and physical); 6. housing loss or lack of affordability; 7. addictions; 8. deinstitutionalization; 9. social safety net failure (lack or public/social support). These factors explain 59.2 percent of the variance across the 22 reasons. The internal consistency of each of the seven factors was then eva- luated by calculating the ? coefficients for reliability for each factor. With the exception of addictions, which was composed of only one item, the ? coefficients for reliability ranged from 0.7 to 0.9 for the poverty, interper- sonal conflict/violence, health, housing loss/affordability, deinstitutiona- lization and social safety net failure. The dependent variables, representing each of the seven factors identified above, were created by assigning a value of 0 to each dependent variable if the participants did not specify any of the reasons included in each factor and 1 if they speci- fied at least 1 of the reasons included in each factor. The risk variables (covariates) were chosen based on key sociode- mographic characteristics and traits identified in the research literature: ? age; ? gender; ? marital status; ? ethnicity; ? education; ? number of times homeless; ? amount of time homeless.1 In order to assess the linkage between the reasons for homelessness and the sociodemographic traits of the homeless, a series of logistic re- gression models were estimated for each of the dependent variables re- gressed on the risk variables (covariates). For each of these analysis, di- agnostics were run to ensure that the fit of each model was supported over the set of covariates and the assumptions of logistic regression were satisfied (Hosmer & Lemeshow, 2000, pp. 167-186). Self-reported reasons for homelessness Two-thirds of the homeless people in this study were men; 50 percent were between the ages of 25-44, with 32 percent less than 24 years of age and 19 percent 45 and older. Close to 60 percent of the sample is com- posed of single persons, 72 percent born in Canada, and 46 percent had 12 years of education or more. About 50 percent of the homeless in this study report that their current spell of homeless is their first time and that they have been homeless longer than one month. Finally, the majori- ty of study participants identified poverty (75 percent), interpersonal conflict (63 percent) and housing loss or lack of affordability (59 percent) as the three primary pathways that took them into homelessness. Over- all, this sample is similar to other previously studied samples of home- less men and women in midsized Canadian cities. See Table 1. 1 Each risk factor was dummy coded as follows: Age: 1 if 25-44, 0 otherwise; Gender: 1 if male, 0 otherwise; Marital Status: 1 if single, 0 otherwise; Ethnic- ity: 1 if Canadian, 0 otherwise; Education: 1 if respondent reported 12 or more years of schooling, 0 otherwise; Number of times homeless: 1 if res- pondent had been homeless more than once, 0 otherwise; Amount of time homeless: 1 if respondent had been homeless 2 or more months, 0 otherwise. Table 1 Table 2 Table 2 provides a breakdown of the seven aggregated reasons for homelessness by age, gender, marital status, ethnicity, education and homelessness background. The data do, indeed, at least partially support the hypothesis that different sociodemographic groups tend to be more likely to follow distinctive pathways into homelessness. For example, women are significantly more likely than men to cite poverty (82.2 per- cent vs. 71.3 percent), interpersonal conflict {77.8 percent vs. 56.2 percent) and housing affordability (66.7 percent vs. 54.5 percent) as their reason for homelessness, while men are significantly more likely than women to report addictions (23.6 percent vs. 14.4 percent) as a reason for being cur- rently homeless. Similarly, significant differences are observed across the other risk variables for some, but, not all pathways into homelessness. For exam- ple, older homeless adults are significantly more likely than those 24 to 44 to report interpersonal conflict or violence, health problems, and ad- dictions as primary reasons for their homelessness, while they are signif- icantly less likely to report interpersonal conflict as a reason than those who are 24 years of age and under. Those with less than 12 years of schooling are significantly more likely to report poverty and interper- sonal conflict as reasons, and are significantly less likely to report hous- ing affordability problems or housing loss as the reason for their home- lessness. Single persons are significantly less likely than their married counterparts to report addictions and social safety net failure as their reason for currently being homeless. Ethnicity (or ethnic background) appears to be associated with interpersonal conflict, with a significantly higher percentage of immigrants reporting the interpersonal conflict as their reason for being homeless. It is noteworthy that none of the risk variables are associated with deinstitutionalization in Canada. Finally, the frequency and duration of homelessness do appear to be significantly associated with several reasons for homelessness. Specifical- ly, the more frequent a person has been homeless the more likely they are to report the following reasons: interpersonal conflict, health prob- lems and social safety net failure, with the longer one has been homeless being significantly associated with health problems and social safety net failure. In sum, these data lend support to the heterogeneity hypothesis, at least in so far as they indicate that different sociodemographic groups tend to be more likely to follow unique paths to becoming homeless. Based on the findings to this point, it does appear that the likelihood of becoming homeless is a function of the sociodemographic characteristics of the population. Pathways to homelessness and their sociodemographic risks A series of logistic regressions were conducted to assess the net effect of each of the risk variables/factors (controlling for the others) on the seven pathways into homelessness identified by the study participants (see Table 3). Looking at the overall fit for each of the seven pathways models, none provide an adequate or good fit to the data. While the model chi- square for poverty, interpersonal conflict, health, addictions and social safety net failure do indicate a significant improvement in fit over the model containing only the constant, the goodness of fit statistics suggest otherwise. For each of the pathways models, the model -2 log likelih- oods, goodness of fit statistics, and the pseudo (Negelkerke) R2 Age, gender, marital status, ethnicity, education, and frequency and duration of homelessness explain: indicate that, taken altogether, the sociodemographic risk factors do not explain a significant amount of the variance in each of the pathways into home- lessness. ? 9.1 percent of the variance in poverty; ? 16.9 percent of the variance in interpersonal conflict/violence; ? 20.3 percent of the variance in health; ? 5.8 percent of the variance in housing loss/affordability; ? 10.2 percent of the variance in addictions; ? 6 percent of the variance in deinstitutionalization; ? 9.9 percent of the variance in social safety net failure. Table 3 Despite the poor fit, each of the pathways models do provide in- sights into the sociodemographic risk factors that are most relevant in terms of predicting the likelihood of identifying a specific pathway into homelessness for each sub-group. With the exception of deinstitutionali- zation, distinct sociodemographic characteristics significantly predict the likelihood of identifying a specific pathway into homelessness. For example, there is a 28 percent reduction in the odds of reporting poverty as a reason for their homelessness for older homeless adults, as compare to those aged 25-44. Similarly, there is a 50 percent reduction in the odds of reporting poverty for those who are single as compared to all other marital statuses. There is a significant reduction in the odds of re- porting interpersonal conflict for those who are under 24 (36 percent), single (42 percent), immigrants (47 percent), and those who have less than 12 years of schooling (54 percent). Similarly, men are 50 percent less likely to report health as a reason for their homelessness. On the other hand, older homeless adults (7.3 times more likely) and those who have been homeless more than once (3.7 times more likely) are significantly more likely to report health as a reason. Like interpersonal conflict, the odds of reporting housing loss/affordability as a reason for homelessness are significantly affected by gender and education. There is a 50 percent reduction in the odds of reporting housing as a reason for men, while those with 12 years of schooling or more are nearly twice as likely to identify housing as the reason for their homelessness. Unlike those who reported poverty and interpersonal conflict as a reason for their homelessness, both younger and older adults are at least three times more likely to report addictions as a reason than those 25-44. The only risk factor that is related to reporting social safety net failure or a loss of public supports is the number of times homeless. Those who have been homeless more than once are more than twice as likely to re- port this reason for homelessness as compared to the newly homeless. Finally, none of the risk variables significantly predict the likelihood of specifying deinstitutionalization as a reason for homelessness. Making sense of the findings The findings of the present study provide only limited support for the connection between sociodemographic traits and pathways into home- less. Specifically, the sociodemographic groups specified below are at greater risk for homeless due to: 1. Poverty: those aged 25-44 and those who are married, separated, divorced or widowed (Age and Marital Status). 2. Interpersonal Conflict: those aged 25-44, women, immigrants, and those with less than 12 years of education (Age, Gender, Ethnicity, and Education). 3. Health: those aged 45 and older, women, and people who have been homeless more than once (Age, Gender, and Frequency of Home- lessness). 4. Housing Loss or Housing unaffordability: women and those with more than 12 years of education (Gender and Education). 5. Addictions: those 24 and under or 45 and older (Age). 6. Deinstitutionalization: None. No sociodemographic trait predicts the likelihood of reporting this reason. 7. Social Safety Net Failure: those who have been homeless more than once (Frequency of Homelessness). Consistent with other research (Roll et al., 1999; Tessler et al., 2001) the findings from the logistic regressions differ somewhat from the biva- riate analysis of the association between demographic traits and aggre- gate reasons for homelessness. When one examines the simple relation- ship between aggregate reasons and sociodemographic traits, it does appear that the pathways into homelessness vary considerably depend- ing on the sub-group the homeless fall into. But when each of the soci- odemographic covariates are examined controlling for the others, a very different pattern emerges ? one in which the likelihood of following a specific pathway into homelessness is consistently affected only by gender and/or age. The odds of falling into homelessness due to poverty, interpersonal conflict, health, and addictions vary considerably by age, while gender affects the likelihood of reporting interpersonal conflict, health problems, and housing issues as reasons for homelessness. The other sociodemo- graphic variables in the analysis are not as robust as gender and age in their association with the reasons for homelessness. Unlike gender and age, each of the other variables in the analysis is associated with a specif- ic pathway into homelessness. For example, the only pathway associated with ethnicity is interpersonal conflict or violence. Similarly, marital sta- tus is associated only with poverty. Education is slightly more robust in that it is associated with both interpersonal conflict and housing. Overall, these findings are consistent with earlier research that indi- cates that while the homeless population is heterogeneous, there is greater homogeneity in the reasons or pathways into homelessness that people report (Burt et al., 2001; Ji, 2006; Rosenheck et al., 1999; Tessler et al., 2001). These data indicate that poverty, housing issues, and interper- sonal conflict are the three main pathways into homelessness for home- less Canadians and that the odds of following one of these pathways are significantly affected by age and gender. It is, however, important to recognize that these data also indicate that none of the pathway models are significant and that the sociodemo- graphic variables included in the analysis explain only an insignificant proportion of the variance in the self-report reasons for homelessness. This would suggest that there are unidentified confounding variables that play a more important role in explaining the pathways that people take into homelessness. Furthermore, additional research needs to ex- amine a broader range of covariates in order to explain the varying rea- sons that the homeless report. These findings do suggest that the reasons for homelessness re- ported in this study depend on the specific group reporting them. Men, women, youth, adults, and seniors have differing etiologies for their homelessness, which suggests that preventive interventions probably should be tailored to meet their specific needs, particularly in terms of poverty, housing, and interpersonal conflict and violence. For example, the data suggest that poverty interventions need to address the unique needs of those 25 to 44 years of age and those who are married, sepa- rated, divorced or widowed. Similarly, separate interventions for youth, women, immigrants and those with less than 12 years of education are needed to address the issue of interpersonal conflict and violence. The data, however, are limited in terms of addressing program and service development based on the needs of different demographic groups as they pertain to health issues, addictions problems and lack of public supports and services. Given the poor fit of the pathways models, these data suggest that many of the pathways into homelessness are rooted in individual problems and issues, not the demographic traits of the population. The findings also indicate that homeless persons expe- rience significant personal problems that cause them to fall through the gaps in or be excluded from Canada's system of public assistance, i n- come supports, and health care system. The results of this study provide only limited support for the hete- rogeneity hypothesis. Some sociodemographic groups are at risk for fol- lowing specific pathways into homelessness ? poverty, housing prob- lems, and interpersonal conflict or violence ? and require specific interventions to address their unique needs (e.g., men, women, youth, and seniors). The findings of this research, however, also suggest that the more critical issue in the development program and service responses is that of unmet need. This study indicates that some of the key pathways into homelessness cannot be explained by sociodemographic factors alone. Pathways such as addiction problems, limited education or access to education, mental and physical health problems, and exclusion from Canada's social safety net are key issues that need r edress in terms of program and service development, especially for youth, seniors, women and those who have been homeless more than once. Tracy Peressini is an Associate Professor of Sociology at Renison University College, University of Waterloo. Based on an article published in the Canadian Journal of Urban Research. Winnipeg: Summer 2007. Vol. 16, No . 1; pg. 112 - 126. References Blau, J. (1992). The visible poor: Homelessness in the United States. New York: Oxford University Press. Burt, M. (1993). Over the edge: The growth of homelessness in the 1980s. New York: Russell Sage Foundation. Burt, M., Aron, L. Y., Lee, E., & Valente, J. (2001). Helping America's homeless: Emergency shelter or affordable housing. Washington: Urban Institute Press. Chamberlain, C., & MacKenzie, D. (2006). Homeless careers: A framework for intervention. Australian Social Work, 59(2), 198-212. Clapham, D. (2003). Pathways approaches to homelessness research. Journal of Community and Applied Social Psychology, 13(2), 119-127. Crane, M. (1996). The situation of older homeless people. Reviews in Clinical Ger- ontology, 6(4), 389-398. DeVenanzi, A. (2008). The institutional dynamics of homelessness: The United States of America and Japan compared. International Journal of Sociology 28(3-4), 129-145. Garibaldi, B., Conde-Martel, A., & O'Toole , T. P. (2005). Self-reported comorbidi- ties, perceived needs, and sources for usual care for older and younger homeless adults. Journal of General Internal Medicine, 20(8), 726-730. Hirschl, T. (1990). Homelessness: A sociological research agenda. Sociological Spectrum, 10(4), 443-467. Hosmer, D. W., & Lemeshow, S. (2000). Applied Logistic Regression (2nd ed.). New York: John Wiley and Sons, Inc. Ji, E. (2006). A Study of the structural risk factors of homelessness in 52 metropol- itan areas in the United States. International Social Work, 49(1), 107-117. Keigher, S. M., & Greenblatt, S. (1992). Housing emergencies and the etiology of homelessness among the urban elderly. The Gerontologist, 32(4), 457-465. Main, T. (1998). How to think about homelessness: Balancing structural and in- dividual causes. Journal of Social Distress and the Homeless, 7(1), 41-54. Munoz, M., Panadero, S., Santos, E. P., & Quiroga, M. A. (2005). Role of stressful life events in homelessness: An intragroup analysis. American Journal of Community Psychology, 35(1-2), 35-47. Norusis, M. J. (2008). SPSS 16.0 statistical procedures companion. Toronto: Prentice Hall. Roll, C. N., Toro, P. A., & Ortola, G. L. (1999). Characteristics and experiences of homeless adults: A comparison of single men, single women, and women with children. Journal of Community Psychology, 27(2), 189-198. Ropers, R. H. (1991). Persistent poverty: The American dream turned nightmare. New York: Plenum Press. Rosenheck, R., Bassuk, E., & Salomon, A. (1999). Special populations of homeless Americans. In L. B. Fosburg, & D. L. Dennis (Eds.), Practical lessons: The 1998 National Symposium on Homelessness Research. Washington: U.S. Depart- ments of Housing and Urban Development and Health and Human Servic- es. Rossi, P. H. (1989). Down and out in America: The origins of homelessness. Chica- go: University of Chicago. Shinn, M., Gottlieb, J., Wett, J. X. , Bahl, A., Cohen, A., & Ellis, D. B. (2007). Predic- tors of homelessness among older adults in New York City. Journal of Health Psychology, 12(5), 696-708. Tessler, R., Rosenheck, R., & Gamache, G. (2001). Gender differences in self- reported reasons for homelessness. Journal of Social Distress and the Homeless, 10(3), 243-254. Toro, P. (2007). Toward an international understanding of homelessness. Journal of Social Issues, 63(3), 461-481. Varney, D., & van Vliet, W. (2008). Homelessness, children and youth: Research in the United States and Canada. American Behavioral Scientist, 51(6), 715- 720. Wright, J. D., Rubin, B. A., & Devine, J. A. (1998). Beside the golden door: Policy, politics and the homeless. New York: Aldine de Gruyter. Chapter 8.3 Towards a Strategy for Counting the Homeless TRACY PERESSINI, LYNN MCDONALD , AND J. DAVID HULCHANSKI Conventional strategies for surveying and counting the general popula- tion (e.g., traditional probability techniques using households and census blocks) do not apply when enumerating the homeless. Although re- searchers have attempted to adapt conventional survey methods in esti- mating the size of the homeless population, their experience has demon- strated the limited utility, reliability, and accuracy, as well as the exorbitant costs involved in adapting them for use with the homeless population; particularly when attempting to sample from the street- dwelling population (see, for example, Dennis, 1993; Rossi, 1989). As a result, researchers in the United States have struggled to develop new techniques for sampling and counting the homeless population. After two decades of developing, testing, and modifying their methods, re- searchers have reached a consensus that service-based methods1 1 The discussion in this section has been derived from the following sources: Bentley, 1995; Dennis, 1993 and 1991; Dennis and Iachan, 1993 and 1992; Burt, 1993 and 1991; Taueber, 1991; Burnam and Koegel, 1988. produce the most accurate and reliable results. Service-Based Methods: Description & Overview In general, service-based methods refer to a class of survey techniques that sample from or count homeless people in a variety of service system locations, including shelters, soup kitchens, day programs (e.g. drop-in centres), congregate areas or outdoor encampments,2 Typically, two factors affect the choice of sampling frames: the defi- nition of homelessness and the cost of sampling from a variety of loca- tions. In general, definitions can be categorized as falling into one of two types: relative and absolute. Relative definitions tend to be very broad and inclusive and, therefore, use the most expansive range of sampling frames to ensure a representative sample of all the constituent groups is selected. Absolute definitions tend to focus on only those who are literally homeless ? rough sleepers and/or those who physically live on the street. street and mobile health care programs, street outreach programs, casual labour offices, etc. Some studies based on these techniques also sample from street loca- tions, which may include those individuals who sleep on the street, on river banks, under bridges, on rooftops, in vacant buildings, in a public or commercial facility (e.g., library, city hall, shopping mall), in a city park, in a car, or in any other place not meant for human habitation. While there are no hard-and-fast rules about the number of sampling frames that should be included in a study, it is clear that sampling from the broadest range of locations provides the highest degree of coverage. In this context, then, the greater the number of sampling frames included in a study, the greater the coverage of the population, and, hence, the greater the reliability and accuracy of the count. That said researchers have found that sampling from shelters, drop-in centers and soup kitch- ens provides approximately 90-95 percentage of coverage of the urban homeless population, and yield more reliable population estimates than the Canadian Census. Researchers have consistently demonstrated that the majority of other homeless service users can be captured in one of the sample frames noted above over a 30 day period. 2 Congregate areas consist of outdoor sites where the homeless are known to gather on a regular basis during the day. Outdoor encampments, on the other hand, are parks, campgrounds, and vacant lots where the homeless not only congregate, but live as well (e.g., tent cities or squatter encampments). Relative definitions require the researcher to sample from all possible locations where the homeless may be found. Absolute definitions focus on the street dwelling population via a process of preliminary counts and interviews with the homeless and those who serve them (i.e. key informants) in order to create a map the geographical areas where the street homeless may be located. The preliminary interviews also allow the researcher to calculate probabilities of locating the homeless in vari- ous locations and, therefore, to maximize the likelihood of capturing the street dwelling population in a either a snap shot survey (one night) or contiguous surveys over an extended period of time. For example, if the goal of a study is to estimate the size of the pop- ulation of homeless women who are victims of family violence, it makes sense to sample locations that maximize the odds of encountering them. Therefore, an investigator may exclude shelters for runaway youth, de- tox centres that service alcohol and drug addicts, flop houses and men?s shelters where it is very unlikely that they will find battered women and their families. The problem with approaching a study of the homeless in this way is that there will always be some margin of error involved (that cannot be predetermined or estimated) in excluding locations or subsets of the population from a study. This is not as problematic as it sounds as studies of any population, using any type of random or probability sam- pling technique will necessarily result in some margin of error. The goal then would be to collect enough information at the outset to minimize sampling error while maximizing coverage of the population. Apart from shelters for the homeless, most of the services that the homeless use are not specifically targeted to meet just their needs. A wide variety of individuals use food services, social and community drop-in centers, health care services, and employment services and the homeless constitute only a portion of all the people serviced by these types of organizations. An investigator, therefore, may exclude organiza- tions that service a small proportion of homeless people. In doing so, they automatically exclude that proportion of the population who do not use services and, hence, run the risk of compromising the accuracy and reliability of their count. For example, an investigator interested in single homeless women may choose to sample from women?s shelters. The problem is that single women do not necessarily use the women?s shelter system; they can be found in men?s shelters, on the streets and in welfare motels. In addition, they may use other services that address a variety of other problems that they may have. For example, they may have problems with alcohol and drug abuse, mental health problems, employment problems, etc., and may contact and use services that are designed specifically to deal with those problems. While single homeless women do not constitute the ma- jority of their client base, many such organizations do serve a proportion of the population. Thus, excluding services where the likelihood of en- countering single homeless women is low, but not zero, results in an un- derestimation of the size of the population. Thus, most researchers agree that it is preferable to start with as wide a net as possible, in order to cap- ture as many homeless people as possible, and then narrow the focus of the analysis to the groups of interest after the data have been collected. Cost is another factor that is directly related to the number of sam- pling frames to be selected for use in a study. Investigators have found that the greater the number of locations in which counts must be taken, the greater the cost of the project, for two primary of reasons. First, a census or survey must take place at roughly the same time3 Second, the time frame associated with preparing, training, and coordinating both staff and the sites included in the study increases with the number of locations selected. The greater the number of sampling frames, the more time is required to contact the locations, gain entry, gather preliminary information about the client-base using the service, screen respondents about their usage of other services that makeup the other sampling frames in the study, and organize a count at that site. across all of the locations in the study in order to reduce the possibility of double- counting. Therefore, a large staff is required to carry out a simultaneous enumeration across the locations forming the basis of the study. Increas- ing the sampling frames, therefore, will require a proportional increase in the number of staff required to carry out the task at hand, which will, in turn, increase the cost of the project. 3 The sa me ti me ca n be interpret ed in two wa ys: (1) either at the same ti me over suc c essiv e da ys/ ev enings; and (2) acro ss all lo ca ti o ns in one eveni ng. Thus, the more preparatory time and time in the field spent, the greater the overall cost of the project. The cost of counting the homeless has varied depending on the size of the geographic area targeted for enumeration and the number of sam- pling frames to be used. Various researchers throughout Canada and the United States have reported project costs ranging anywhere from $30,000 (in Calgary), to $800,000 U.S. (the Washington DC Metropolitan Area Study), to the $10 millions the U.S Bureau of the Census budgeted to car- ry out the homeless component of the decennial census. Counting the homeless is an expensive and time-consuming process. Efforts to count the homeless at the national level require an inordi- nate amount of resources, both human and financial, as well as extensive advanced planning and coordination. Yet to date no national effort, Ca- nadian or American, has been deemed successful. And, while the U.S. government is continuing its efforts to improve the coverage of the na- tional census to include the homeless, the Canadian government has dis- continued its efforts to capture the homeless in the Canadian census. After a thorough investigation into the possible methodologies available for taking a census of the homeless population, the U.S. Bureau of the Census adopted a service-based method for use in their efforts to include the homeless in a national enumeration of the population. For the 1990 Census, counts in shelters and pre-identified street locations were carried out in a process known as the Shelter and Street Enumera- tion (S-Night). The S-Night enumeration counted persons in emergency shelters and visible in street locations, as well as persons who reported they had no permanent home elsewhere during the standard census of special places and group quarters (e.g., jails, institutions, etc.) on the night of March 20 and the early morning hours of March 21, 1990. Prior to S-Night, the Census Bureau compiled a national list of shelters from administrative records and requested every local jurisdiction, nation- wide, to supplement the list of shelters, street, and open public locations used by homeless persons at night. As anticipated, counting the home- less in the pre-identified street locations proved to the most problematic component of the S-Night enumeration. Indeed, the Census Bureau elected to exclude street locations from the 2000 census.4 The main criticism levelled against the S-Night efforts was with the site selection for the street component. The S-Night street counts were restricted to predesignated areas that the Census, working with local authorities, identified as high-density homeless areas. As Wright and Devine (1992) explain: In essence, the S-Night street enumeration was restricted to homeless per- sons who spent the night somewhere in these predesignated areas; street people outside those areas were not enumerated. Because Census S-Night resources were limited, coverage of entire cities was clearly out of the question; at the same time, the restricted nature and number of sites that were in fact searched strictly limited the completeness of the count (362). This problem is not restricted to the S-Night effort. Virtually every study that has attempted to count the street homeless has been limited by the difficulties inherent in trying to accurately count the number of homeless who live on the street, in abandoned buildings, on rooftops, in cars, under bridges, etc. There is no reliable method for selecting ?high- density? locations or for choosing street locations in which the homeless are most likely to be found. To date, researchers have relied on service providers and local authorities to identify the most likely sites. But re- gardless of the accuracy of the local authorities? knowledge about the street locations of the homeless, the homeless are extremely mobile and move on a constant basis. By the time the study goes to field, the home- less are likely to have moved to new locations not identified prior to the study. There is no reliable way of predicting the street locations where the homeless may be found on any given day. Both Rossi (1989) and Dennis (1993) employed a stratified sampling design based on census blocks to avoid the site identification problems associated with attempts to count the street component of the homeless population. While this is the most rigorous method for counting the homeless, and is considered the most scientifically valid method, both researchers report that it is an extremely expensive and logistically diffi- 4 For a detailed account of the problems and difficulties encountered during the S-Night enumeration, see Evaluation Review, 1992, Volume 16, Number 4. cult method to implement. Dennis (1993) reports that the costs associated with the street component of the DC*MADS study were so high (over half of the total budget) that the project was halted and redesigned to exclude the street sampling frame. Given these experiences, most researchers concur that the street component of any method for enumerating the homeless is extremely costly and produces the least satisfactory results. Moreover, the propor- tion of homeless actually living on the street is small, and the majority of street homeless can be captured in the shelter, soup kitchen and en- campment sampling frames. We therefore conclude that the following sampling frames represent the minimum number of locations required to ensure the best coverage of the population: shelters, soup kitchens, day programs, and outdoor encampments. The Mechanics of Service-Based Methods Service-based methods are principally carried out in two stages: the pre- sampling and the survey/census phases. During the pre-sampling or counting phase, detailed lists are developed of the sites and services where the homeless are to be found. Decisions are made concerning the time frame within which the count will take place (e.g., in one night, or over a period of nights). The services are contacted, information is re- quested on the number of homeless using the service, the optimal time for taking a count on site is determined, and permission is sought to car- ry out a count/survey on the site. Outdoor locations are canvassed and counts of the numbers of homeless are generated. The proportion of homeless using the services and found in the outdoor locations is esti- mated based on the information collected. Sample probabilities are then generated from the estimates of the numbers of homeless in each of the sampling frames. Finally, the screening tool and survey instrument are developed. During the survey/census phase, field personnel are sent out to count and survey the population. Individuals are approached at each of the sites, their participation in the study is requested, and, if they agree to participate, they are administered a screening instrument to determine whether or not they are homeless and what services they use. Finally, if the project involves collecting detailed information from the homeless, the questionnaire is administered to the appropriate respondents. The following description of the service-based method used in the Calgary survey of the homeless (McDonald and Peressini, 1991) provides a detailed example of the activities and tasks involved in implementing such a design. The goal was to survey a sample of the homeless popula- tion, not to take a census of the population. The primary difference be- tween sampling from and taking a census of the homeless population is that because a census requires counting/surveying every individual in each of the sampling frames, it does not require probability estimates of the proportion of homeless using each of the services. Probability esti- mates are used to determine the number of individuals to be selected from each of the sampling frames and are, therefore, not required for a census. However, estimates of the size of the population can be generat- ed based on the probabilities associated with the proportion of homeless in each of the sampling frames in a study. The Calgary Survey of the Homeless The first Calgary survey of the homeless (CSH) was carried out as part of a larger research project from January, 1990 to 1991, for the City of Cal- gary. The CSH used a sampling design developed by Burnam and Koe- gel (1988) in their study of the Los Angeles Skid row. Burnam and Koe- gel note that the main stumbling block in drawing a representative sample of homeless is the construction of an accurate sampling frame (e.g., a complete list of all homeless individuals in a population), and in selecting a sample from this listing such that each person has a known probability of being chosen for inclusion in the sample (1988: 118). They developed a method of selecting homeless individuals such that every homeless person in the Los Angeles Skid row area would have an equal chance of being selected into the sample. The strategy involved: estimating the relative proportions of the homeless population that ?passed through? various facilities over a month?s time (including facili- ties which served the unsheltered sector of the homeless population), and then randomly sampling, within these facilities, numbers of persons that were directly proportional to the average proportion of the population utilizing the facility over the period of a month (122). Before selecting their sample, they had to determine which facilities were used by the homeless and what services each facility provided. Af- ter conducting an investigation into the facility and service utilization of the homeless population, they determined that for their sample to be representative, three sampling sectors had to be distinguished: (1) Beds: persons using temporary sleeping quarters or beds made available to the homeless in shelters or through the provision of hotel vouchers; (2) Meals: homeless persons receiving free meals from missions or other programs, but not using beds; and (3) Congregating areas: homeless in- dividuals who made some use of missions and drop-in centres (Burnam and Koegel, 1988: 123). The next phase involved collecting data from each of the facilities included in each of the sectors, as well as from surveys of homeless indi- viduals themselves, to estimate the proportion of the population falling into each of the sectors over a month?s time (1988:123). Burnam and Koe- gel then determined the proportion of individuals using one or more of the facilities or services (that is, the overlap between the facilities) and adjusted their estimates of the numbers of people falling into each of the sectors. Finally, they randomly selected a sample of individuals within each sector on a given day. The result was the selection of a sample that consisted of a representative cross-section of homeless individuals ?on an average day? in Los Angeles?s Skid row. A preliminary survey of the homeless population in Calgary?s Skid row was carried out using similar methods and procedures. The goal of the initial survey was to determine the services and facilities used by the homeless and to estimate the proportions of homeless using each of the service sectors in order to develop a sampling list from which to draw a sample. Development of the Sampling Frames for the Survey In February 1991, the data required to determine the proportions of the homeless population using each facility included in the beds, meals and congregate area sectors were gathered. This information is needed to calculate the probabilities of homeless persons using the services and facilities in each sampling frame. First, the different type of services for the homeless in the Calgary?s skid row area was determined. Then in- formation was collected on the characteristics of the facilities such as: numbers of beds, eligibility for beds, length of stay in beds, numbers and times of meals served, other services (e.g., drop-in or congregate servic- es) and number of persons using other services during the study month. Table 1 presents a summary of this information. Table 1: Sampling Information f or the Calgary Survey of the Home less Facility No. of transient beds Number of beds Feb 1991 Meals served to transients Total meals Feb 1991 Congregate areas Single Men?s Hotel 138 3,811 Breakfast Lunch Dinner - - - - - - Calgary Drop- In Centre 100 2,992 Mid-morning Mid-afternoon 2,992 2,238 1,230 St. John?s Soup Kitchen - - Lunch 3,352 Salvation Army Soup Line - - Dinner 1,292 Source: McDonald and Peressini, 1991. Two facilities offered beds to the homeless, and three facilities of- fered meals to individuals other than those whom they were housing as part of a program, together providing approximately 233 meals per day to the homeless and transient population. However, these figures did not accurately reflect the number of homeless individuals served, because more than one meal was served per day at one of the facilities, individu- als could eat at more than one facility in a day, individuals were allowed to have more than one serving during the same serving period, and meals were not restricted to homeless individuals. Only one facility contained an indoor congregating area. Because of the constant movement of individuals in and out of the facility, the re- search team found it extremely difficult to determine exactly how many individuals the facility served. This facility took a daily count of the number of individuals present at noon. Using this information, the re- search team calculated that the total number of individuals (who may or may not have been homeless at the time) using this congregate area for the study month was no more than 1,230 people or approximately 44 people per day. The researchers concluded that, taking the size and characteristics of the homeless population into consideration, it would be redundant to partition the sample into three sectors. Because only one facility in Cal- gary?s skid row area offered a drop-in service (that is, provided an in- door congregating area) and that same facility provided both beds and meals, it was decided that the portion of the homeless using this service would be captured, or represented, in the bed and meal sectors. Thus, the services provided to the homeless were assigned to the meals and beds sectors by the research team in the following way: Beds Sector: The Single Men?s Hostel, The Drop-In Cen- tre (Night Program) Meals Sector: The Drop-In Centre (Meals service), St. John Soup Kitchen, Salvation Army Soup Line The next stage required first, the identification of all other congregat- ing areas outside Calgary?s Skid row attracting concentrations of home- less individuals, and second, taking a census of homeless persons in these areas. They included City Hall (on Sunday mornings), the Public Library, Devonian Gardens (part of a major shopping mall in the down- town core), and three fast-food outlets. In total, 75 people were counted in these areas over several periods between mid-February and the end of the month. This number, however, was not considered an accurate ref- lection of the number of people who actually passed through each of the congregate areas. Thus, similar to Burnam and Koegel?s survey, it was decided that this subpopulation of the homeless would be captured in the samples of the other two sectors. The decision not to sample from the other congregating areas al- lowed for the possibility that a portion of the homeless who did not use the bed and meals services and did not congregate at the drop-in centre would not be selected for inclusion in the final sample. Thus, ultimately, the final sample chosen would not represent the entire homeless popula- tion. To ensure that a minimal amount of bias was introduced into the sample by excluding this sampling sector, a short survey of the individ- uals in the ?other? congregating areas was carried out in order to deter- mine their likelihood of being captured in the beds and meals sectors. The sample design of the auxiliary survey involved sampling a quo- ta (a set number of individuals based on their proportions, derived from the observations of the areas) of those individuals available in each of the areas at the time of the survey. The survey was conducted over four days. Individuals were asked a series of questions that established whether they were homeless (that is, they did not have a room, apart- ment, or house of their own, or had not been in their own place in the previous month). If they met the criteria for homelessness they were asked three additional questions to determine whether, in the previous month, they had: (1) slept in a bed in any of the facilities in the beds sec- tor; (2) eaten a meal at any of the settings in the meals sector; or (3) spent time in the drop-in centre?s congregate area. Thirty-six persons were approached. Five refused to participate in the mini-survey, for a completion rate of 86 percent. Of the 31 individu- als agreeing to participate, 25, or 80 percent, had passed through the bed sector, meal sector, or both during the study month. Only 3 people (10 percent) had used the congregating area at the drop-in centre. Further- more, these three people reported using at least one of the facilities in both the beds and meals sectors over the month prior to being surveyed. These results, therefore, support the decision to exclude the ?other? con- gregating areas from the overall sampling frame. The majority of people in the outdoor congregating areas would be represented in the final sample by virtue of the fact that they were also using the services availa- ble in the Bed and Meal sectors. Thus, it was concluded that allocating the sampling frame to the Bed and Meal sectors would produce a sample of homeless individuals which would be representative of the entire homeless population in the city. Allocating the Sample across the Bed and Meal Sectors The next step was to determine how to proportionately sample from each of the sectors. Following Burnam and Koegel?s reasoning (1988: 133), the researchers concluded that the population using a bed in the bed sector was a subset of the population that receives meals (e.g., the meals sector is more inclusive than the beds sector). Thus, as Burnam and Koegel put it, ?as one moves from the category of beds to meals... one casts a wider net, drawing in individuals who are less and less in- volved in service utilization? (1988: 134). It was decided that the strategy would be to sample those in the beds sector first, thereby allocating the maximum proportion of the sample to this category, and then sample the meals sector (accessing those individuals who would have a high proba- bility of being excluded from the sample derived from the beds sector). Before this step could be carried out, an estimate of the amount of overlap between the two sectors had to be determined. A third survey of the facilities included in the meals sector ? the most inclusive of the sec- tors ? was therefore completed. The Meals Enumeration consisted of four questions: 1. Do you currently have a room, apartment or house of your own? 2. Have you stayed in your own place in the last 30 days? 3. Have you slept in a bed at the Single Men?s Hostel or the Drop-In Centre in the last 30 days? 4. In the last week have you eaten at any of the following places? The Salvation Army Soup Line? The 10:30 and 2:30 meals at the Drop-In Centre? and The St. John?s Soup Kitchen? A complete census of all individuals using the meal services pro- vided at the Soup Line, Drop-In Centre and Soup Kitchen at four differ- ent times over the study month was carried out. In total, across the three locations, 264 individuals were ap- proached. Of these, 26 declined to be interviewed, resulting in a comple- tion rate of 90 percent. From the survey, 61 people were defined as not homeless (e.g., they answered ?yes? to questions 1 or 2). Thus, informa- tion on sector overlap was available for 177 people. Of these 177 individ- uals, 78 percent (138) had slept in a bed at either the Single Men?s Hostel or the Drop-In Centre in the study month, while 22 percent (39) had re- ceived meals but had not slept in a bed in the beds sector in the month prior to being interviewed. The goal of the CSH was to obtain 100 interviews of the homeless. Factoring in a refusal rate of 10 percent (based on the refusal rate ob- tained in the meals enumeration survey), it was calculated that a sample of 110 individuals would have to be initially selected to achieve a final sample size of 100. Knowing that 22 percent of the population used meals but not bed services, it was calculated that 24 people needed to be selected from the meals sector. The remaining 86 people (78 percent) would be drawn from the beds sector. All that remained to be done was to ensure that the individuals to be sampled across the different facilities within each sector had an equal probability of being selected. Sampling within the Beds Sector Two primary facilities are available to people seeking beds for which they did not have to pay in the City of Calgary: The Single Men?s Hostel (SMH) and The Calgary Drop-In Centre (CDIC). Before determining the proportions of interviews to take place in each facility, however, the de- gree of utilization overlap between the two facilities had to be calculated. In other words, to ensure that interviews were allotted to each facility in the correct proportion, this overlap had to be accounted for and con- trolled in the overall sampling design. Using Burnam and Koegel?s design, an estimate of the amount of overlap between the facilities was derived and subsequently controlled for by going through the following steps. In the first step, a list of differ- ent people using each facility for the study month was compiled. This list provided a count of the different people who had slept in each facility in a 30-day period, thus providing an estimate of the number of people that the beds in each facility represented. In total, 510 different people slept in a bed at the SMH and 529 different people slept in a bed at the CDIC. Next, the list from each was compared to the other to see which people had slept in a bed in more than one place. Where overlap was found, it was split evenly or proportionately weighted between the two lists. For example, a person who had slept in a transient bed at the SMH and in a bed at the CDIC during the designated period (February) was counted as one-half in each of the two categories. This procedure served to adjust the estimated number of persons represented by a facility to account for the overlap between facilities. In total, 239 people out of 1,039 had slept in a bed at both the facilities in February. A weight of 0.5 was assigned to the individuals whose names appeared on both of the lists. As a result, the total number of different people using a bed in the beds sector in one month was calculated to be equal to 800; after adjusting for the overlap between the two facilities. Finally, using the figures derived in the first two stages, the propor- tion of the total population of different people using beds for each site in the beds sector, controlling for overlap, was calculated as follows: the SMH = 48. 81 percent and the CDIC = 51 .19 percent. The sample of de- sired interviews, 86, was proportionately allocated to transient beds in each of the facilities as follows: SMH, 42 interviews and CDIC, 44 inter- views. Sampling Within The Meals Sector Three organizations provide meals to homeless and transient people in the city of Calgary: the CDIC (2 meal settings at 10:30 am and 2:30 p.m.), the Salvation Army Soup Line (dinner), and the St. John?s Soup Kitchen (lunch). The goal here was to control for the overlap in usage between the meals settings, such that each individual in the total population availing themselves of meals in each of the facilities would be counted only once. Burnam and Koegel?s design was once again employed. Table 2 presents a summary of the steps and calculations that were taken in the CSH to replicate Burnam and Koegel?s design for deriving an estimate of the proportions of eligible persons served by each meal site, adjusted for the overlap in usage between the four sites. Using these proportions the desired number of interviews to be conducted in the meals sector, 24, was allocated as follows: ? St. John?s Soup Kitchen: 19 interviews ? Drop-In Centre (10:30 am): 1 interview ? Drop-In Centre (2:30 p.m.): 1 interview ? Salvation Army Soup Line: 3 interviews ? Total: 24 interviews Table 2: Summary of Steps Used to Derive Proportions of Interviews to be Conducted at Ea ch Facility in the Meals Sector Procedure Soup Kitchen Drop-In Centre Salvation Army Total N a.m. p.m. Not Homeless 23 22 10 6 61 Slept in a Bed in the Beds Sector 36 69 28 5 138 Eaten a Meal but Did Not Sleep in a Bed 27 2 4 6 39 Average # of Meals (Feb) 10.5 3.4 2.4 1.9 --- Total # of Meals Served (Feb) 3,352 2,992 2,238 1,292 9,874 Est. of the # of Different Persons served (Total/Average # of Meals) 334 883 948 663 2,828 Proportion of Eligible Persons at each site .69 .05 .10 .15 --- Total # of Eligible Persons served by each site (Est. of Different Persons x Prop. eligible) 231 45 97 102 475 Eligible # of Persons eating at each site, adjusted for overlap 22 6.5 3 7.5 39 Proportions of Persons eating at site, adjusted for overlap .56 .17 .08 .19 --- Total # of eligible persons served by each site, adjusted for overlap 130 8 8 20 166 Proportions of Eligible persons served by each site, adjusted for overlap .78 .05 .05 .12 --- Table 3 presents a summary of the sample of the sub-groups mak- ing up the homeless population, living in the skid row area of the city that was surveyed. The proportion of each group is outlined and the number of interviews that were conducted in each stratum specified. Table 3: Sampling Strata Summary No. of People to be Randomly Sampled Proportion of the Population (%) Beds: Single Men?s Hostel Calgary Drop-In Centre Sub-Total 42 44 86 38 40 78 Meals St. John?s Soup Kitchen Calgary Drop-In Centre (a.m.) Calgary Drop-In Centre (p.m.) Salvation Army Soup Line Sub-Total 19 1 1 3 24 17 01 01 03 22 Total Number of Interviews 110 100 In all, 110 homeless men and women were administered a 14-page questionnaire containing questions that would allow us to derive a com- prehensive description of the respondents, including their housing histo- ries; specific needs for help; problems with health, drugs, alcohol and mental disorders; a demographic profile; early childhood experiences; problems with the police; work histories; and other relevant experiences. The survey instrument was developed, pre-tested, and revised over the course of carrying out the interviews in the meals sector. The revi- sions involved simplifying questions and shortening the questionnaire (it was originally 25 pages long). The interviews took roughly 25 to 65 mi- nutes to complete, with an average completion time of 45 minutes. In total, 159 persons were approached and asked to participate in the study. Of these, 35 people refused to take part. The remaining 124 people who agreed to take part in the study were asked two screening questions designed to ensure that the individuals chosen for inclusion in the study were, in fact, homeless. As a consequence of the screening pro- cedures used, 14 people did not meet either criteria and, hence, were ex- cluded from the study. In the beds sector, 24 interviews were conducted. Altogether, 35 people were approached. Of these, four refused and seven were found not to be homeless according to the criteria used. Three interviewers car- ried out the survey at each of the locations in the meals sector: St. John?s Soup Kitchen (N=19), the Salvation Army Soup Line (N=3) , and the Cal- gary Drop-In Centre (a.m. N=1; p.m. N=1). People were approached as they lined up for a meal at each of the services. Starting from the first person in the line, every fifth individual was selected and asked to participate in the survey. The selection proce- dure continued until the pre-determined number of individuals to be interviewed at each site had been achieved. In some instances, depend- ing on the setting, the participants, the number of interviews to be con- ducted at the particular site, and the amount of time allowed the inter- viewers at each site, the interviewers had to return over a number of days and repeat the selection procedure until the desired number of in- terviews was completed. It took five interview sessions to complete the required 19 inter- views at the St. John?s Soup Kitchen. The soup kitchen served only one meal per day from noon to 1 p.m. The interviewers were granted an ad- ditional hour in which to carry out their interviews. Since only two in- terviews could be carried out per interviewer during any one sitting, the 19 interviews were completed over five consecutive days. Both the drop- in centre and the Salvation Army soup line required only one sitting to obtain the required number of interviews. In the beds sector, 86 interviews were completed: 44 at the drop-in centre and 42 at the Single Men?s Hostel. Like the St. John?s Soup Kitch- en, because of time limitations, interviewing for the beds sector took place over seven days. A total of 124 people were approached and asked to participate. Of these, 31 refused and seven were found not to be home- less according to the study criteria. Between six and eight interviewers were used, depending upon the site and the number of people to be interviewed. Only two interviews could be completed each time. The Single Men?s Hostel presented a unique problem. The residents at the hostel were required to be out of the building by 8:00 a.m. and were not allowed back into the hotel until 4:00 p.m. After dinner they were free to spend their time in whatever way liked, as long as they were in their bed or room for ?lights-out? at 10:30 p.m. Given the limited amount of time that all of the residents would be available for interview- ing, interviews were conducted as the residents lined up for breakfast (between 7 and 8 a.m.) and dinner (4 to 6 p.m.) to maximize the time the researchers had for interviewing and the pool of residents from which a sample could be selected. Individuals were selected from the meal line-ups at both settings in the beds sector. Initially the research determined that every tenth person would be selected. However, because the residents at each facility did not use the facility at the same time, this sampling interval was too broad. The researchers therefore set the sampling interval at five. Inter- viewing took place over three consecutive days at the Drop-In Centre and over two days at the Single Men?s Hostel. Service-Based Methods: Further Requirements Employing a service-based method for counting the homeless requires both a screening instrument and survey instrument (intake form and questionnaire). Screening instruments allow the interviewer to identify which of the individuals using services are homeless. For example, the Calgary Survey of the Homeless defined persons as ?homeless? if they did not currently have a room, apartment or house of their own, or had not been in their own place within 30 days of participating in the study. Accordingly, the screening criteria consisted of the following two ques- tions: (1) Do you currently have a room, apartment or house of your own for which you pay to live in? (2) Have you stayed in your own place within the last 30 days? A screening instrument may also contain questions designed to col- lect information on the other services that the homeless use. Questions collecting information on the socio-demographic characteristics of the respondents may be incorporated in either the screening instrument or the survey instrument. Both types of additional information can be used to develop and assign a unique identifier to each individual counted. Unique identifiers permit the investigator to cross-reference cases from one sampling frame to the next to eliminate the problem of double or multiple counting across frames.5 Unique identifiers can take a number of forms. The U.S Bureau of the Census uses the respondent?s Social Security Number. Another unique identifier consists of a composite descriptor based on the indi- vidual?s date of birth, gender, race, and name. For example, a person whose birth date is January 1, 1965, who is male, black, and whose name is Fred James Smith could be represented by the following number: 01016501FJS, where 010165 is their birthdate, 0 the male code for gender, 1 the black code for race, and FJS the first initials of their name. This in- formation could be collected from everyone surveyed and the final count adjusted for multiple occurrences in the database. This type of identifier is the most common one used and has been employed in service-based methods, as well as computerized information management systems. Because the homeless may use more than one of the services included in a count, a method for identifying cases that distinguishes between individuals is required. The last issue is that of the types of data to be collected. The most important limitations on the amount of information collected from res- pondents are the amount of time that the investigator has to ask ques- tions (which may be determined by the operating procedures of the or- ganizations involved) and the respondent?s willingness to cooperate. The questionnaire employed in the CSH included a range of ques- tions or variables about respondents? sociodemographic and socioeco- nomic background, their work history, their levels of income, their health problems (mental and physical), their use of drugs and alcohol, their family background, their social networks, their daily activities, their rates of service utilization, and their history of homelessness. Similar types of information are collected by investigators using computerized information management systems (CIMS), such as the ANCHoR system. Typically, however, those employing CIMS collect substantially less information from the homeless than researchers carry- 5 The issue of double counting is only problematic for those designs consisting of an enumeration that takes place over an extended period of time (e.g. more than one night). For designs other than one-night blitzes, it becomes necessary to develop a unique identifier to control for the possibility of counting an indi- vidual more than once of the period in which the enumeration takes place. ing out actual surveys. The following example of the questions included on the application for hostel assistance used by the Municipality of Met- ropolitan Toronto shows that usually only information such as the res- pondent?s basic demographic characteristics (e.g. age, gender, race, date of birth, last place of residence), reasons for request of services, disposi- tion of case, and the relevant accounting information is collected. 1. Full Name 2. Gender 3. Date of Birth 4. If accompanied by spouse, give first name 5. If accompanied by children, give ages 6. Last permanent address SERVICE INFORMATION 1. Major Reason for service: A. Spousal abuse B. Spousal abuse - psychological C. Parental abuse ? sexual D. Parental abuse ? other E. Family breakdown ? general F. Eviction - landlord G. Eviction ? other H. Transient lifestyle I. Moving to City J. Stranded in City K. From treatment ? psychiatric L. From treatment ? other M. From corrections N. Fire/unsafe premises O. Other (specify) 2. Disposition of case: A. Found new address in community B. Returned to spouse/parents C. Continued on at another hostel D. Moved in with friends/relatives E. Left the city F. Admitted to hospital G. Whereabouts unknown H. Other (specify) ACCOUNTING INFORMATION 1. Name of Hostel 2. Month 3. No. of persons 4. Admission date 5. Leaving date 5. Date during which client(s) slept in hostel 6. Total nights service X per diem rate = amount of this claim $ These data collected by the Municipality of Metropolitan Toronto are most relevant to their needs. For other studies, the type of questions asked will reflect the interests and needs of those collecting the data. Conclusions Conventional strategies for counting individuals based on households simply do not apply to the homeless. The homeless are a transient, mo- bile, and elusive population which cannot be consistently located in a single place. They do, however, tend to aggregate in known locations for short periods of time. These places include shelters, soup kitchens, medi- cal services, outdoor congregate areas, indoor drop-in centres, employ- ment offices, and the like. In the mid-1980s American researchers acknowledged this fact and began to develop strategies for sampling from the population in these locations. However, carrying out counts in these types of locations re- quired permission and cooperation from those providing the services that the homeless use. Initially, most, if not all, met with some degree of resistance. It took at least ten years to develop and build up a level of trust between the service provision community and researchers so that the methods for counting the homeless can be easily implemented and carried out. Researchers still run into resistance, but a tacit agreement has been made that, in most instances, researchers and providers are working towards the same goal. In Canada, government officials are skeptical of the estimates of the numbers of homeless that the service community has produced. The ser- vice community is equally skeptical of government estimates, and ques- tions the value of trying to count the homeless. For them, the numbers are not important, rather it is the amount and quality of care and services that they can provide which is foremost in their minds. This difference must be addressed before any progress can be achieved and a common goal devised in order to create the level of co-operation required to suc- cessfully count the homeless. This chapter is drawn from a report called Estimating Homelessness: To- wards a Methodology for Counting the Homeless in Canada published by CMHC, Ottawa, in 1996. Tracy Peressini is an Associate Professor (Sociology and Social Development Studies) and Director of the Centre for Social and Community Development at Renison University College, University of Waterloo. Lynn MacDonald References is a Professor in the Faculty of Social Work and Director of the Institute for Life Course and Aging at the University of Toronto . J. David Hulchanski is an Associate Director of Research for the Cities Centre and a Professor in the Fa- culty of Social Work, University of Toronto. Bentley, D. (1995). Measuring homelessness: A review of recent research. Winnipeg, MB: Institute of Urban Studies. Burnam, M. A., & Koegel, P. (1988). Methodology for obtaining a representative sample of homeless persons: The Los Angeles skid row study. Evaluation Review, 12(2), 117-152. Burt, M. A. (1991). Causes of the growth of homelessness during the 1980s. Hous- ing Policy Debate, 2(3), 903-936. Burt, M. A. (1992). Practical methods for counting homeless people. Washington, DC: Interagency Council on the Homeless and the U.S. Department of Housing and Urban Development. Burt, M. A. (1993). Over the edge: The growth of homelessness in the 1980s. New York: Russell Sage Foundation. Dennis, M. L. (1991). Changing the conventional rules: Surveying homeless people in nonconventional locations. Housing Policy Debate, 2(3), 701-732. Dennis, M. L. (1993). Coverage of service-based methodology: Findings from the DC*MADS homelessness study. Paper presented at the Towards Census 2000 Conference, Arlington, VA. Dennis, M. L., & Iachan, R. (1992). Sampling people who are homeless: Implications of multiple definitions and sampling frames. Paper presented to the American Psychological Association?s 100th Annual Conference, Washington, D.C. Dennis, M. L., & Iachan, R. (1993). Multiple frame approach to sampling the homeless and transient population. Journal of Official Statistics, 14(5), 1-18. McDonald, P. L., & Peressini, T. L. (1992). The East Village community study: Final report. Calgary: The City of Calgary. Rossi, P. H. (1989). Down and out in America: The origins of homelessness. Chicago: The University of Chicago Press. Taueber, C. (Ed.). (1991) Enumerating homeless persons: Methods and data needs [Conference Proceedings] . Washington, DC: U.S. Department of Com- merce, Bureau of the Census. Wright, J. D., & Devine, J. A. (1992). Counting the homeless: The Census Bureau?s S-Night in five U.S. cities. Evaluation Review, 16(4), 355-364. Chapter 8.4 Ethics in Research with Homeless Women EMILY K. PARADIS Women?s homelessness is a specific, concrete product of an oppressive social system critiqued by feminist and community psychologists. It emerges from a complex interplay of factors, including economic injus- tice, racism, sexism, and marginalization of people with disabilities (Bax- ter, 1991; Farge, 1989; Ralston, 1996). Feminist and community psychologists have argued that psycholo- gy traditionally has functioned alongside other social institutions to prop up and justify the existing social system (Fine & Gordon, 1991; Mulvey, 1988; Serrano-Garcia, 1990). One role of the institution of psychology in the maintenance of an oppressive status quo has been to obscure social injustice by locating the deficits responsible for marginalization within marginalized communities and individuals (Fine & Gordon, 1991). The conventions of traditional quantitative research?including the study of individual traits without reference to social context, and an emphasis on ?objective? measurement of variables?produce research that fulfils this role. For example, Shinn (1997) has pointed out that epidemiological re- search focused on characteristics of ?the homeless? constructs homeless- ness as an individual trait, rather than a state situated in a social and his- torical context. Such research implicitly holds victims of injustice respon- sible for their own marginalization, because it seeks explanations for homelessness in the behaviour of individuals. Feminist psychologists have described the structure of traditional research as exploitative (Grossman, Gilbert, Genero, Hawes, Hyde, & Marecek, 1997). Community psychologists Chavis, Stucky, and Wan- dersman (1983) advocate a shift away from the traditional relationship between researchers and communities, described in some communities as ?experimental colonialism? (431). Indeed, research is exploitative if the researcher?s interests alone shape every step of the research process, from the formulation of the question through the collection of the data to the write-up and dissemination of the results. And research resembles a colonial economy when researchers enter uninvited into the world of participants, extract a resource called data, process this resource into a product called theory, and use the product only for their own ends. On the other hand, transforming exploitative aspects of the traditional mod- el can yield research that promotes the interests of marginalized people and advocates for changes to an oppressive social system. Traditional ethics codes governing psychological research (Ameri- can Psychological Association, 1992; Canadian Psychological Associa- tion, 1991) are written with traditional, quantitative research in mind.1 1 These codes have been revised since the original publication of the article that is the source for this chapter (APA, 2002; CPA, 2000). As a result, the sections cited may not be valid for the current versions of the codes. In some respects, the codes? responses to some of the issues raised in this chapter have evolved: for example, the Canadian code now places more emphasis on ethical obligations to communities. The ethical issues explored here, however, remain relevant. They reproduce a positivist scientific model in which the researcher strives to be an objective observer and the advancement of science is constructed as inherently valuable. The research participant?s role is to provide data, and it is assumed that participants lack the ability to in- terpret their own experiences. Both researcher and participant are re- duced to their roles in the process; their social positions, vested interests, histories, and contexts are seen as factors to be set aside lest they impede objectivity (Oakley, 1981). While recent revisions of ethics codes refer to issues of context and difference, the depoliticized manner in which these are presented allows for interpretations that may be antifeminist rather than feminist, for ex- ample, when guidelines that warn against discrimination based on gender are invoked to justify sexual assault research in which the defini- tion of coercive sex includes male participants? experiences of being ?coerced? into foreplay as well as female participants? experiences of unwanted sexual activity (Condor, 1991). Add-on statements about is- sues such as gender or research with marginalized communities do not account adequately for the role of the institution of science in maintain- ing systemic dominance and oppression. Arguing that in traditional research, claims of ?objectivity? often disguise the researcher?s investment in an oppressive status quo, femin- ist and community psychologists (Fine, 1992; Mulvey, 1988; Serrano- Garcia, 1990) have called for research that is explicit in its support for the interests of marginalized individuals and communities. In this context, research ethics aim to prevent exploitation or oppression within the re- search relationship (Grossman et al., 1997), and to ensure that research promotes the interests of marginalized groups (Rappaport, 1990). Just as each stage of traditional research can reenact social relationships of do- minance and oppression, feminist and community psychology research should manifest a liberatory potential in its purpose, planning, methods, and outcome (Fine & Gordon, 1991; Serrano-Garcia, 1990). While feminist and community psychologists share the goal of transforming psychology (Mulvey, 1988), they have described different strategies for creating liberatory research. Feminist psychologists often study aspects of women?s lives that have been ignored or distorted by traditional research (Grossman et al., 1997). While all methods of inquiry may be used in feminist research (Grossman et al., 1997; Peplau & Con- rad, 1989), emphasis is placed on qualitative methods that capture nuance and complexity, or on politicized analyses of quantitative data. Feminists attend to the intersubjective nature of the research relationship and attempt to share power between researcher and participant (Oakley, 1981). Articles on feminist research may reflect these aspects of the process?for example, in the use of first-person voice to challenge the idea of the researcher?s objectivity. Community psychologists, instead of studying the intrapsychic lives of individuals, focus on issues of concern to marginalized commun- ities (Fawcett, 1990). Research questions seek solutions to community problems through the identification of strengths and resources (Rappa- port, 1990), the development of strategies for empowerment or preven- tion, and the explication of the role of social structures in oppression. Community research sometimes employs participatory methods, in which the community collaborates in the design, data collection, analy- sis, dissemination, and implementation of research (Serrano-Garcia, 1990). Instead of keeping the research within psychology through the publication of journal articles, community psychologists may return it to the community (Chavis, Stucky,& Wandersman, 1983) through reports, public forums, or action adapted to community needs. Unfortunately, research by feminist and community psychologists has not always lived up to a stated ethical commitment to liberation. Most research reports published in feminist journals in the late 1980s adhered to traditional, positivist methods in the research and writing (Walsh, 1989). As well, the individualistic focus of most feminist research fails to promote the collective action necessary for liberation (Fine & Gordon, 1991). Meanwhile, macro-level theories in community psychol- ogy have not necessarily translated into micro-level research on specific, real-life situations of oppression (Mulvey, 1988). In spite of embracing an ethic of empowerment, community psychology as a discipline has large- ly failed to take a stand on specific social issues (Mulvey, 1988). This chapter argues that feminist and community psychologists can and must take a stand on a particular social issue: homelessness. Re- search offers a means for psychologists to challenge the social injustice at the root of homelessness; however, it is vital that the research itself not reproduce this injustice in its process or products. Feminist and commu- nity psychology offer mutually enhancing approaches to the develop- ment of just and ethical research on homelessness. This chapter applies feminist and community psychology ethics to the analysis of examples of research with women who are homeless. I examine some ethical questions that can arise throughout the research process, from the planning stage, through data collection and analysis, to the writing and dissemination of results. I critique the failure of decisions based on traditional ethics codes to promote the interests of participants, and explore the challenges women?s homelessness poses to the tradi- tional understanding of ethical issues such as consent, privacy, harm, and bias. Finally, I propose some strategies for transforming research with women who are homeless into a means for challenging oppression and promoting liberation. While this paper is a call to action for feminist and community psy- chologists on the issue of homelessness, it also serves a second pur- pose?that of refining feminist and community psychology ethics through application to specific situations. Feminists and community psy- chologists must ground our ethical development in the messiness of eve- ryday situations instead of abstract principles if we hope to do honoura- ble research in a deeply flawed world. To test our ethics with reference to research with homeless women resonates with the community psy- chology value of making research accountable to the most vulnerable group (O?Neill, 1989), and with the Black feminist principle that radical change can result only from bringing the knowledge and interests of those most marginalized to the centre of our analysis (hooks, 1984). Accordingly, this paper brings the interests of women who are homeless to the centre of the discussion of research ethics. The examples that follow illustrate ethical dilemmas that can arise throughout the re- search process, with reference to situations from research with homeless women. Some of these examples are drawn from my own experience; others are based on current published research or are imaginary situa- tions that reflect existing trends or problems. Planning the Research: Begin Before the Beginning Through my work as a graduate assistant, I had access to a database that is maintained by a hospital-based sexual assault care centre. For the past five years, anonymous data have been collected about women and men who have used the centre?s emergency services after experiencing sexual assault. The database contains a wide range of information on clients, including their current living situation, details about the assault, their relationship with the assailant, their history of physical and sexual as- sault or abuse, and their mental health. Using the data about the clients? living situations, I was able to identify a large subgroup of women?10% of those in the database?who were homeless at the time of their contact with the centre. Realizing that the database offered a unique opportunity to study sexual violence against homeless women, I wanted to explore the data and document my findings, but I was uncertain how to ap- proach this exploration. This example from my own experience depicts several concerns that can arise when planning research on women?s homelessness. The ab- sence of participants from a research project means fewer, and simpler, ethical dilemmas at the planning stage for the traditional researcher. Ac- cording to APA standards (APA, 1992, 6.06? 6.10), my only obligation was to plan this project ?in accordance with recognized standards of scientific competence and ethical research,? such as ensuring that the database met criteria for informed consent, minimizing the possibility that the results would be misleading, and performing only tasks I was competent to perform. By contrast, for the feminist or community psy- chology researcher, the absence of participants raises questions with re- gard to accountability, point of view, provenance of the data, and formu- lation of the research question. Far from wishing to remain ?objective,? I was determined that my research should champion the interests of women who are homeless, the community implicated in my study. Instead of reducing my ethical obli- gations, the absence of participants with whom to negotiate the purpose and process of the research left me responsible for determining the community?s interests and evaluating unforeseen consequences. O?Neill (1989) points out that this is a common dilemma for community psy- chologists: how can one remain accountable to the most vulnerable group when they are not directly represented in the research? Regardless of my good intentions toward the community being stu- died, I also had to acknowledge that this research would reflect only my point of view, necessarily shaped and limited by my own experiences, including my roles as researcher and student, my work as an activist and service provider with women who are homeless, and my lack of personal experience of homelessness. Feminist and community psychologists see knowledge as situated (Landrine et al., 1995) or socially constructed (Ser- rano-Garcia, 1990); researchers must articulate their own positions to recognize the limits of their knowledge. Concerned with how to interpret the data without access to participants? interpretations, I prepared by exploring sources of information that could help me ?read? the data, in- cluding sources beyond the traditional psychology literature. These in- cluded women?s stories about homelessness told to me as a service pro- vider, anecdotal information from centre staff, feminist theory on trauma and violence, and qualitative research representing homeless women?s accounts of their experiences (Baxter, 1991; Ralston, 1996). As a feminist I believed I should take a critical stance in interpreting data gathered in an institutional setting. Women who are homeless often must rely on institutions for their subsistence needs, or are forced into contact with institutions such as the criminal justice or medical? psychiatric systems. The dependence or coercion that may characterize their contact with institutions is one aspect of their ?captive? status (Ro- senthal, 1991, p. 109). Rosenthal suggests that research with participants who are ?captive? reveals as much about the institution and its require- ments as it does about the participants, who are likely modifying their behaviour to fit with institutional norms. It was important that my re- search plan account for the circumstances under which the data was provided: in crisis, in a hospital, possibly in the presence of police. I ac- counted for this situation, in part, by questioning ?silences? in the data. I noted topics on which large numbers of participants had not provided information, and wondered which institutional factors might have con- tributed to these silences. Finally, formulating the research questions was an issue of concern. Feminist theorists have noted that questions frame the research by iden- tifying what information is relevant, and that they often incorporate the researcher?s hidden assumptions or biases about the research topic (Lan- drine et al., 1995). Though I did not have to worry about direct harm to participants in the course of the study, I was aware that certain questions could produce results that would be harmful to the community as a whole. I wanted to avoid asking questions that might contribute to the stigmatization of homeless women. For example, data were available on whether the victim or assailant had been using drugs or alcohol at the time of the assault, but I believed that to study this area using only cate- gorical data would erase the nuance and complexity of homeless wom- en?s experiences of substance use, and risk perpetuating both stereotypes about homeless women and victim-blaming myths about sexual assault. The research plan included consideration of a number of factors re- levant to feminist and community research approaches: my accountabili- ty to the community being studied, the limits of my ability to interpret the data, the contributions of my own experience, and the potential for perpetuating stereotypes. As a part-time graduate assistant, I lacked the resources to develop a forum for community participation in the re- search plan, which would have been the ideal approach from a commu- nity psychology perspective (Serrano-Garcia, 1990). Instead of attempt- ing to do conclusive research without community participation, I treated the study as an exploratory first step that could lay the groundwork for community-based and participatory interventions. The data demonstrated that the majority of homeless clients of the centre had multiple experiences of assault in their lifetimes; anecdotal information from staff suggested that homeless women rarely accessed the centre?s free, follow-up counselling services. These results suggested that homeless women who seek hospital services after a sexual assault may be especially vulnerable to post-traumatic stress due to their life- time experience of violence and their current circumstances. Meanwhile, these women are unlikely to use the services that could help them cope with the acute stress of the recent assault. I concluded that while women who are homeless frequently use the centre?s crisis services, it appeared that the follow-up services were not meeting the needs of this group. A fruitful second step for this study would be to ask women who are homeless about their use of services after sexual assault and their ideas for making these services more user-friendly. It may be that these women are coping with sexual assault using other resources, such as cri- sis lines, informal support networks, or counselors at hostels and drop-in centres. On the other hand, women who are homeless might lack access to support after a sexual assault. If that is the case, participants could provide valuable information about factors that would improve the ac- cessibility of the follow-up services offered by the centre. Service provid- ers who have worked with homeless women might be able to suggest some improvements, such as the location of services outside a hospital setting, availability after business hours, and flexibility with regard to appointments. Asking the community directly, though, would enrich the research with women?s concrete expertise about their own needs and situation. Collecting Data: Create an Empowering Process A psychologist receives a grant to do research on stress and coping among homeless women. She approaches the director of a drop-in centre where she has been a volunteer for two years, and receives permission to ask clients to participate. She develops an open-ended interview sche- dule that focuses on stressors, symptoms of depression, and coping strategies. The interviews and debriefing will take about two hours in total. Reasoning that the women should be considered equal partners in the research, she decides to pay them $30 each for the interview, as that is the rate her graduate research assistants are paid. She starts looking for participants, beginning with 10 women she has come to know quite well through her volunteer work. Her initial response rate is very high: only one woman refuses. Soon word spreads about the study, and women begin to approach her to ask if they can participate. In this hypothetical scenario, the context of women?s homelessness reframes ethical concerns raised by traditional codes. In data gathering, the researcher must ensure that consent is both informed and voluntary (APA, 1992, 6.11; CPA, 1991, I.14, I.21), that privacy is respected, (APA, 5.03; CPA, I.33), and that participants are not harmed in the process (APA, 1992, 1.14; CPA, 1991, II.2). The context of homelessness presents particular challenges to ensur- ing voluntary consent during data collection. First, as already noted, women who are homeless may be in a dependent or coercive relation- ship with institutions in which research takes place, or with the research- ers themselves. In the above example, the agency with which the re- searcher is associated might represent participants? sole source of subsistence, advocacy, or emotional support. Therefore the APA (1992, 6.11[c]) and CPA (1991, I.31) guidelines about participants in a depen- dent relationship with the researcher?though usually limited to stu- dents or clients?could apply here. Familiarity with the researcher has an attendant danger of blurring the relationship in the mind of the partici- pant, and may lead her to reveal material she might normally share with a friend or confidante but not with a person doing research (Liebow, 1993; Rosenthal, 1991). The question of monetary payment for participation is a compli- cated ethical dilemma in research with women who are homeless. Hono- raria for participants are a standard feature in much psychology re- search. Some feminist researchers have reinterpreted this tradition, and view money given to participants not merely as an honorarium, but as the compensation due to equal partners in the research enterprise (Lan- drine et al., 1995). In research with women who are homeless, though, financial rewards for participation may function as inducements (APA, 6.14; CPA, III.31), and are therefore ethically problematic. Thirty dollars may not seem like much money to a researcher, but such an amount (in Ontario, the equivalent of about a week?s worth of Personal Needs Al- lowance, the stipend paid to people staying in hostels) might make a woman who is homeless reluctant to refuse or withdraw participation. Some researchers working with people who are homeless have ad- dressed this concern by offering in-kind rewards such as a meal, food vouchers, or bus tokens instead of cash (Unger et al., 1997). Liebow (1993) offered needed services to participants such as driving or running errands; this represents a creative solution for researchers whose work is not funded, and it may place participants and researcher on more equal footing because the underlying message is, ?You are helping me by tell- ing me about yourself; now, what can I do to help you?? But the question of inducement is not the only problem with mone- tary or material compensation. Some researchers point out that partici- pating in traditional research rarely has direct benefits for marginalized populations, including people who are homeless (Rosenthal, 1991), while it often has great professional benefits for the researcher. Paying partici- pants is no substitute for ensuring that research has intrinsic benefits for participants and champions the interests of the community. Feminist and community researchers have demonstrated that homeless participants might benefit from research by making it a medium through which the participants can voice their concerns to policymakers or service provid- ers (Ralston, 1996; Rappaport, 1990). Others have designed projects that include the community?s use of the researcher?s skills and status to ad- vocate for individual participants or to support activism by people who are homeless (Rosenthal, 1991). Harm to participants is another ethical concern complicated by the context of homelessness. As noted in the previous example, research may be harmful to a community as a whole if it promotes negative ste- reotypes or misinformation. The research in the current scenario also has the potential to harm individual participants. The majority of homeless women are survivors of traumatic experiences including sexual and physical abuse (Breton & Bunston, 1992; D?Ercole & Streuning, 1990; Goodman, 1991; Ralston, 1996; Schram & Giovengo, 1991). Given their life circumstances of chronic stress and their limited financial resources, few have had the opportunity to heal after these experiences (Goodman, Saxe, & Harvey, 1991; Milburn & D?Ercole, 1991). Research examining stressors or depressed feelings might trigger women?s carefully managed feelings of rage, grief, or despair. The standard debriefing that follows most research is not sufficient, because homeless women might lack the social resources to access other types of support once debriefing is over. Some feminist researchers have advocated that counselling be made available to homeless women after they participate in research that ex- plores their feelings or experiences (Ralston, 1996); though as seen in the previous example, women may not avail themselves of this service un- less its hours, location, and service model are accessible and culturally appropriate. Feminist psychologists have warned of the potential for exploitation in data gathering (Grossman et al., 1997). For women who are homeless, the combined contexts of dependence on institutions, financial need, emotional vulnerability, and lack of privacy produce a situation that is ripe for exploitation by others, including researchers. Because homeless women are accustomed to accommodating themselves to situations not of their choosing in order to survive (Farge, 1989), it is vital that the re- searcher take full responsibility for identifying and changing aspects of the research process that might be experienced as coercive. Individual solutions might be found for each of the problems in the above scenario, such as offering nonmonetary forms of compensation. Such solutions would respond to the concerns of traditional research codes by protecting participants, but they would maintain a basic re- search structure in which the researcher holds the power and partici- pants are vulnerable. Feminist and community psychologists would argue that the best way to guard against exploitation in the research relationship is to de- velop a process that is inherently empowering for participants (Rappa- port, 1990; Serrano-Garcia, 1990). A minimum standard for empowering research is participant involvement in data collection (Serrano-Garcia, 1990); such involvement should include dialogue between researcher and participants on the effects of the research relationship and power differences (Bond, 1990). Ideally, it would include full collaboration with the community, so that data collection processes share power between participants and researcher and have some intrinsic benefit for partici- pants (Serrano-Garcia, 1990). An empowering approach to the research in the current scenario might begin with the researcher approaching service users as a commu- nity (and not only the agency?s administrator) to request permission to use the drop-in centre as a research site. The data collection process itself could be designed to respond to the needs of the individual participants, the community to which they belong, and the sponsoring agency. For example, the researcher could devote her time and resources to the crea- tion of a community development program for women using the agency. Together, using the researcher?s expertise and their own knowledge, a group of service users could identify personal and systemic stressors, examine the effects of these stressors, and explore coping strategies. They could identify goals for this work, which might include education on stress reduction techniques, creation of an ongoing interpersonal support network, or political activism. The social support of the group, and the focus on systemic stressors and active coping, could mitigate some of the emotional toll of the subject matter; the suggestion that the program is for the women themselves, rather than a favour they are doing for the volunteer, could help ensure that participation is voluntary; and partici- pation would have intrinsic benefits. A group structure could help redi- stribute power between participants and researcher, encourage the par- ticipants to contribute their knowledge and expertise, and explode the myth of the researcher as ?expert.? The research would actively respond to community needs and problems, rather than studying these from a disinterested stance. Ideally, the research would leave the community better able to address their own needs than they were before participat- ing. Analyzing Data: Interrogate Local Meanings A psychologist conducts a longitudinal study of 100 homeless women. Using assessment instruments and observations, she examines the partic- ipants? level of functioning, their patterns of transience, and their mental health over three years. The MMPI and other tests consistently indicate pathology in the whole sample. As well, the psychologist notices that most of the women, when presented with the opportunity to obtain bet- ter housing arrangements, generally fail to take the necessary steps to obtain such housing or end up back on the streets in a matter of months if they do obtain it. She writes a lengthy article identifying borderline, antisocial, and self-defeating personality disorders as major factors in the etiology of women?s homelessness. This hypothetical example reflects many problematic trends in cur- rent mainstream psychology research with women who are homeless. According to traditional ethics codes, during data analysis the researcher must avoid bias (CPA, III.10) and strive to represent the data as accurate- ly as possible (APA, 6.21; CPA, III.5). As in the previous example, a fe- minist or community psychology perspective on the context of women?s homelessness reframes the meaning of terms such as ?bias? and ?accura- cy,? leading to new ethical concerns about data analysis. In traditional research, avoidance of bias in data analysis is one as- pect of research ?objectivity.? For feminist and community psycholo- gists, who recognize that research is ?the subjective creation of a social phenomenon? (Serrano-Garcia, 1990, p. 182), bias represents the tenden- cy of research to reproduce unproblematically the point of view of the dominant group, often to the detriment of the marginalized community being studied. A common example of this tendency is the bias inherent in research using only etic data such as assessment instruments (Koegel, 1992; Landrine et al., 1995). Etic data are shaped by the researcher?s definitions and priorities, giving little information about how participants construct their own ex- perience. Research based solely on etic data is likely to yield a distorted picture of the community being studied. The risk increases if the re- searcher?s definitions are embedded in the dominant culture while those of participants are embedded in marginalized cultures, because the re- searcher?s questions and assumptions may be based on dominant stereo- types of the community. For instance, feminists and community psy- chologists (Herman, 1992; Koegel, 1992; Rosewater, 1985) have pointed out that the effects of trauma or homelessness, such as fearfulness or fa- tigue, may be seen by psychologists as personality disorders or signs of ?mental illness,? reflecting popular stereotypes that depict homeless women as crazy (and stereotypes that depict women in general as emo- tionally unstable). In fact, the emotions and behaviours labeled as ?dis- orders? by the researcher may represent, for participants, the natural outcomes of extreme stress, or an adaptive strategy for surviving trauma or homelessness. To ensure accuracy in the representation of data, a researcher work- ing from a positivist tradition might acknowledge that environmental factors may have confounded the results of some screening instruments, or might use instruments that have been re-normed on a homeless popu- lation to account for the effects these factors. For feminist and communi- ty psychology researchers, a more fundamental issue remains. As some theorists have pointed out, psychological research focused on identifying the individual problems of women who are homeless contributes to so- cietal victim blaming (Shinn & Weitzman, 1990), and generalizations drawn from this literature create a ?pathology? of the poor? (Montgom- ery, 1994, 36). Such research implicitly lays the blame on marginalized people for the difficulties they experience in living, instead of accounting for environmental factors that shape participants? experiences. Some feminist and community researchers have responded to the pejorative and pathologizing trend in the psychology literature on home- lessness by creating research focused on the strengths (Montgomery, 1994) and coping skills (Banyard, 1995) of women who are homeless, in which the definitions of ?strength? and ?coping? emerge from partici- pants? frames of reference rather than from standard indices. Instead of viewing the participants as the problem, such research constructs them as active subjects grappling with the problems presented by homeless- ness. The research question implicitly shifts from ?Why do homeless women fail to integrate into society?? to ?How do women who are homeless survive in a society that fails to serve the needs of all of its members?? In the current scenario, an analysis that views participants as active copers might begin from the understanding that for many women, homelessness itself is a product of their determination to escape and sur- vive violence in their homes (Breton & Bunston, 1992). Rappaport suggests that empowering research in community psy- chology should ?give voice to participants? definition of reality? (1990, 56) through the use of methods such as interviews, which enable partici- pants to influence the researcher. Feminist researchers, as well, point out that research should ?interrogate local meanings? (Fine, 1992, 13), using an analysis that emerges from the perspective of participants. Koegel (1992) applied these principles to the study of transience in homeless men, asking participants to explain why they often rejected what ap- peared to be ?better? housing. Participants explained that they remained in or returned to the hostels due to loneliness, or in response to problems such as the distance of housing from needed services. A comparable ap- proach in the study of transience in homeless women might yield similar explanations. This approach to research has a number of benefits. First, it empow- ers individual participants because it validates their ?expert? knowledge about their own needs and experiences. Also, it challenges dominant stereotypes of the community, depicting women who are homeless as competent copers rather than as sick, irresponsible, or lazy. Finally, and perhaps most importantly, its conclusions have concrete beneficial appli- cations to the development of appropriate housing for women who are homeless, whereas the information that homeless women can be diag- nosed with personality disorders contributes little to changing their situ- ation. Writing and Disseminating Research: Making Voices Heard For a course on qualitative research, I worked with a woman living with her son at a hostel. When I first approached her, I planned to interview her about her experiences with mental health interventions. As we worked together to plan and undertake the research, however, a much more complex analysis developed. We explored several issues including, (a) her experiences of oppression and how these affected her well-being; (b) the stigma of being labeled ?mentally ill? and how service providers treated her as a result of this label; and (c) her strategies for surviving and challenging the system. We developed this analysis over a two- month period of ?hanging out? that included formal taped interviews, informal discussions, and my participation in her meetings with institu- tional gatekeepers. This process had intrinsic benefits for both of us. My co-researcher taught me lessons I still refer to daily in my work with people who are homeless, and she states that she benefitted from my advocacy and validation. One of the course requirements was to write a final report on the process and results of the research. The few guidelines that ethics codes provide for the write-up stage?such as accuracy and honesty in reporting (APA, 1992, 6.21; CPA, 1991, III.5) or concealing the identity of participants (APA, 1992, 5.08; CPA, 1991, I.40)?do not adequately support feminist and community psychologists? aims of producing research that privileges participants? accounts of their experiences (Grossman et al., 1997; Rappaport, 1990), that honours the mutuality of the research relationship (Fine, 1992), and that responds to participants? needs (Fawcett, 1990). These were my chief concerns in writing the research report. First, I wanted to leave my co-researcher?s voice in its rightful place at the centre of the text and not marginalize it. Also, I wanted the writing to reflect the mutuality of our relationship, rather than turning my core- searcher into an object I was writing about. Finally, I intended the text to be useful to both myself and my co-researcher, just as the process had been. Bringing these feminist concerns into academic writing can be risky, as work that challenges the marginalizing tendencies of traditional research may itself be marginalized in psychology (Grossman et al., 1997). I decided to write the report to my co-researcher, in the first and second person rather than the third person. This approach required me to situate myself in the text, own my opinions, and abandon the authori- ty conferred by an anonymous stance, all crucial elements of feminist scholarship (Fine, 1992). This writing style honoured the relationship by keeping the report within our mutual dialogue, rather than placing it within an academic discourse that excluded the co-researcher. To keep my co-researcher?s story and her voice at the centre of the work, I organized the report around long quotes transcribed from the interviews, and based the analysis on the language and content of these quotes. In this process I was sensitive to the concerns Fine (1992) points to about the use of voice in research. I wished neither to exploit my co- researcher?s voice by choosing quotes to suit my analysis, nor to roman- ticize her voice by presenting it uncritically. The dialogical style of the report enabled me to present my own critiques and arguments alongside my co-researcher?s accounts, in keeping with the feminist and communi- ty psychology view of research as a pooling of resources between re- searcher and participant (Bond, 1990) in which both parties bring valua- ble contributions. Instead of presenting a monolithic analysis of the data, as a tradi- tional report would, the text of this report laid bare the dialectical process of the analysis, in which the co-authors? differing perspectives on the same ?facts? were situated in our different social contexts. As a re- sult, one topic in the paper was the tension between how the mental health system is experienced by a woman with a disability who is poor, Black, Jamaican-born, and a service user, and how it is experienced by a woman without a disability who is middle-class, white, Canadian-born, and a service provider. Finally, I wrote the report in the same language my co-researcher and I used when speaking, ensuring that it would be accessible to her. We each kept a copy of the report for our own uses. I was initially con- cerned that an academic report would not fulfil any of my co- researcher?s needs. When I asked her about this recently, she informed me that she refers to the report when she is feeling crazy and needs to be reminded of her strength. She also uses it in her interactions with service providers and family members, to demonstrate to them that an ?expert? believes she is worth listening to. It is important to note here that my professor?s support of a partici- patory feminist approach made it feasible for me to write the report as I did. Too often, psychologists are forced to choose between doing empo- wering work and meeting the demands of the academy (Grossman et al., 1997). While I was satisfied that the write-up honoured my relationship with my co-researcher, it failed to fulfil some ethics of participatory fe- minist scholarship. For example, writing the report took my time away from practical work we were doing to try to change my co-researcher?s situation. Her goal for the research was to produce a document that could be used to raise public awareness and rally support for herself. In the end, an academic report written in an alternative narrative style did not fit this bill. Other formats for the report?such as writing it as an ar- ticle for the local newspaper? might have been more useful in effect, if less radical in form. This disparity between our goals for the dissemina- tion of the research reflects the unequal distribution of power in our rela- tionship, a reality even in the most radical research (Bond, 1990). In feminist and community work with people who are margina- lized, ownership of the products of research should be returned to the community (Chavis et al., 1983). In research with women who are home- less, it is not adequate simply to hand over the research product to the participants. The researcher should also support the participants in ac- quiring the means to use the research for concrete changes in their own lives. This may mean sticking around after the research is over (that is, after the researcher?s goals have been met) to support participants? ac- tion. In my own case, it means planning future writing with my co- researcher as the circumstances of our lives permit, and remaining avail- able to her in the meantime as an advocate with an inside knowledge of the social services system and the persuasive power of a white, educated voice. Conclusion: Uses of Research Toward the Elimination of Oppression and Inequality In Ontario, a task force was created to study and make recommendations on homelessness in the province. After a process that included consulta- tions with ?stakeholders? and service providers in several municipali- ties, the task force released its findings. The task force report (Provincial Task Force on Homelessness, 1998) grouped ?the homeless? in Ontario according to the main ?causes? of their homelessness, including addic- tion, mental health problems, and failed immigration sponsorships. Grinding poverty?exacerbated by the current provincial government?s 21.6% cut to welfare rates ?was not mentioned as a cause of homeless- ness, nor was it considered a factor in people?s substance use, mental well-being, or relationship breakdowns. The message was clear: people become homeless because they are addicts, ?mentally ill? and untreated, or irresponsible immigrants; government policies on welfare and social housing have nothing to do with it. The recommendations accordingly proposed means for controlling these ?high-risk groups? and eliminating the problems ?the homeless? cause for businesses, public safety, and the social assistance system. Research is always political, whether it uncritically reflects the sta- tus quo or explicitly challenges it (Mulvey, 1988; Serrano-Garcia, 1990). This example from homelessness research in Ontario illustrates that re- search on homelessness is especially political because it can be used by governments to legitimize the very economic policies that cause home- lessness. Psychological research on homelessness lends itself to such mi- suse when its questions construct the ?subjects? as the problem; when its methods disempower participants and communities; when its analyses measure the complex experiences of marginalized people by compari- sons to dominant norms; when its writing silences or distorts the voices of participants; and when its dissemination delivers ownership of the research only into the hands of ?experts? and institutions. Ethical feminist and community psychology research with women who are homeless must begin with consideration of the personal, inter- personal, community, and political ramifications of research. Recogniz- ing that homeless women are vulnerable to harm as individuals and as a community because of the extreme victimization, stigmatization, and marginalization they endure, researchers? first challenge is to plan our research so that it does not exacerbate the suffering of individual partici- pants. We also must be aware of the potential for our research to rein- force stereotypes and contribute to discrimination against the ?commu- nity? of women who are homeless. Finally, we must grapple with the reality that economic and social exclusion of women who are homeless makes it unlikely that they will benefit in a meaningful way from tradi- tional psychological research. The scenarios discussed in this paper demonstrate that some ethical concerns may be addressed by building in safeguards while retaining a traditional research structure. But feminist and community psychology research must move beyond the ethical necessity of avoiding harm, and make an active commitment to the well-being of participants and their communities. This shift entails a transformation of the very structure of research. Serrano-Garcia (1990) proposes six steps that community research- ers should take in order to direct our efforts toward the elimination of oppression and inequality: (1) inform the community about the research and the research plan; (2) obtain ongoing, informed consent from the community via its representatives; (3) maintain confidentiality; (4) in- volve community members in the collection of data; (5) seek community authorization for publication of research; and (6) disseminate the data to the community. While these steps represent a collaborative model in which researchers retain most of the control, Serrano-Garcia asserts that communities included through these steps will begin to demand a more participatory approach in which the community shares control of the research process. Given the social context of marginalization of women who are homeless, and the individual sense of vulnerability or disentitlement that may result from internalized oppression, transformation of the research process is necessary to create an alternative context in which participants and communities are empowered to use research in their own interests. While it may not always be possible for researchers to follow all of Ser- rano-Garcia?s guidelines, this paper has proposed some methods to promote the well-being of women who are homeless and their communi- ties at each stage of the research process. At the outset, we can evaluate the feasibility of community partici- pation, and plan the research accordingly. If participation is not possible, we can attempt to predict possible consequences of the research ques- tions and process, and take responsibility for unforeseen consequences. The research plan can draw upon alternative sources of information in- cluding homeless women?s own accounts. Instead of treating quantita- tive research without community involvement as an end in itself, we can construct it as a foundation for the development of future participatory research. While gathering data, we can use our time and resources creatively, striving for a process that has intrinsic benefits for participants and func- tions as a site for community empowerment. When analyzing data, we can champion the point of view of participants, and represent women who are homeless as competent copers. When writing research, we can position ourselves as interested subjects rather than objective experts, maintain the integrity of participants? voices, and ensure that the lan- guage and structure of the writing is accessible and useful to partici- pants. When disseminating research, we can respect the community?s ownership of the research, and assist participants to acquire the means to use the research for their own benefit. Transforming the structure of research can be difficult and risky. Marginalized communities may have neither the skills nor the energy for the level of participation this type of research entails (Bond, 1990; Serra- no-Garcia, 1990); in such a case the researcher is challenged to fit the me- thods to the capacities of the community (Bond, 1990), or train the com- munity to employ the methods (Chavis et al., 1983). This problem is exacerbated when doing research in which the community is not directly represented (O?Neill, 1989), or with a group such as women who are homeless, who may not see themselves as a ?community? at all. As re- searchers trained in traditional methods, we too may lack the skills re- quired to mobilize community involvement (Serrano-Garcia, 1990). If the community does collaborate, we risk losing control of the re- search, particularly if our aspirations are different from those of the community (Chavis et al., 1983). The planning, process, and writing of such research may be more time- and resource-consuming than tradi- tional methods (Serrano-Garcia, 1990). Meanwhile, research that stays true to feminist and community psychology values may not be competi- tive for funding (Grossman et al., 1997). This is particularly true in re- search on homelessness, where governments and their granting agencies are far more likely to fund large-scale, quantitative, epidemiological re- search on the ?deficits? of people who are homeless than to fund re- search that empowers homeless women or examines political and eco- nomic policies responsible for homelessness (Shinn & Weitzman, 1990). This might place researchers in a difficult ethical position, forcing us to choose between our professional self-interest and doing research that is truly beneficial and meaningful. Luckily, taking the high road is not all about self-sacrifice. As psy- chologists doing ethical research with women who are homeless, we have the opportunity to participate in work that is more interesting, complex, challenging, meaningful, and valuable than most traditional research. To examine homelessness dispassionately through the eyes of dominance not only is an abdication of our ethical responsibility, but also impoverishes our work. Our efforts (as imperfect as these may be) to embody feminist and community psychology values in our research will enrich the field and contribute to the transformation of lives, including our own. This chapter is based on Paradis, E. (2000). Feminist and community psychology ethics in research with homeless women. American Journal of Community Psychology, 28(6), 839 ? 858. Emily Paradis is Research Manager at the Cities Centre, University of Toron- to, and an instructor in the Department of Adult Education and Counselling Psychology at Ontario Institute for Studies in Education, University of Toron- to. This chapter was written during her M.Ed. studies, which were funded by the Fonds pour la Formation de Chercheurs et l?Aide a la Recherche (FCAR) of the government of Qu?bec. References American Psychological Association. (1992). Ethical principles of psychologists and code of conduct. American Psychologist , 4 7(12), 1597? 1611. American Psychological Association. (2002). Ethical principles of psychologists and code of conduct. Washington, D.C.: American Psychological Association. http://www.apa.org/ethics/code/index.aspx. Banyard, V. (1995). ??Taking another route??: Daily survival narratives from mothers who are homeless. American Journal of Community Psychology , 23 (6), 871? 891. Baxter, S. (1991). Under the v iaduct : Homeless in beautiful B.C. Vancouver: New Star. Bond, M. (1990). Defining the relationship: Maximizing participation in an un- equal world. In P. Tolan, C. Keys, F. Chertok, & L. Jason, (Eds.), Researching community psychology (pp. 183? 187). Washington, DC: American Psychologi- cal Association. Breton, M., &Bunston, T. (1992). Physical and sexual violence in the lives of homeless women. Canadian Journal of Community Mental Health, 11 (1), 29? 44. Canadian Psychological Association. (1991). Canadian Code of Ethics for Psy- chologists. Canadian Psychological Association. (2000). Canadian code of ethics for psycholo- gists. Third Edition. Ottawa, ON: Canadian Psychological Association. http://www.cpa.ca/cpasite/userfiles/Documents/Canadian%20Code%20of% 20Ethics%20for%20Psycho.pdf Chavis, D., Stucky, P., & Wandersman, A. (1983). Returning basic research to the community: A relationship between scientist and citizen. American Psycho l- ogist, 38 (4), 424? 434. Condor, S. (1991). Sexism in psychological research: A brief note. Feminism and Psychology , 1 (1), 430? 434. D?Ercole, A., & Streuning, E. (1990). Victimization among homeless women: Im- plications for service delivery. Journal of Community Psychology , 18 (2), 141? 152. Farge, B. (1989). Homeless women and freedom of choice. Canadian Journal of Community Mental Health, 8 (1), 135? 145. Fawcett, S. (1990). Some emerging standards for community research and action: Aid from a behavioral perspective. In P. Tolan, C. Keys, F. Chertok,& L. Ja- son, (Eds.), Researching community psychology (pp. 64 ? 75). Washington, DC: American Psychological Association. Fine, M. (1992). Disrupti v e v oices: T he possibilities of feminist research. Ann Arbor. University of Michigan Press. Fine, M., & Gordon, S. M. (1991). Effacing the center and the margins: Life at the intersection of psychology and feminism. Feminism & Psychology , 1 (1), 19? 27. Goodman, L. (1991). The prevalence of abuse among homeless and housed poor mothers: A comparison study. American Journal of Orthopsychiatry , 6 1(4), 489? 500. Goodman, L., Saxe, L.,&Harvey, M. (1991). Homelessness as psychological trau- ma: Broadening perspectives. American Psychologist , 46 (11), 1219? 1225. Grossman, F., Gilbert, L., Genero, N., Hawes, S., Hyde, J., & Marecek, J. (1997). Feminist research: Practice and problems. In J. Worrell & N. Johnson (Eds.), S haping the future of feminist psychology : Education, research and practice (pp. 73? 91). Washington, DC: American Psychological Association. Herman, J. (1992). Trauma and recov ery. New York: Harper Collins. hooks, b. (1984). Feminist theory from margin to center. Boston: South End Press. Koegel, P. (1992). Through a different lens: An anthropological perspective on the homeless mentally ill. Cultur e, Medicine and Psychiatry , 1 6 (1), 1? 22. Landrine, H., Klonoff, E., & Brown-Collins, A. (1995). Cultural diversity and me- thodology in feminist psychology: Critique, proposal, empirical example. In H. Landrine (Ed.), Bringing cultural di v ersity to feminist psychology : T heory , re- search , practice (pp. 55? 75). Washington, DC: American Psychological Asso- ciation. Liebow, E. (1993). Tell them who I am : T he li v es of homeless women. New York: Free Press. Milburn, N., & D?Ercole, A. (1991). Homeless women: Moving toward a compre- hensive model. American Psychologist , 4 6(11), 1161? 1169. Montgomery, C. (1994). Swimming upstream: The strengths of women who sur- vive homelessness. Adv ances in Nursing Science , 1 6(3), 34 ? 45. Mulvey, A. (1988). Community psychology and feminism:Tensions and commo- nalities. Journal of Community Psychology , 16 , 70? 83. Oakley, A. (1981). Interviewing women: A contradiction in terms. In H. Roberts (Ed.), Doing feminist research (pp. 30 ? 61). London: Routledge and Keegan Paul. O?Neill, P. (1989). Responsible to whom? Responsible for what? Some ethical issues in community intervention. American Journal of Community Psychology , 1 7(3), 323? 341. Peplau, L. A., & Conrad, E. (1989). Beyond nonsexist research: The perils of fe- minist methods in psychology. Psychology of Women Quarterly , 1 3 , 379? 400. Provincial Task Force on Homelessness. (1998). Report of the Prov incial Task Force on Homelessness. Toronto, ON: Government of Ontario. Ralston, M. (1996). ?? Nobody wants to hear our truth??: Homeless women and theories of the welfare state. Westport, CT: Greenwood Press. Rappaport, J. (1990). Research methods and the social empowerment agenda. In P. Tolan, C. Keys, F. Chertok, & L. Jason (Eds.), Researching communit y psy- chology (51? 62). Washington, DC: American Psychological Association. Rosenthal, R. (1991). Straighter from the source: Alternative methods of research- ing homelessness. Urban Anthropology , 20 (2), 109? 123. Rosewater, L. (1985). Schizophrenic, borderline, or battered? In L. Rosewater & L. Walker (Eds.), Handbook of feminist therapy : Women?s issues in psychotherapy (pp. 215? 225). New York: Springer Publishing Company. Serrano-Garcia, I. (1990). Implementing research: Putting our values to work. In P. Tolan, C. Keys, F. Chertok,&L. Jason (Eds.), Researching community ps y- chology (pp. 171? 182). Washington, DC: American Psychological Associa- tion. Shinn, M. (1997). Family homelessness: State or trait? American Journal of Comm u- nity Psychology , 25 (6), 755? 769. Shinn, M., & Weitzman, B. (1990). Research on homelessness: An introduction. Journal of Social Issues , 46 (4), 1? 11. Schram, D., & Giovengo, M. (1991). Evaluation of Threshold: An independent living program for homeless adolescents. Journal of Adolescent Health, 12 (7), 567? 572. Unger, J., Kipke, M., Simon, T., Montgomery, S., & Johnson, C. (1997). Homeless youths and young adults in Los Angeles: Prevalence of mental health prob- lems and the relationship between mental health and substance use disord- ers. American Journal of Community Psychology , 25 (3), 371? 394. Walsh, R. (1989). Do research reports in mainstream feminist journals reflect fe- minist values? Psychology of Women Quarterly , 1 3, 433? 444. Chapter 9 .1 Burned by the System, Burned at the Stake: Poor, Homeless and Marginalized Women Speak Out F ORWAR D REP ORT T O THE UNITED NATI ONS C OMMITTEE ON EC ON OMIC , S OCIAL AND CULTURAL RIGHTS History of FORWARD FORWARD1 1 At the time of this report, FORWARD stood for Feminist Organization for Women?s Advancement, Rights and Dignity. Since 2008, the group?s full name is For Women?s Autonomy, Rights and Dignity. began in April 2005 as Claiming Our Rights, a human rights education program at Sistering, a drop-in centre for women who are homeless and marginally housed in Toronto. With the help of two edu- cator/facilitators, women began coming together every week to learn about human rights, analyze our experiences of having our rights vi- olated, and discuss ways to make change. In December 2005, after the planned program of workshops and group meetings had come to an end, women who had been working on Claiming Our Rights decided to con- tinue to meet each week to work together on actions to challenge pover- ty, homelessness, and violence against women. We chose the name FORWARD to express the optimism and determination that the group embodies. In 2006, FORWARD presented this report to the review of Canada by the United Nations Committee on Economic, Social and Cultural Rights (CESCR). Doreen Silversmith presented an oral submission to the Committee in Geneva. The CESCR Concluding Observations on Canada, released May 23, 2006, contain many strongly worded findings and rec- ommendations on women?s poverty and homelessness in Canada.2 Since that time, FORWARD has continued its self-advocacy work through a number of initiatives, including attendance at the June 2006 Grassroots Women?s International Academy and the World Urban Fo- rum; a submission to the UN Special Rapporteur on the Right to Ade- quate Housing when he conducted his fact-finding mission in Canada in October 2007; a report to the October 2008 review of Canada by the UN Committee for the Elimination of Discrimination Against Women (CEDAW), based on research conducted with women at twenty drop-in centres and Aboriginal organizations in the Toronto area; and a fall 2009 project to offer peer-led Women?s Health Circles at Toronto drop-in cen- tres. For projects such as these, FORWARD was recognized with an Ur- ban Health Award by the Wellesley Institute in 2008. Reporting on Human Rights Most of the reports prepared for the CESCR review of Canada rely on statistics and research to show how Canada?s federal and provincial pol- icies have led to violations of economic, social, and cultural rights. Our report is complementary to these, because it uses women?s own words to develop a picture of how rights violations unfold in the lives of the people most affected: migrant and Canadian-born women who are poor, homeless, racialized, Aboriginal, psychiatrized, older, disabled, queer, mothers, and survivors of violence. This report was based on discussion 2 United Nations Committee on Economic, Social and Cultural Rights. (2006, May). Concluding Observations on Canada. http://daccess-dds- ny.un.org/doc/UNDOC/GEN/G06/427/83/PDF/G0642783.pdf?OpenElement, UN document numbers E/C.12/CAN/CO/4 and E/C.12/CAN/CO/5. and analysis that emerged in the group during weekly meetings held from May 2005 through March 2006, as well as formal testimonies deli- vered by group members, and notes made specifically for this report at a series of meetings held in February and March 2006. In some cases these are supported by institutional documents, academic literature and media reports. In other cases, the reader will have to rely on what we know from our own experiences, because this knowledge has not yet found its way into the media or the academy. Our experiences show that social and economic rights are indivisi- ble from civil and political rights. Violations of social and economic rights make women more vulnerable to violations of our civil rights, and violations of civil rights contribute to our social and economic marginali- zation. Though the testimonies that ground this report are from a rela- tively small group of women, they should not be seen as representing unusual or anomalous experiences. In fact, the experiences we describe are common. The commonalities across women?s stories prove that Ca- nadian governments? policies and practices operate in similarly harmful ways in the lives of many poor women. We strongly urge the Committee on Economic, Social and Cultural Rights to consider the voices, knowledge, and experiences of poor wom- en when formulating its Concluding Observations on Canada. Our lives bear the scars of Canadian governments? flagrant disregard of the Cove- nant in the period covered by the review. The rights violations we have experienced demonstrate that an adequate standard of living and ade- quate housing are fundamental to all human rights. The poverty, home- lessness, and social exclusion we live with are contrary to the basic prin- ciples of human dignity and worth. Our demands, if acted upon in a meaningful way, would aid governments in Canada in fulfilling the true spirit of the Covenant and all human rights documents. Violations under the Covenant Articles 2 & 3: Non -Discrimination and Equal Enjoyment: Poor Women Targeted and Marginalized ?Women of a certain economic class are stereotyped as whores. This is used as an excuse for violence.? Poor women 3 Poor women experience social exclusion in Canada . Right now across Canada, poor women are excluded from social forums that influ- ence our own lives and society as a whole. Our voices, faces and know- ledge are not in evidence in the mainstream media, in education, or in government. Our needs and priorities are not considered in business, policy, or legislation. Our interests are not served by the systems of law don?t have equal enjoyment of economic, social, and cultural rights. Articles 2 and 3 of the Covenant guarantee non- discrimination in economic, social and cultural rights, and equal enjoy- ment of these rights by women and men. In Canada, this is not the case. Women?and especially women who are racialized, Aboriginal, dis- abled, psychiatrized, and migrants?are disproportionately represented among Canadians who experience poverty and severe housing problems including homelessness (CERA, 2002). We are disproportionately af- fected by cuts and eligibility restrictions in welfare, unemployment in- surance, and housing; and by legislation that mandates increased sur- veillance and coercion by social assistance, child protection, and psychiatry (Mosher, 2002). When women are marginalized economically and socially in this way, we become receptacles for mistreatment, degra- dation, and violence. And then, in a cycle that is truly vicious, the de- graded image of poor women is used to justify the very policies and laws that increase our poverty. 3 In this report, ?poor women? is to be understood as referring to low-income and homeless women who may be migrant or Canadian-born, racialized, Abori- ginal, psychiatrized, disabled, older, queer, mothers, and/or survivors of vi- olence. The rights violations, marginalization, and targeting described are on the basis of the interaction of income and gender with these other factors. enforcement, child protection, social assistance, health care, and mental health care and psychiatry. Abuse of poor women is pervasive. FORWARD members and oth- er women we know have endured countless incidents of discriminatory surveillance and intrusion by child protection and law enforcement agents, verbal and physical abuse by strangers on the street, sexual ha- rassment by landlords, mistreatment by medical and psychiatric profes- sionals, hostility from public transit drivers, exploitation by employers, arbitrary refusal of benefits by social assistance workers, and humiliation by shelter staff. The aggressors in these incidents are targeting women on the basis of gender, class, race, colour, disability, age, citizenship, lan- guage, sexual orientation, family status, and/or appearance of homeless- ness. We do not see these incidents as discrete, separate violations of our rights. Instead, we see them as part of an overall perception of poor women as convenient and appropriate receptacles for abuse by anyone in a position of power, however slight that power may be. The perception of poor women as inferior is reinforced by State policies and actors. This perception of poor women as inferior and de- serving of mistreatment has been reinforced by state policies, and by the actions and statements of state actors. In the period covered by this re- view, policies in social assistance, child protection, and mental health law in Ontario have come together to paint a picture of poor women as cheats, bad mothers, and dangerous.4 4 In 1995-96, the Ontario government cut welfare by 21.6%, toughened anti -fraud laws, increased fraud investigations, and introduced a ?snitch line? encouraging people to call and report on their neighbours and family members. Victims of these policies included Kimberley Rogers, a woman who died while under house arrest for a conviction of welfare fraud. She had been cut off welfare entirely and was eight months pregnant. Moreover, in 2000, the Ontario government intro- duced new child protection legislation that increased scrutiny of women who entered domestic violence or homeless shelters with their children. Workers in such settings were mandated to notify child protection agencies of any child whose mother was escaping abuse. Finally, in 2000, the Ontario government amended the Mental Health Act and the Health Care Consent Act to widen criteria for involuntary admission to psychiatric facilities, and to allow for the forced drugging of people living in the community. Though advocates argued that the This picture has been perpetuated by public statements by the highest officials of government.5 Social exclusion and stereotyping make poor women targets for violence. When poor women are portrayed as society?s garbage cans, that?s how we get treated. The effects are terrifying. FORWARD mem- bers have endured harassment, threats, attacks, and sexual assaults. These experiences are not unique to members of FORWARD, they are well documented across Canada. The Native Women?s Association of Canada and Amnesty International report that over 500 Aboriginal women?most of them impoverished and homeless?have disappeared or been murdered, and that the murders often were marked by extreme brutality (Amnesty International, 2005). In Vancouver?s Downtown East- side and in Edmonton, Aboriginal and non-Aboriginal women working as prostitutes have been disappearing for years. It took years for police to begin to investigate, and a site containing the remains of dozens of the women was discovered (see www.missingpeople.net). In Toronto, a se- ries of murders targeted transsexual and transgender women working as prostitutes (Commercial Sex Information Service, 1997). Poor wom- en, and especially women who are homeless, racialized, Aboriginal, mothers, and psychiatrized, have been the targets of a hate campaign in Ontario, and in many other jurisdictions across Canada as well. Poor women?s survival strategies are pathologized and crimin a- lized . Of course, poor women don?t just lie down and take the abuse and degradation. Unfortunately, though, our strategies for survival have been portrayed as crimes or pathologies. Working under the table to supplement starvation welfare rates is defined as fraud; refusing to take harmful psychiatric medications can get you locked up; working on sa- These examples represent the extreme end of a continuum of exploitation, cruelty, and violence that poor women face every day. existing test for involuntary admission?that a person present an immediate danger to herself or others ? was strong enough, the government justified the new legislation by suggesting that people with psychiatric diagnoses were un- predictably dangerous and violent. 5 For example, former Ontario Premier Mike Harris commented that he was ter- minating the $26 pe r month nutrition allowance for pregnant social assistance recipients because, ?We want to make sure they?re not spending it on beer.? fer, better-lit streets can get you arrested for soliciting. We liken this campaign to the witch hunts of medieval Europe, in which women seen as threats to the status quo were burned at the stake. Now, instead of being burned at the stake, poor women are ?burned? by the system. Conclusion. We demand that the federal, provincial and territorial gov- ernments acknowledge the connections between violence against women and social and economic exclusion. These governments must re-evaluate all policies and practices in social assistance, child protection, and mental health to ensure that they reflect the worth and dignity of poor women, instead of contributing to our degradation. Articles 6, 7 & 8: Work and Working Conditions: The Exploitation of Poor Women ?I was fighting for more shifts and they didn?t like that so they terminated me. They didn?t like older workers, because we tried to organize unions. So they fired us and hired 19-year-olds. And then, because I couldn?t get the extra shifts I needed, I was 24 hours short for E.I. (Employment Insur- ance). At another job, they paid me $400 per mont h for four hours per day, but I had to stay for a fifth hour that they called ? lunch,? even though I was the only one there and had to answer the phones. After my contract ended, I asked for my Record of Employment (ROE) so I could get E.I. They must have realized that $400 per month for 25 hours per week was only $4.00 per hour, much lower than mini mum wage, and they might get in trouble. So they put much fewer hours on my ROE, and I didn?t qualify for E.I.? ?When I was working they said I was too slow and I had to speed up. They forced me to work faster even though I couldn?t because of my dis- ability. My supervisor made fun of me, saying, ? My dog has thyroid prob- lems.? When I was in the hospital, my employer called the hospital and tried to get them to tell him confidential medical information about me. Because of my disability, I have trouble staying warm. I had to wear a lot of clothes because the heat wasn?t on in the building. They threatened to fire me because of the clothes. I still have threatening letters from the em- ployer.? ?At a factory where my aunt worked for years, they forced her to pay for E.S.L. (English as a second language) and other training as a condition of her employment, and promised to refund it, but never did. Also, she lost her seniority when she returned from this mandatory training. She had been there for 15 years! They did this to lots of the Latin American people who worked there. They made them sign contracts they couldn?t read.? Many poor women are excluded from labour law protection . These testimonies illustrate how poor women?especially women who are migrants, racialized, disabled, and older?are easy targets for exploi- tation by employers. Though labour laws about working conditions, sal- aries, and unions exist, they are insufficiently enforced. Often if we at- tempt to invoke these laws we could be fired. It is also important to note that there are workers in Canada who are not protected by these laws. For example, those of us who are on welfare in Ontario are forced to work in exchange for social assistance, but we are legally forbidden to form unions or to strike, and the social assistance we receive is far below minimum wage. Women admitted to Canada seasonal agricultural workers? program (SAWP) are not covered by the Employment Stan- dards Act and are barred from unionizing.6 The right to work is connected to the rights to social security, housing, and health . These testimonies also illustrate the interconnec- tions between the right to fair working conditions and other rights under the Covenant, including the rights to social security, to housing, and to health. Because our jobs are more likely to be temporary, part-time, ca- sual, home-based and piecework, we usually don?t qualify for Employ- ment Insurance when unemployed. Finally, many industries in Canada, including the garment industry, rely on the work of non-status migrants? mostly racialized women?who are not protected by any labour laws. 7 6 Since the time of writing, an Ontario court decision upheld the rights of farm- workers, including migrant workers, to organize. See Fraser v. Ontario (Attorney General), 2008 ONCA 760, viewed February 4, 2010, at http://www.ontariocourts.on.ca/decisions/2008/november/2008ONCA0760.pdf. The case is currently before the Supreme Court. For more information, see http://www.justicia4migrantworkers.org/supremecourt.html. Many FORWARD members have never been able to collect E.I. Meanwhile, when we lack access to social 7 In 1994, changes in eligibility requirements for federal unemployment insurance decreased the percentage of eligible unemployed workers to 36% from 74% in 1989. assistance, we must accept any work we can find in order to survive, even work that is exploitative or dangerous. Women?s low wages make adequate housing unaffordable, and leave us vulnerable to exploitation by landlords. We are forced to seek housing in inappropriate places, such as rooming houses that are well- known to be dangerous for women. Low wages and insecure employ- ment also mean that we are unable to leave housing where we are expe- riencing racial or sexual harassment or abuse. And even when employed, we may not be able to afford housing at all. Many FORWARD members have held jobs while living in homeless shelters. Finally, the right to safe and healthy working conditions interacts with the right to health. For women with disabilities, requests for ac- commodation of a disability can lead to harassment or dismissal. In or- der to maintain our employment, women with disabilities and chronic illnesses must often subject ourselves to working conditions that are harmful to our health. Conclusion: We demand that all workers in Canada, including migrant women and women receiving social assistance, be protected by all labour stan- dards. We demand stronger mechanisms and better enforcement of these stan- dards by governments. Article 9: The Right to Social Security: No Security for Poor & Homeless Women ?I am sixty-three years old. I came to Canada from Italy in 1966, when my son was two years old. I never went to school, so I had to learn from my kids. I never got a job. I raised my kids whole and healthy. I am living in a house that my husband left me when he passed away. I was 59, my hus- band was 68, when he died. My daughter was still only a child. We got only $400 a month. My friend told me not to bother applying for welfare because I owned a house. ?I have a big problem that the government doesn?t give me financial help because I?m not 65. They only give me $451 a month widow?s pension. It started five years ago at $400 and went up slowly to $451. There is no money in the bank, because the funeral took all the money. The $451 is not enough to take care of th e house. I have to pay property tax six times a year?three months in a row, then another three months in a row. Last year it was $409 (each month). In July it went up to $456. In the months that I pay property tax, sometimes my son gives me the bal- ance ? $5. Hydro is too much ? $ 127 a month sometimes. My house in- surance is $762 for the year. When I?m 65 , I can apply for another $400 [per month]. Then I will be rich! I?ll be able to pay everything. ?When my husband died my [ youngest] daughter was only 13. The school board gave her bus tickets to go to school. My husband?s brother- in-law would bring food every month because they have a restaurant. For clothes sometimes I get from [ a women?s drop-in centre]. ?My [oldest ] daughter from America helps me out for water, hydro, etc. because she doesn?t want to lose that house, for respect for my hus- band, for the blood he put into that house. My husband died for that house. Too much work. He worked in construction. Because my daughter from America helps me I can stay above water. Without her I would sink.? Social security rates across Canada are too low . Many NGO reports will undoubtedly describe the inadequacy of social assistance rates across Canada to cover the basic necessities of life such as food and hous- ing. FORWARD members know first-hand about the inadequacy of wel- fare and disability, and we?ve experienced homelessness and hunger as a direct result of the low rates of social assistance in Ontario. Many poor women have no access to social security programs . Poor women, and especially women who have been homeless, fall into gaps between social security programs. In the period covered by this review, such gaps in Canada have grown wider. In the past decade, so- cial security in Canada has transformed from a safety net of universal entitlements to a loose patchwork of programs whose eligibility criteria often exclude those most in need. The testimony above provides an ex- cellent example. The witness is not old enough to qualify for old age benefits, yet the widow?s pension to which she is entitled does not ac- count for the fact that she is caring for a minor child. Her lifetime full- time work of childrearing is not considered eligible for a pension or in- come replacement of any kind. As an older woman with no primary education, very little English, and no experience in the workforce, she is not employable. If she were to apply for welfare, she would have to re- port all the family support on which she relies to survive, or face being charged with fraud. The financial value of this support and of her wi- dow?s pension would be deducted from her welfare, leaving her with nothing. Meanwhile, ironically, her greatest expense is taxes, paid to a government that fails to provide for her. Women who are homeless cannot access social assistance. Another gap that is of great concern to FORWARD members and others who have experienced homelessness is the requirement that a person have a permanent address in order to begin receiving social assistance benefits in Ontario. This ridiculous rule also exists in many other provinces and territories. Women who are being released from hospitals and jails with no fixed address, and women who are living on the streets, in shelters, or in places not intended for human habitation, have to get a landlord to sign a ?promise to rent? form in order to begin receiving benefits. It is almost impossible to get this letter signed without money to offer as a down payment. Once again, women are made particularly vulnerable by this requirement because of the sexual exploitation that may be the price of the promise to rent. In addition, this requirement alerts potential lan- dlords to the fact that women are currently homeless and applying for welfare, two more strikes against us in competing for tenancy, along with discrimination we may face on the basis of race, citizenship, disabil- ity, language, family status, and sexual orientation. Also, if a woman?s identification has been lost or stolen while homeless, or if it was not re- turned upon release by a jail or hospital, she has to replace it before ap- plying for welfare, a process which costs money and takes weeks. Final- ly, many women?especially mothers, who risk the apprehension of our children if we go into a shelter?keep our homelessness hidden by stay- ing with friends or relatives. When the person providing a woman with temporary shelter is also in receipt of social assistance, it may jeopardize their own benefits if the woman uses their address to begin her claim. Poor women often do not qualif y for Employment Insurance . As noted above, another issue that has affected many members of FORWARD is lack of access to Employment Insurance (EI). Most poor women?s jobs do not provide the ?insurable hours? required to qualify for EI, even though insurance premiums have been deducted from our pay cheques. The massive annual surplus in this program is in part a result of the ineligibility of workers to receive benefits after paying pre- miums.8 The application process for Ontario disability benefits is d e- signed to exclude . Denial by process in applying for Ontario disability benefits is another concern (Income Security Advocacy Centre, ___ _) . It is very difficult to navigate this complex and lengthy process while liv- ing on an income that is about half of what we would receive on disabili- ty benefits. As a result we are forced to subsist with inadequate food and housing and under conditions of stress that may exacerbate our physical or mental health conditions. Routine denial at the first stage of applica- tion means that almost all applicants must also go through an appeals process in order to be successful. Because the adjudication process is not transparent, a woman might receive no information about why her ap- plication was rejected or what might be required in order to strengthen her claim. Many women with disabilities?especially women who are homeless, who have mental health problems and/or who do not read or speak English?are unable to complete the process within the required timelines and are never able to access the supports to which we should be entitled. The existence of this surplus is also in direct contradiction of Article 2, which commits governments to use the maximum available resources for the realization of social and economic rights. Conclusion. We demand that governments in Canada ensure that all so- cial security programs are adequate, available and accessible to all women who need them, especially women who are poor, homeless, disabled, psychiatrized, and migrants. 8 The 2004 report of the Auditor General of Canada to the Standing Committee on Human Resources, Skills Development, Social Development and the Status of Persons with Disabilities, states: ?I draw Parliament's attention to the surplus in the Employment Insurance Account. This is the fifth year that we have done so. The surplus reached $44 bi llion by March 2003 and continues to grow.? Viewed March 19, 2006, at http://www.oag- bvg.gc.ca/domino/other.nsf/html/04hr01_e.html . Article 10: Protection of the Family : Poor Mothers Targeted by Child Protection Agencies ?I came to North America in the early 1980s from [ South America]. One of my worst moments in Canada was when there was a Christmas party at one of the neighbours?. I was coming in with two of the children and the police were waiting there. They grabbed me by the hair and took my six- month-old grandson. They lied about all of this. The social worker was going so fast she fell and dropped my grandson down the stairs. All through Christmas we didn?t know the fate of my grandson. No-one told us if he was dead or alive. It was a real kidnapping. ?I requested a five-bedroom [ subsidized housing unit] so that I could get guardianship of my two grandchildren, but was denied so I couldn?t qualify to be their guardian. CAS [ Children?s Aid Society] has come into my apartment many times and taken my papers, journals and valuables. The CAS workers have told me that they will put me on the street and make me into a bag lady?that I deserve the worst. The allega- tions of CAS were that we were full-time prostitutes moonlighting as mothers?that we weren?t real mothers. It has been eleven years of physi- cal, emotional and spiritual torture. They have destroyed my whole fami- ly, starting with me and ending with my grandchildren. ?I am in desperate need of a five-bedroom place because I want to pull what?s left of my family together. My beloved aunt, who raised me, had a heart attack and died when she heard of this situation. I want to bring back my grandchildren. This December it will be two years since I have seen my grandchildren.? ?When I was 21 I got raped, and got pregnant with my daughter. I hid it from my family. I knew the guy. We were partying. I said no. It was very frightening. But I blamed myself, told myself that I shouldn?t have been drinking. So I came to Toronto. I took a bus. I don?t know why I chose To- ronto. I didn?t know anybody here. I was on the streets in Toronto. I didn?t have anybody or know anything. ?So then I?m tired cold hungry, walking up Yonge St., and the la- bour pains started. I was very scared. Somehow I found Mount Sinai hos- pital and the baby was born after 12 hours. The only child I ever had was a daughter. CAS took my child as soon as she was born. I was all alone. So of course I had postpartum depression, and other issues to deal with such as defending myself on the street against attempted physical and sexual assault, and hunger. It is my experience that it is really demoraliz- ing to the spirit when you are homeless, faced with the knowledge that I am walking on land that my ancestors had walked on since time imme- morial. ?Most of [ that year] I was in and out of shelters?it was packed? one night there were six of us on one bunk bed. So I started drinking again, and got in trouble?not with police, just with fighting with men. I woke up one morning with blood splatters all over me. I had no recollec- tion of what had happened. I was in a blackout. ?They put me in Queen Street (a psychiatric facility) because they said I was a ? drunken crazy Indian.? A nurse there helped me. She said, ? Why don?t you go to this program in Beaverton?? It was run by (a wom- en?s shelter). It was the first time in that vicious cycle that I actually lis- tened to somebody. I listened to myself. If you don?t listen to yourself, you?re dead. So I got a drive up to Beaverton. I felt very connected there. It was still Mother Earth, even though it wasn?t First Nations territory. I stopped drinking there. ?The whole time I was fighting for custody of my daughter. On Jar- vis Street [Family Court ] I lost custody permanently because of my drink- ing and because I was still in recovery. They didn?t give me a chance. ?When my child was being taken away, I had a flashback of an old newspaper photo I had seen in a review in 1979 of children being taken away to the residential schools. To lose your child does something to your soul.? Poor women are targeted, not supported, by child protection . Many of the women in FORWARD can testify to the irreparable harm our families have experienced from child protection agencies. These agencies are supposed to investigate any allegations of harm to children, but in practice, mothers who are Aboriginal, poor, homeless, racialized, migrants, young, and psychiatrized are vastly overrepresented among those involved with CAS (Young Mothers In/From Care, 2001). Child protection agencies are mandated to take whatever action is needed to protect and support a child at risk, but the actions they take rarely in- clude practical support measures to provide the necessities of life, im- prove the family?s living conditions, and support mothers in being good parents. Instead of protecting the child by supporting the family unit, child protection agencies put us under a microscope and make it even harder for us to take good care of our kids. To get our kids back, we are required to meet conditions that are almost impossible without practical and moral support. Other research shows that mothers in British Colum- bia have the same problems as we have had (Kellington, 2002). The right to protection of the family is related to the right to ad e- quate housing . As the testimonies demonstrate, the right to adequate housing has a complex relationship with the right to protection of the family. Women who are homeless?especially young women?are at very high risk of having our children apprehended at birth. Though the testimony above is of events that occurred decades ago, current child protection practice with young homeless mothers remains the same: more often than not, their babies are apprehended at birth, they are left to cope with the grief and loss while still homeless, and they lose custody permanently because they lack the housing and supports to parent. As also described above, inadequate housing also often prevents family members from obtaining guardianship so that the child can remain in the family. Housing is a factor in one out of five cases of child apprehension in Ontario (Chau et al., 2001). Meanwhile, poor women who lose custody of children also lose access to the social assistance benefits and social housing units associated with the child. It is even more difficult to meet stringent conditions for reunification (which may include an apartment of an appropriate size) while on a dramatically reduced income. Poor women are also at a disadvantage in attempting to fulfill conditions be- cause we cannot afford the time off work, or the transit fare, for atten- dance at mandated counselling sessions, legal appointments, and child- parent visits. The child protection system discriminates against poor women . Discrimination in child protection is also a big problem, especially for women who are homeless, Aboriginal, and psychiatrized. Many young women who are homeless have been in State care, and a young woman who is well-known to CAS is much more likely to have her own child apprehended (Young Mothers In/From Care, 2001). The well- documented mass abduction of Aboriginal children by the Canadian state (also known as the ?60?s scoop?) placed many women who are now in their 30s and 40s into State foster care or non-Aboriginal adoptive families (Aboriginal and Indigenous Social Work, n.d.). Many of these women?especially those who were placed in foster care or whose adop- tion arrangements broke down?now find themselves under increased child protection surveillance as parents. Poor women who are psychia- trized are presumed to be unfit parents due to our psychiatric diagnosis. The conditions for reunification with our children may include taking psychiatric drugs that make us tired, depressed, and unable to function. Why should CAS be able to control what we do with our own bodies? The child protection system does not serve poor children well. Fi- nally, those of us who were in care as kids know that the child protection system also did not serve our needs as poor children. For example, why did the State place us in care due to abuse by a household member, in- stead of removing the abusive person from our home? When in care, some of us experienced abuse, a lack of consideration of our needs and culture, and a profound sense of alienation. These negative impacts of State care have contributed to our emotional struggles as adults. Conclusion. We demand that the money spent maintaining children in state care be redirected to providing supports for poor families. Poor mothers need access to adequate and affordable housing, nutritious food, respite care and voluntary ongoing counselling. The punitive and degrading nature of the cur- rent child protection system does nothing to improve our parenting. Article 11(1): The Right to Adequate Housing : Not Every Dwelling Is a Home ?I was born in [ South America], and I came here many years ago. I have been experiencing housing problems pretty much since we arrived here. Pretty much we were always living in shared accommodations to make ends meet. My cousins, everybody, they said, ? You?ll have enough power, now that you?re going to Canada.? And actually, I couldn?t confide in them. When I wrote them letters, I couldn?t tell them, ?I am homeless.? It?s like an insult to their dreams or something. I didn?t want to crush their expectations. ?Anyway, so my family, they never knew until I landed in a shelter here in Canada. I lived in a shelter for four years minus a day. My first shelter was at [a Catholic agency ] and they refused to believe I was preg- nant. They said, ? Yeah, you look a little bit pregnant.? The workers humi- liated me. That was one of the worst things that happened to me?that they refused to believe that I was pregnant. They were saying that I was pretending to be pregnant so that I could get some compassion, you know people feeling sorry for me, as though I am somebody who likes to mooch for attention or something like that. ?They used to put my food in the garbage. The food was horrible. And I was vomiting and feeling more sick, because the baby needed food and so did I! I was six months pr egnant, and the doctor told me ? I?m scared about your pregnancy.? Rather than gaining weight, I was losing weight! So I used to go and eat at ( [my aunt?s] place. And then when I came back, they knew that I was full. They used to shout at me and tell me, ? Oh you must have got a rich pimp, that?s why you got fed.? So they told me, ? You have to get out of the shelter. If you have somebody, you have money, get the hell out of here.? They put me outside. It was winter- time, it was January, and they put me outside to sit on the floor. They said, ?You have to tell us the truth!? So many terrible things those people did to me in the shelter.? ?I?ve been thrown out of my home after 30 years, because the house got sold. There were some promises made by the new landlord that I could stay as long as I wanted?.anyways, long story short, I?m faced with evi c- tion? I just find that it?s incredible, you know, to think that after 30 years all you have is two months to look for a place. It?s outrageous. Especially being in a position of financial constraints when you?re looking at now apartments being like bachelors $700 and up [ from $500 rent on the cu r- rent place] . ?So of course I went to fight it, to the [ Ontario Rental Housing] Tri- bunal. So, anyway, I?m just talking about how cruel the system is. Section 162 of the Tenant Protection Act, 1997, is the piece that I?d like to see changed. It?s one thing for a landlord to say, ? I?m moving in? or ? My fami- ly is moving in? if that is legitimate. But it?s another if (they are just claim- ing to move in so that they can evict the tenant). Nobody does any inves- tigating, for one. And I really don?t think that this clause, this piece, is meant for somebody who is a real estate agent, who owns five properties. Who, you know, has got a huge house in High Park [a high-income area of Toronto]. I mean, it?s just, I don?t know. I just don?t think it?s right. ?I found that there?s a certain amount of discrimination. And that has to do with the class, and your source of income. Now, if I had money, I would have gotten myself a Bay St. lawyer. I would have got to hire a private eye and have done some investigating. Right? And you know what, I may have won in the first place, because I would have had money to be able to prove, you know? And I myself, if I were investigating them, I could be set up for stalking. But it?s all really about the income.? ?I was working, but getting no pay increases, so I couldn?t afford my apartment. Then I moved to a rooming house. But these people were not related to me, not part of my family, not anyone I knew. And if I played East Indian music people were offended?they said there were no East Indian people living there, so I should go and live with my own people. They said that because I was working, I had to pay more for heat and hy- dro. When I was in the kitchen, people would come in and accuse me of things I hadn?t done. Other people said the landlord sexually harassed them as well. So there were many problems with this house. The rent was too high, there was racism and discrimination, it was untidy, there was too much noise. I complained about these things, but I was told that I had no priority, because I didn?t own the house. The landlord ignored my complaints, and said, ? You should go live with people of East Indian ori- gin.? I couldn?t live there. They were raising the rent too much. So I ap- plied to the government for housing. But I didn?t have priority.? Poor women across Canada live in conditions of extreme housing inadequacy . These testimonies illustrate the conditions of extreme hous- ing inadequacy?including hidden homelessness, absolute homeless- ness, and insecurity of tenure?in which poor women across Canada are forced to live. Housing for poor women in Canada is characterized by enforced dependency that is in contradiction to human dignity and worth, the cornerstones of all human rights. Whether poor women live in shelters, private market apartments, social housing, with family or friends, in rooming houses or in group homes, we often find ourselves at the mercy of other people?staff, landlords, cohabitants?who treat us in an exploitative, abusive or degrading manner. Problems with governments? responses to homelessness: A shelter is not a home. In the period covered by this review, Canada?s principal response to homelessness has been through the provision of funding for shelters and other services for people experiencing absolute homeless- ness.9 9 The major federal homelessness program, the Supporting Communities Part- nership Initiative (SCPI), targets funds to ten Canadian municipalities with high rates of homelessness. SCPI funding is available for services for homeless people, including services to help people move into housing, however this funding does not increase the availability of affordable or supportive housing, which most Though shelters are a vital resource for women with no place to go, and conditions in some shelters have improved, those of us who have stayed there know that shelters are not housing: the food is sometimes terrible, there?s noise all night, things get stolen, and there?s no privacy. Sometimes women are treated very badly and humiliated by staff. Wom- en can?t stay for very long, and we have to leave during the day. In some shelters the conditions are bad, for example it?s cold, it?s dirty, or there are not enough showers or toilets or no doors on the toilets. Women can be locked out or ?barred? with no place to go. Shelter life is especially bad for pregnancy or if you have a health condition. There is no security of tenure for poo r women. As illustrated by the second testimony, insecurity of tenure haunts the tenancies of most poor women in Canada. In Ontario, legislation passed in 1997 allows landlords to raise rents on vacant units an unlimited amount, providing an incentive to evict tenants, especially those considered ?undesirable? (on the basis of race, source of income, disability, citizenship, gender, age, and family status) and those paying relatively low rents. The same legislation made the eviction process much easier and allowed for land- lords to evict tenants without cause if the landlord planned to move into the unit. Tenants may be evicted for non-payment of rent, even after pay- ing back all the rent they owe. These changes have led to steadily in- creasing eviction rates in Ontario.10 homeless women require in order to be rehoused. During the period of this re- view, there was no federal funding for new affordable housing (except Aborigin- al housing) until the 2001 Federal-Provincial Housing Framework Agreement. However, due to cost-sharing requirements there has been little up-take of these funds in most provinces. For example, in Ontario, only 63 new homes have been built as a result of this funding (National Housing and Homelessness Network 2005 Report Card, viewed 20 March 2006 at http://www.tdrc.net/05-NHHN- Housing%20Report%20Card.pdf ). In contrast, the waiting list for subsidized housing in Toronto alone stands at about 70,000 households. Some landlords now use the easy-to- 10 For example, in 2005 eviction orders issued by the Ontario Rental Housing Tribunal increased by 9% from the previous year, to 64 864. (see Wellesley Cen- tral Pre-Budget working paper, viewed March 20, 2006 at http://www.wellesleycentral.com/ITGUpload/doc/83/ontario_prebudget_backgr ounder_2006.pdf ). More than half of the households evicted lose their housing by default, because they fail to respond to the Tribunal within the five-day time obtain eviction notices as a form of harassment, so that even if the evic- tion is not carried out, tenants live in a state of uncertainty and fear. Eviction affects mental and physical health . Security of tenure in- teracts with the right to health, especially for women who have disabili- ties and who have been psychiatrized. Threats of eviction, the search for housing, and the process of moving are all extreme stressors that desta- bilize women?s physical, mental and emotional well-being, and may in- terfere with our ability to work or parent. For example, one FORWARD member who was evicted felt so anxious about it that she needed medi- cation, but the medication also made her sleepy, depressed, and less productive. Eviction may also affect women?s physical health when we are forced to move into accommodations that are more expensive, more crowded, more dangerous, in worse condition, and unhealthier than the place from which we are evicted. This is especially the case for women who are forced to move after living for a long time in a rent-controlled apartment. The woman who testified about her eviction above now has no food money after the first week of the month, once her new higher rent and utilities are paid. Social housing is vital, but it is often substandard, dangerous , and degrading. Housing advocates say the government should build more subsidized housing and we agree, but those of us who have lived there know that there are many problems with social housing projects. In some developments, we?re made to feel like a ?welfare case? if we go to the management office with a complaint. Many women in social housing are treated with disrespect and racism by our neighbours and by the man- agement. We have experienced violence and intimidation, and the police and security guards refuse to take our situations seriously. Because of the stealing, drugs, violence, and weapons in our communities, we live in fear. The buildings are old and falling apart and the government says there?s no money to fix them. Now that the management of public hous- limit due to a misunderstanding of the process. This procedure continues in spite of the current government?s 2003 promise to create new legislation, and in spite of a letter of concern from the Ontario ombudsperson?s office directing the gov- ernment to act without delay (see Toronto Star article, February 15, 2004, viewed March 20, 2006 at http://www.ontariotenants.ca/articles/2004/ts-04b15.phtml). ing has been taken over by private management companies, we believe that they are trying to get rid of subsidized tenants in some buildings and replace them with tenants who will pay market rent. Even in subsi- dized housing, we don?t have security of tenure. We worry about being evicted if we draw attention to ourselves through complaints, and some of us have had our privacy compromised by video surveillance and by arbitrary extra checks on our income. The homelessness of many poor women is hidden . Hidden home- lessness?which often takes the form of staying with family or friends? is a reality for increasing numbers of Canadian women. It is particularly prevalent in rural areas where few services exist, in First Nations and Northern communities where there is a grave shortage of housing, and among migrant women whose families often must live in conditions of extreme overcrowding due to extremely low wages and high rents in urban centres.11 Rooming houses are not adequate housing . As shown by the third piece of testimony above, living in rooming houses is very hard for women and it is often the only housing a single woman can afford, even if we are working. In rooming houses we face sexual and racial harass- ment, homophobia, and racist comments about our music or the smell of our food. Like in shelters, sharing space with strangers compromises women?s privacy and increases the risk of theft and violence. Rooming house owners may exploit their power by controlling residents? sche- dules or guests, and setting other arbitrary and invasive rules. For exam- ple, one woman was told by the owner to use a bedpan at night, instead of the toilet. And because there is no regulation of rents on rooms, own- ers may charge widely varying amounts even among tenants in the same house, getting as much as they can from each resident. Besides placing people in overcrowded conditions, hid- den homelessness has other consequences, including stresses on family relationships that might otherwise provide vital sources of support. 11 In First Nations communities, housing shortages might be exacerbated when private companies mine, log, or build on land over which title is in dispute, as is now happening at one FORWARD member?s community of Six Nations in On- tario, for example. See http://ocap.ca/1stnations/sixnations/courtorder. Poor women are abused and exploited in group homes and care homes. Another form of housing that is well-known among women who are homeless but less visible to the general public is group homes?also known as lodging homes and care homes?for people with disabilities, psychiatric survivors, youth in state care, and older people. Though housing advocates rightly call for additional supportive housing, many women who are homeless would never consider living in a group care setting. For those of us who know about group homes through our own experiences or through our work, the reasons are not difficult to under- stand. In Ontario, privately-run lodging homes?a type of housing women are often referred to if they are moving off the street?receive much more money to house poor women than women would receive themselves if they were on a disability. Homes are paid a per diem top- up per resident (not per room) in addition to the housing and board por- tion of residents? disability or pension cheques. Residents may be re- quired to share rooms, with up to four persons per room. We have known of many homes that require that residents appoint the home as trustee, and disburse residents? monthly incomes as they see fit, often deducting large sums for items such as soft drinks, cigarettes, and sham- poo ?purchased? from the house. Staff who are often underpaid and un- dertrained oversee residents of widely varying needs, from frail elderly to people with developmental disabilities to psychiatric survivors. Many homes deliberately overmedicate residents in order to maintain com- pliance and quiet. There are documented reports of abuse, deaths, and poor nutrition in Ontario group homes (Toronto Star, n.d., 2003; Hedley et al., 1994). Conclusion. We demand that every person in Canada have access to a home that is adequate, affordable, secure and dignified. Governments in Canada must ensure that all forms of state-provided housing, including social housing and group homes, meet the highest standards of dignity and safety. Article 12: The Right to Mental Health : Poor Women Harmed By the Psychiatric System ?My first time in the hospital was in my mid-twenties. The first time, I was in hospital for three months. The second time was one and a half years. I was misdiagnosed with chronic paranoid schizophrenia, and that has upset me very much. The doctors prescribed a lot of medications and I didn?t know what they were for. They violated my rights by things they gave me that they weren?t supposed to because of the wrong diagnosis. Later when I went to other doctors they diagnosed other illnesses. ?I want them to be more sure of diagnoses before giving so much medication. Certain things they gave me, they weren?t supposed to, be- cause of the wrong diagnosis. I?m very upset about the wrong diagnosis on my file and about the wrong medications that were given to me. I was taking calcium pills, thyroid medication, haldol, congentil, chlorproma- zine, laxapine. I kept complaining about the medications and refusing to take them. So they sent me to my family doctor and he gave it to me in an injection. I took it for one and a half years. I was all lost in my mind? someone would be talking to me and I would be somewhere else. Right now I?m not taking any medication for that diagnosis. I have been ex- amined by many doctors and nurses and they all say I don?t have it. Now they say what I have is anxiety and things caused by my thyroid. ?I was forced to stay at the hospital and sometimes police came to get me and took me there. One of my arms was damaged in the hospital. My calcium was very low. They held my arm very tight to force me to take a drug?to give me an injection. One and a half years later I was working and was in extreme pain with my arm. I didn?t know what was wrong. I went to another hospital and they said it had been broken. I still experience pain in my arm all the time. In the 90s I couldn?t raise my arm. ?I really desire that this diagnosis be removed from my file. I?m con- cerned that my files contain information that is not true. It?s on my mind all the time to get this diagnosis removed from my records. My doctor says there?s nothing that can be done. If I die it will still be on my record. I have been to medical records four times and they say it can never be re- moved because it is on microfilm. I want to get the correct things on my files. I called back an officer of the hospital and he said there is nothing that can be done. I don?t know who can help me get it off.? ?People should not be under arrest for living their lives. I had my own life until I was systemized. I was persecuted and hauled away in handcuffs. The [psychiatric] system is the one who threw us out on the street in the first place. We?re told, ?Co nform to society or go in a mental institution.? I feel raped by the psychiatrists. They raped my brain. They took my intel- ligence and tried to zap it out of my head.? Poverty affects women?s physical and mental health . Poor women in Canada face an array of health problems related to social and econom- ic marginalization. The physical health effects of poverty, homelessness, and racism have been the subject of extensive documentation. These problems are also of great concern to members of FORWARD; however, this report will focus on an area that is unlikely to receive much attention in other reports: psychiatric treatment as a violation of the right to the highest attainable standard of physical and mental health. Poor women?s human rights are violated in the psychiatric sy s- tem. As the testimonies above illustrate, psychiatric diagnosis, incarcera- tion, and medication have a powerful negative impact on the lives of poor women. FORWARD members have experienced abuse and degra- dation in the psychiatric system, including being physically restrained, forcibly injected, and sexually assaulted. The medications we have been prescribed by psychiatrists or family doctors often make us feel even worse. We do not receive the full information we need to make an in- formed decision about medication, and often when we complain to our doctors about the unwanted effects of medication, they don?t listen. If we refuse to take medication, it might be forcibly administered. Psychiatric labels expose poor women to further discrimination . We are treated in a discriminatory manner by medical professionals, shelter staff, welfare workers, child protection authorities, police, em- ployers and landlords as a result of being labelled with a psychiatric di- agnosis. This discrimination, like the label, is often lifelong. In the words of one member, ?Once you?re labelled, it?s for life. Once you?re diag- nosed, they put you on medications for life. They don?t think you can get better.? And as can be seen from the first testimony, there are no me- chanisms to enable a person to have such a label removed from her med- ical files, even once it is proven to be false. Legislation enables increased surveillance and control of psychia- trized women . In the period covered by this report, state surveillance and control of psychiatrized women has increased. In Ontario, as already noted, new mental health legislation has given police, doctors, and fami- ly members increased powers to incarcerate a woman in a psychiatric institution against her will. The same legislation has created Community Treatment Orders, binding orders that require people to comply with prescribed psychiatric treatment (usually medication) while living in the community, or be returned to the hospital (Szigeti, 2001). Women?s counselling services are cut while antidepressant pr e- scriptions increase. In Ontario and across Canada, the past decade has seen deep cuts to free and low-cost non-psychiatric services to support the mental and emotional well-being of poor women, such as communi- ty-based counselling centres, and counselling programs for women sur- vivors of violence. In the same period, there has been an enormous in- crease in prescriptions of mood-altering drugs by family physicians and psychiatrists (Currie, 2005). Psychiatric labels mask the devastation of poor women?s lives . Across Canada in the past decade, poor women?s lives have been devas- tated by abuse, violence, poverty, homelessness, hunger, the loss of our children, and the scapegoating we experience every day in society. Then the label of ?mental illness? is applied to our basic human reactions to the extreme stress under which we live. Instead of receiving support to empower ourselves, we are subdued with medications. Calling us crazy makes it easier for the public, the media, the government, shelter work- ers, welfare workers, police, and doctors to disregard our opinions and experiences. The right to an adequate standard of living is fundamental to the right to physical and mental health. Conclusion. We demand supports for independent living, and programs to improve our mental and emotional well-being, such as counselling and access to traditional remedies. The most important mental health ?trea tment? is an adequate standard of living. Article 1: The Right to Self-Determination: Poor Women Are Deprived of Our Own Means of Subsistence ?I mean, I could tell you stories that you are not going to be able to sleep tonight because of the humiliation that we have to go through. Now I know why people live on the street, they?d rather be on the street than in these so-called shelters. Because the social workers are horrible. They treat you like they are doing you a favour. But in time, it clicked on me: if we didn?t exist, people in need, they wouldn?t have a job. It is because of us that they got in there.? Women?s poverty and homelessness enables the creation of mid- dle-class jobs. Poor women across Canada are kept in a state of depen- dence. For our income, we are at the mercy of employers who would exploit us, or a social assistance system that would disqualify us or cut our benefits to the minimum possible level. Our families and homes are at the mercy of child protection agencies whose policies discriminate against poor women. In our housing, we are too often at the mercy of shelter workers, landlords, cohabitants, group home staff, and abusive partners. Our emotional well-being is too often in the hands of doctors who don?t listen to us, and who have increasing powers to coerce us to take whatever treatment they deem best. It hasn?t escaped our notice that the people on the other side of these relationships?employers, welfare workers, CAS workers, landlords, shelter workers, doctors?have a much better standard of living than ours. Our poverty, homelessness, marginalization and psychiatrization keep them in business. Government spending does not reflect poor women?s priorities . Government decisions about expenditures do not include our input and they do not reflect our priorities. It is well-known that shelters cost more than housing, psychiatric treatment costs more than counselling and a healthy diet, state care for children costs more than providing adequate resources to families. Yet government spending in these areas remains oriented towards the systems that maintain state control and poor wom- en?s dependence, instead of promoting our autonomy and dignity. Poor women?s lives are not for sale . Through this report, we de- mand our right to self-determination. We insist that we be included in the decisions that are made about our lives at every level?from the wel- fare office to government. Financial resources dedicated to improving poor women?s lives should be placed in our hands, not used to perpe- tuate dependence on professionals and agencies. Conclusion. We demand our right to self-determination. We insist that women who are poor, homeless, and marginalized be included in the decisions that are made about our lives at every level? from the welfare office to govern- ment. Financial resources dedicated to improving poor women?s lives should be placed in our hands, not used to perpetuate dependence on professionals and agencies. This project owes a great deal to the support of its agency partners: Sistering, the Centre for Equality Rights in Accommodation (CERA) , and the Women?s Counselling, Referral and Education Centre (WCREC). Claiming Our Rights/FORWARD was a participatory research project by Emily Paradis, who prepared this report in consultation with the women of FORWARD. The re- search was supported by a doctoral fellowship from the Social Sciences and Hu- manities Research Council of Canada. The research project ended in December 2006 and FORWARD co ntinues to operate as a self-coordinated organization of women who are homeless and marginally housed. References Aboriginal and Indigenous Social Work. ?The ?60s Scoop.? http://www.aboriginalsocialwork.ca/special_topics/60s_scoop/ , viewed March 20, 2006. Amnesty International Canada Public Brief, October 2005, http://www.amnesty.ca/campaigns/resources/sisters_brief_oct2005.pdf . CERA, Women and Housing in Canada: Barriers to Equality, March 2002 Chau, S., Fitzpatrick, A., Hulchanski, J. D., Leslie, B., & Schatia, D. (2001). One in five?Housing as a factor in the admission of children into care . University of To- ronto, Centre for Urban and Community Studies, Research Bulletin 5. Commercial Sex Information Service. (1997). ?Three Toronto Pros Murdered with the Same Gun.? News clipping from 1996, http://www.walnet.org/csis/news/toronto_96/triplemurder.html Currie, J. (2005). The marketization of depression: The prescribing of SSRI anti- depressants to women. Women and Health Protection, May 2005. Viewed March 30, 2006 on the Canadian Women?s Health Network site at http://www.whp-apsf.ca/pdf/SSRIs.pdf. Fraser v. Ontario (Attorney General), 2008 ONCA 760, viewed February 4, 2010, at http://www.ontariocourts.on.ca/decisions/2008/november/2008ONCA0760. pdf. Hedley, M.R., Tomasick, H.H., & Woolcott, D.M. (1994) Nutritional quality of food served in contracted second level lodging homes. Journal of the Cana- dian Dietetic Association, 55 (2), 75-8. Viewed March 21, 2006 at http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMe d&list_uids=10134918&dopt=Abstract Income Security Advocacy Centre. (2003) ?Denial By Design,? viewed 19 March 2006 at http://dawn.thot.net/denial_by_design.html#8 . Kellington, S. (2002). Missing Voices: Mothers at risk for or experiencing appre- hension in the child welfare system in BC. Prepared for National Action Committee on the Status of Women. http://www.nac- cca.ca/about/regions/bc/Final_Report2.pdf Mosher, J. (2002). The shrinking of the public and private spaces of the poor. In Hermer, J. and J. Mosher, (Eds.), Disorderly people: Law and the politics of ex- clusion in Ontario (pp. 41-53). Halifax: Fernwood Publishing Szigeti, A. (2001). Ontario?s community treatment orders: how did we get here and where do we go now? An advocate?s perspective. Health Law in Canada 21 (3), 66-83. Toronto Star. (n.d.). ?Adults beaten in group homes: Hundreds of cases may just be tip of iceberg.? viewed March 21, 2006, at http://www.ont- autism.uoguelph.ca/adults.PDF Toronto Star, June 20 2003, ?Deaths of young people in Ontario institutions,? viewed March 21, 2006, at http://www.namiscc.org/News/2003/Spring/MentalHealthDeaths.htm Young Mothers In/From Care Project. (2001). School of Social Work, University of Victoria.