Autism Spectrum Disorder In Girls: Sex Differences, Subtypes, And Symptom Structure

The present study investigated how autism spectrum disorder manifests in female children through three separate studies. Two separate datasets were used. The Surrey Place (SP) dataset (N=1725) was composed of children screened to participate in an intensive treatment program. The SFARI dataset (N=2757), is a collection of data representing autistic children who participated in the Simons Simplex Collection (SSC; Fischbach & Lord 2010). Each study boasts a large sample of autistic females (n=301, and n=375, respectively). The datasets differed from one another in many important ways, such as the characteristics of the sample, method of data collection, and the measures used.

Study 1 comprised an investigation of sex differences between female and male autistic children across a wide range of developmental and diagnostic characteristics. Groups of male and female children were compared using frequency data, correlations, and Multivariate Analysis of Covariance. Results of this study showed significant differences based on sex. That is, boys tended to have higher FSIQ, Non-Verbal IQ and adaptive abilities than girls. No differences were found in terms of total autism symptom scores. Boys were found to have higher levels of RRBIs and Self Injurious Behaviour.

Study 2 used Latent Profile Analysis to explore subtypes within the sample of girls; that is, autistic girls with specific characteristics that cluster together, and likewise within the samples of boys. Results indicated three distinct profiles (or classes) for the SP dataset and these were similar in boys and girls. Class 1 consisted of children who were slightly younger than the sample mean, with adaptive scores close to the sample mean, and autism scores that showed mild to moderate symptoms of autism. Class 2 again consisted of younger children with better adaptive skills, and mild to no autism symptoms. Class 3 consisted of older children, with lower adaptive skills, and more severe autism symptoms. The SFARI sample included more developmental variables and allowed for a more nuanced analysis. As a result, many possible combinations of profiles emerged. Using LPA fit statistics and clinical judgement, a five-profile solution was deemed to best fit the data. Visual comparison across the profiles for girls and boys found that four of the five profiles were largely similar across boys and girls. One of the five profiles showed unique and distinct profiles for girls and boys.

Study 3 used confirmatory factor analysis to investigate the symptom structure of autism between females and males, based on factor structures reported in previous research. Overall, the factor structure was found to be largely consistent for males and females.

These results suggest that sex differences in the current study were minimal, although some subtle differences were noted. More research is needed to better understand the unique qualities of autistic females. Neuro-affirming diagnostic measures and more representative research samples may allow for the discovery of more sex differences.