Social Strengths of Autistic Girls: Sex Differences in Clinician-Rated and Parent-Reported Autistic Traits

Background: Females are diagnosed with autism spectrum disorder (ASD) at significantly lower rates than males, with particularly discrepant rates among those without co-occurring intellectual disability (ID; Loomes et al., 2017). A growing body of research indicates that females with ASD without co-occurring ID may have important social strengths that “camouflage” autistic traits, making the condition harder to capture (Lai et al., 2016). We have previously found sex differences in parent-reported autistic traits among school-age children with ASD, but not on gold-standard diagnostic measures (Ratto et al., 2017).

Objectives: The aim of the present study was to further investigate sex differences in autistic traits among school-aged children on a clinician-rated and parent-report to extend prior findings.

Methods: The sample was composed of participants ages 8-18 years enrolled in the Gender Exploration of Neurogenetics and Development to Advance Autism Research (GENDAAR) study. From a total sample of 523 participants, a subsample of 196 children (n=85 female) were identified who had been administered the Module 3 of the Autism Diagnostic Observation Schedule-2 (ADOS-2) and had a clinician-confirmed diagnosis of ASD. There were no significant differences by sex in age [t(194)= -.22, p=.83] or full-scale IQ [t(194)= -1.11, p=.27]. Independent samples t-tests and chi-square analyses were used to evaluate sex differences in ADOS-2 performance, as well as parent report on the Social Responsiveness Scale (SRS) and the Vineland Adaptive Behavior Scales-II. Chi-square analyses were used to examine differences on select items on the ADOS-2, which prior literature suggested may be strengths for girls with ASD. Social items included Conversation, Facial Expressions, Empathy, Social Insight, Imaginative Play, and repetitive items included Stereotyped Language, Restricted Interests, and Compulsive Behaviors.

Results: Overall, girls were significantly less likely to meet criteria on the ADOS-2 [X²(1, N=196)= 4.80, p=.03], though there were no significant sex differences in total scores on the diagnostic algorithm. Socially, girls showed better conversation skills [X²(2, N=196)= 5.85, p=.05], empathy [X²(2, N=196)= 18.00, p<.0001], and imaginative play than males [X²(3, N=196)= 14.63, p=.002], but no differences in facial expressions or social insight. With respect to repetitive features, girls showed fewer compulsive behaviors [X²(2, N=196)= 7.19, p=.03], but no differences in stereotyped language or restricted interests. Parents of girls reported greater communication [t(179)= -3.01, p=.003] and social [t(180)= -2.10, p=.04] skills on the Vineland-II, though not on the SRS. All results retained significance after controlling for multiple comparisons using the Benajmini-Hochberg procedure.

Conclusions: Although there is a growing literature indicating that non-intellectually disabled girls with ASD may show more subtle autistic traits, most prior studies have not identified differences in performance on gold-standard diagnostic measures. The sample used in this study was specifically recruited to be sex-balanced and thus better able to assess possible sex differences. Girls in this study showed important social strengths in both clinician and parent ratings, as well as reduced compulsive behavior. Despite comparable average algorithm scores, girls were less likely to meet ADOS criteria, suggesting that an adjusted cut-off score may need to be explored in future studies.