Differences in the Anxiety Profiles of Middle Childhood Boys and Girls with ASD

Background: Symptoms of anxiety are reported commonly in ASD, with approximately 30-50% of individuals with ASD meeting criteria for an anxiety disorder. In typical development, anxiety disorders affect boys and girls at similar rates during early childhood, but at higher rates in females by adolescence. It is unclear whether these sex differences exist in the context of ASD. The current study evaluates a longitudinal dataset of children with ASD to examine potential predictors of the development of clinically significant anxiety in children with ASD.

Objectives: To examine anxiety prevalence and severity in boys and girls with ASD. To define early childhood predictors of the development of clinically significant anxiety at middle childhood.

Methods: Participants studied to date include 30 children with ASD (20 male/10 female; IQ range 39-123; ADOS-2 composite score range 5-10) enrolled in the ongoing Autism Phenome Project. Early childhood behavioral measures (Autism Diagnostic Observation Schedule-Generic [ADOS-G], Childhood Behavior Checklist for Ages 1½-5 [CBCL], Short Sensory Profile [SSP], and Vineland Adaptive Behavior Scales-II [VABS-II]) were collected at approximately 3 years-of-age. The children are currently between 9 and 13 years of age and clinically significant anxiety was evaluated using the clinician administered Anxiety Disorders Interview Schedule (ADIS-IV) parent version with the Autism Spectrum Addendum. Children with a clinical severity rating of 4 or greater were classified as having an anxiety disorder.

Results: In middle childhood, there is a trend towards a greater proportion of girls with ASD who have clinically significant anxiety than boys with ASD (80% of girls vs. 65% of boys, X²=0.71, p=0.39). Girls with ASD also have more severe anxiety than boys as measured by the ADIS-IV clinical severity rating (girls=5.20, boys=3.70, F=5.6, p=0.024). Interestingly, for girls with ASD, but not for boys, parent-ratings of behavior on the CBCL at 3 years-of-age predicts anxiety severity in middle childhood for three subscale scores: Internalizing t-score (F=17.62, p=0.002, R²=0.716), Withdrawn t-score (F=10.01, p=0.016, R²=0.539) and Emotionally Reactive t-score (F=12.96, p=0.009, R²=0.649). Scores on the ADOS-G, SSP, and VABS-II were not predictive of anxiety for boys or girls.

Conclusions: By middle childhood, a greater proportion of girls with ASD develop anxiety disorders and with greater severity, compared to boys with ASD. In typical development, sex differences in the prevalence of significant anxiety are present around this age (approximately 12% of girls versus 8% of boys have an anxiety disorder). It should be noted that both girls and boys with ASD are presenting with anxiety at rates much higher than their typically developing peers. In girls, the CBCL also may provide an early predictor for the future development of anxiety, which holds clinical implications for specific early intervention of anxiety symptoms. This study provides evidence for sex differences in the anxiety profiles of girls and boys with ASD.