Gender Identity in Autism Spectrum Disorder and Attention Deficit Hyperactivity Disorder

Background: Evidence suggests a higher-than-expected co-occurrence of autism spectrum disorder (ASD) and gender dysphoria (GD) in children, adolescents and adults (van der Miesen et al., 2015). GD is characterized by distress due to incongruence between assigned gender and experienced/expressed gender. Recent studies have found that ASD is more common among children who are clinically referred for GD compared to those referred for other psychiatric/psychological concerns (Leef et al., in preparation). Among youth with ASD, elevated traits of GD have also been found (e.g., May et al., 2017). These findings imply a unique relationship between ASD and GD. However, one study reported similarly elevated GD traits among youth with ADHD (Strang et al., 2014), suggesting that elevated rates of GD traits may not be unique to ASD versus other neurodevelopmental conditions.

Objectives: To explore whether traits of GD are elevated in children and adolescents with ASD relative to ADHD and controls.

Methods: Data from 644 participants enrolled in the Province of Ontario Neurodevelopmental (POND) Network were analyzed: ASD (n=370, M age=11.02, SD=3.32), ADHD (n=115, M age=10.03, SD=2.90), and typically developing controls (TD; n=159, M age=11.35, SD=3.20). We used one conventional method of assessing the propensity towards GD by analyzing a parent-reported item on the Child Behavior Checklist (CBCL) “wishes to be of opposite sex” (Item 110).

Results: Fisher’s Exact Test revealed no significant association between endorsement of Item 110 and group (Fisher’s Exact=1.38, p=.48; 2.2% in ASD, 1.7% in ADHD, and .6% in controls). Notably, of those for whom this item was endorsed, 73% had ASD, 18% had ADHD, and 9% were controls. Compared directly to controls, the ASD group had an Odds Ratio of 3.50; compared to ADHD, the Odds Ratio for ASD was close to 1 (.80).

Conclusions: This is one of the largest studies to explore a specific GD trait in a research-based sample of youth with ASD, ADHD and TD controls. The rate of endorsement of CBCL Item 110 did not significantly differ between ASD and ADHD. This finding is in line with Strang et al.’s (2014) report of clinically-referred youth with ASD and ADHD. The rate in the current controls was consistent with the CBCL non-referred standardization sample (i.e., .7%). It is noteworthy that, among those with this particular GD trait, the vast majority had ASD, and participants with ASD were 3.5 times more likely to have this GD trait than controls. This is consistent with recent findings that a significant proportion of children with GD have ASD (>20%; Leef et al., in preparation). Although this is a large sample, the number of individuals with this GD trait was quite small, making it difficult to conclude whether the link between ASD and GD is unique to ASD compared to other neurodevelopmental conditions. Further analyses will explore associations with other characteristics, including participant sex (i.e., birth-assigned gender), birth weight, repetitive behaviours, and pubertal stage.