(PS11-D81) Specific Autism Symptoms Differentially Predict Self- and Parent-reported Social Anxiety in Youth with and Without ASD

Youth with autism spectrum disorder (ASD) may be at elevated risk of experiencing social anxiety (SA) due to the impact of social-communicative challenges characteristic of ASD on social interactions and relationships (Spain et al., 2018). While ASD symptoms are continuous, quantitative traits that extend into the general population (Hoekstra et al. 2008), it remains unclear if the relationship between ASD symptoms and SA is different in typically-developing (TD) youth. Considering both parent- and child-reported measures of SA may be important, as reports may differ due to parents’ reliance on overt symptoms despite the covert nature of many anxiety symptoms and children’s challenges with introspection (Comer & Kendall, 2004; Storch et al., 2012). For instance, some studies suggest social-communicative challenges are associated with only parent-reported SA (Hallett et al., 2013) versus only self-reported SA (Spain et al., 2018). Given the heterogeneity in ASD symptoms, this study considers how specific subdomains of ASD symptoms may relate to SA in youth and how this relationship may differ between youth with and without ASD.

Participants included 167 youth (M = 14.0 years, SD = 1.96, 49% ASD; 72% male) and their parents, who reported on ASD symptom severity (SRS-2; Constantino & Gruber, 2012). SA was measured by parent- (MASC-P; March, 2013) and self-report (MASC-C & SAS-A; March, 2013, La Greca, 1998). Correlations were used to establish associations between overall SRS-2 and SA. Multiple linear regression with SRS-2 subscales entered simultaneously was used to examine the role of each SRS-2 subscale on SA while controlling for other subdomains. Moderation analyses were conducted to examine differences by group status (ASD versus TD).

SRS-2 was correlated with MASC-P SA (r = .373, p< .001) and SAS (r=.173, p< .05), but not MASC-C SA. MASC-P SA was predicted by higher SRS-2 social motivation scores (B=.541, p< .001) and lower SRS-2 social awareness scores (B=-.345, p< .001). Conversely, SAS was only predicted by higher SRS-2 social motivation scores (B=.477, p< .001).

Group status moderated the relationship between MASC-P SA and overall SRS-2 (B=.299, p< .05), such that the effect was stronger in TD (B=.532, p< .001) than ASD (B=.234, p< .05) youth. The effect for MASC-P and SRS-2 social motivation was also moderated by group status (B=-.237, p< .05), with a stronger effect in TD (B=.630, p< .001) than ASD (B=.393, p< .001) youth. There was no interaction for SAS and SRS-2 by group status (p >.10).

Different ASD symptom profiles, specifically fewer challenges with social awareness or greater difficulties with social motivation, predicted higher SA in youth with and without ASD. The relationship between SRS-2 (specifically the social motivation subscale) and parent-reported SA in TD youth was stronger than in ASD youth. These findings suggests that universal (e.g., classroom-based) interventions for ASD social symptoms may elicit corollary improvements in SA across populations. These efforts may be especially pertinent considering the high prevalence and chronicity of SA, particularly in ASD youth and in the absence of therapeutic intervention (Keller, 2003; Spain et al., 2018).