Investigation of Individual Factors Associated with Anxiety in Youth with Autism Spectrum Disorder

Background:  Prior research has indicated anxiety symptoms and disorders are highly prevalent in youth with autism spectrum disorder (ASD) and may impart additional impairment (Van Steensel, Bogels, & Perrin, 2011).  The extant literature yields conflicting findings regarding factors related to anxiety (e.g., cognitive functioning, social impairment) and the nature of such relations. Validating and extending prior anxiety risk factor research is necessary to improve assessment and guide treatment.

Objectives:  To identify individual factors associated with anxiety risk and explore whether cognitive functioning moderates such associations in a large sample of youth with ASD.

Methods:  The present study used extant data from the Simons Simplex Collection (SSC), a database of genotypic and phenotypic information obtained from 4- to 18-year-old youth with ASD (n=2662).

A dichotomous anxiety risk outcome variable was produced from scores on parent-rated Child Behavior Checklist (CBCL; Achenbach & Rescorla, 2001) DSM-IV Anxiety Problems scale based on the CBCL-defined cut point (scores 65 and above = high risk). Factors with hypothesized relations to anxiety included social withdrawal, adaptive social behavior, autism severity, intellectual functioning, age, and gender.  Factors were derived from the Aberrant Behavior Checklist (ABC; Aman et al., 1985), Autism Diagnostic Observation Schedule (ADOS; Lord et al., 2002), Vineland Adaptive Behavior Scales, 2^nd Edition (Vineland-II; Sparrow et al., 2005), and multiple measures of intellectual functioning (e.g., Differential Ability Scales, 2^nd Edition [DAS-II; Elliot, 2007]).

A multivariate logistic regression model was built to examine the factors independently associated with anxiety risk while controlling for all other factors.  Then, a hierarchical logistic regression model was created to investigate whether intellectual functioning moderated the relation between hypothesized factors and anxiety risk.  Finally, the nature of any significant moderation was probed by examining the conditional effects of factors on anxiety risk for youth with high and low cognitive functioning.

Results:  Results from the first model indicated anxiety risk was related to four individual factors at the α=.05 level.  High-risk youth were less likely to be female (odds ratio [OR] = 0.69, 95% CI [0.54, 0.89]) and more likely to have average or greater cognitive functioning (OR = 1.96, 95% CI [1.58, 2.43]).  Likelihood of being classified as high risk increased as age (odds ratio = 1.11, 95% CI [1.08, 1.13]) and social withdrawal increased (OR = 1.07, 95% CI [1.06, 1.09]).

Moderation analysis indicated intellectual functioning moderated the relation between adaptive social behaviors and anxiety risk (OR = 0.97, 95% CI [0.96, 0.99], p < .05).  Simple slope analyses revealed that higher adaptive social behaviors were associated with membership in the high anxiety risk group for youth with lower cognitive functioning (OR = 1.02, CI [1.00, 1.03]).  There was no relation between adaptive social behaviors and anxiety risk for youth with average or higher cognitive functioning.

Conclusions:  Results from the present study provide support for previously identified risk factors; however, further investigation is necessary to uncover additional risk factors and to explore how their relation to anxiety differs across individuals with ASD and varying levels of cognitive functioning.