Exploring Relationships Between Anxiety Symptoms and Repetitive Behaviors in Children with Autism Spectrum Disorders

Background: Repetitive behaviors, a core deficit of autism spectrum disorders (ASD), are often interfering and problematic in the daily lives of individuals with ASD. At present, we know little about the etiology of these behaviors or the extent to which they may be governed by endogenous processes (e.g., cognition, affect). Hutt and Hutt (1965) hypothesized that repetitive behaviors may serve to regulate arousal level: upwards for hypo-aroused individuals and downwards for individuals with elevated anxiety levels (potentially as a means of self-soothing or distraction).  Others (Baron-Cohen, 1989; Carruthers, 1996) have posited separate models wherein social anxiety triggers repetitive behaviors in individuals with ASD.  However, the exact nature of the relationship between anxiety symptoms and repetitive behaviors in children with ASD remains unclear.

Objectives: This study tests the hypothesis that repetitive behaviors are related to anxiety symptoms in individuals with ASD by examining data from two parent-report measures: the Repetitive Behavior Scale-Revised (RBS-R) and the Screen for Child Anxiety Related Emotional Disorders (SCARED). 

Methods: Data are presented on 20 participants between the ages of 6 and 16 years (90% male; mean age=10.1 years, mean IQ=99.1; range: 47-158) diagnosed with ASD using gold standard diagnostic measures (ADOS, ADI-R) and expert clinical judgment; assessment of intellectual functioning was also completed via the Differential Ability Scales Second Edition. Analyses consisted of general linear model-based comparisons of anxiety symptoms and repetitive behaviors, and post-hoc analyses including IQ. Data collection is ongoing. 

Results: When controlling for IQ, preliminary results indicate that the Self-Injurious Behavior subscale on the RBS-R is significantly correlated with the SCARED overall score (r=.58, p<.05), Panic Disorder subscale (r=.53, p<.05), and Separation Anxiety Disorder subscale (r=.63, p< .01). Interestingly, preliminary analyses show no significant relationships between the Social Anxiety Disorder subscale on the SCARED and any of the RBS-R subscales.

Relationships between IQ and both anxiety symptoms and repetitive behaviors are observed, such that overall IQ is significantly related to both the Generalized Anxiety Disorder subscale on the SCARED (r=.56, p<.05) and the Stereotyped Behavior subscale on the RBS-R (r=-.53, p<.05). Further mediation analyses will be conducted with a larger sample.

Conclusions: Preliminary results suggest that self-injurious behavior may represent a maladaptive coping mechanism for anxiety in individuals with ASD. If this finding holds, it opens up potential avenues for the treatment of repetitive behaviors (particularly self-injury) that focus on anxiety management. Recent studies indicate that behavioral strategies can mitigate anxiety symptoms in ASD (Wood et al. 2009); it is therefore worth exploring whether behavioral strategies for anxiety could also lead to decreases in repetitive behaviors. Additionally, results raise the possibility that higher intellectual functioning in children with ASD may serve as a potential risk factor for generalized worry, and a potential protective factor for stereotyped behaviors.  Future studies in this area will refine our understanding of the precise relationship between anxiety and repetitive behaviors in ASD, and expand our knowledge of the role of behavioral strategies in reducing both of these symptoms.