Factors that Mediate Effective Treatment (CBT) of Anxiety in Children with Autism Spectrum Disorders: Lessons Learned and Potential Impact on Implementation in New Settings

Background: Anxiety disorders are the most commonly reported concurrent diagnoses in individuals with autism spectrum disorders (ASD; White et al., 2009); however, there is a paucity of information examining the factors that might contribute to or underlie anxiety. Studies examining the use of cognitive behavioral therapy (CBT) to treat anxiety in ASD have recently begun to receive empirical attention (Lang et al., 2010). Although results are promising, CBT is not effective for all children with ASD. Review of previous treatment research shows significant reductions in parent reported anxiety symptoms after delivery of CBT; however, approximately one third of children still had clinically significant levels of anxiety. In order to improve treatment effectiveness, examination of mediational processes is essential to ensure that we understand how and why treatments work.

Objectives: The current study investigates mechanisms that mediate treatment responsiveness of children with ASD receiving CBT for anxiety. This is a pilot study to assess feasibility and variance in the mediator and outcome variables to calculate sample size requirements for a larger, randomized control trial study.

Methods: Literature examining mediators and moderators of CBT will be reviewed and discussed in the context of the current pilot project. Twenty-one participants (from 6 groups of approximately 5 children each) with ASD (IQ>70) and their parent(s) participated in a group CBT program for individuals with ASD, Facing Your Fears (i.e., Reaven et al., 2009). Groups were run at Chedoke Hospital, York University and Holland Bloorview Kids Rehabilitation Hospital. Participants were recruited from referrals from clinicians and advertisements on autism-related websites. The current pilot project examined four potential mediators of therapeutic effectiveness: cognitive (i.e., executive function-EF & attentional flexibility), social (i.e., social skills/awareness), parental/family (i.e., family functioning & parental stress), and therapeutic variables (i.e., support & therapeutic alliance). Measures of anxiety and potential mediators were administered pre-treatment, at two time points during treatment, and immediately and at 4-6 months post-treatment.

Results: We have completed data collection across all phases of this project, and are currently assessing how change in child anxiety is explained by changes in the child, family, and interpersonal factors. The CBT groups were run in university settings as a research project, as well as a clinical setting where the added research component was optional. Implementation of measures to assess the various mediators was challenging given the frequency of assessments, but this process presented with unique challenges in the clinical (versus research) context, which will be addressed. 

Conclusions: There is a tremendously high need for services for individuals with ASD, and yet there are very few public services for this group, as they often ‘fall through the cracks’ in our mental health care system, thus improving treatments is a priority. It is important to expand our understanding about the underlying mechanisms of change in order help guide treatment programs in effectively addressing these issues and understanding factors that may support implementation in new settings.