Multimodal Anxiety and Social Skills Intervention for Adolescents with Autism Spectrum Disorders (ASD): Feasibility and Preliminary Efficacy

Background:  Anxiety is more common in children and adolescents with ASD than age mates without ASD (White et al., 2009; Merikangas et al., 2010). When present, anxiety may compound the social disability that characterizes these disorders. Moreover, untreated anxiety may hinder clinical efforts solely aimed at improving social skill deficits. There is evidence that anxiety may be most problematic during adolescence for people with ASD (Tse et al., 2007; Witwer & Lecavalier, 2010). There are few manual-based treatment curricula, however, developed specifically for adolescents with ASD.

Objectives:  We assessed the feasibility (i.e., consumer acceptability and satisfaction, treatment fidelity, and subject compliance with treatment) of a cognitive-behavioral intervention program (Multimodal Anxiety and Social Skills Intervention for Adolescents: MASSI) that addresses anxiety and ASD-related social disability in adolescents. Secondarily, we explored preliminary outcome data to judge whether further investigation is warranted.

Methods:  We conducted a randomized controlled trial (RCT) with 30 adolescents (ages 12 – 17) with ASD and at least one co-occurring anxiety disorder. All participants were in the average IQ range  (mean verbal IQ = 97.07).  Social Phobia was the most common anxiety disorder in the sample (77%), but most subjects had multiple anxiety disorders. Participants were randomized, 15 per group, to MASSI or a 14-week waitlist condition. Intent-to-treat and completer analyses are presented.

Results:   The intervention was acceptable to the adolescents and their parents as evidenced by average attendance at 93% of required treatment sessions, and the high study completion rate (83.3%). Adolescents reported that the group therapy meetings were the most useful component of the intervention. By contrast,  parents reported that the individual sessions were  most helpful. Mean subject adherence across sessions, measured by homework completion, was 58% (range 0% to 100%).  Therapist fidelity ranged from 87.5% to 100% (mean 94.09%). Preliminary efficacy results on parent-reported symptoms of ASD-related social disability and anxiety were inconsistent. There was a large, within-group effect size of 1.18 on parent-reported social disability (Social Responsiveness Scale; Constantino & Gruber, 2005).  However, the within-group effect size on parent-reported anxiety symptoms was modest at .55 (Child and Adolescent Symptom Inventory-4 Anxiety Scale; Sukhodolsky et al., 2008).

Conclusions:  Results from this pilot study suggest that the intervention is acceptable and can be delivered reliably. The outcome data are promising and suggest that a larger trial is warranted.