Deconstructing the Literature on the Treatment of Anxiety in Youth with ASD

Friday, May 13, 2016: 11:30 AM-1:30 PM
Hall A (Baltimore Convention Center)
B. M. Marro1, R. J. Weber1, L. Bungert1, S. Sommer1, C. M. Kerns2, L. J. Moskowitz3, E. Cohn4 and M. D. Lerner1, (1)Stony Brook University, Stony Brook, NY, (2)A.J. Drexel Autism Institute, Philadelphia, PA, (3)Psychology, St. John's University, Queens, NY, (4)Center for Health Innovation, Adelphi University, Garden City, NY
Background: Anxiety symptoms in Autism Spectrum Disorders (ASD) are highly prevalent (Simonoff et al., 2008). Interventions aimed at treating ASD and anxiety include different elements, each defined as a “discrete clinical technique or strategy” (Chorpita et al., 2005). Characterizing interventions by their base elements streamlines terminology, and illuminates variation and overlap amongst interventions, which may be particularly diverse when targeting comorbid syndromes. No study has reviewed discrete elements in the literature on treating anxiety in ASD. Doing so provides a uniquely informative bird’s-eye view of current evidence-based treatment in this literature. Further, as only two treatment approaches, Cognitive Behavioral Therapy (CBT) and Applied Behavioral Analysis (ABA), are considered “empirically supported” for anxiety in ASD, it is useful to see whether patterns of their use can be discerned by examining distributions of their discrete components.    

Objectives: (1) To explore which elements are used most commonly in the empirical literature treating anxiety in ASD youth. (2) To categorize those elements that are best characterized as CBT or ABA strategies. (3) To examine how participant age and gender relate to element distribution.

Methods: Researchers reviewed PubMed, PsycINFO, and Web of Science. This process yielded excellent agreement (ICC(1,2)=.957), and N = 40 articles. They were coded for elements compiled by a research consortium as representing common evidence-based practices in the field, and for participant age/gender. Inter-rater reliability for these codes was excellent (ICC(2,5)=.792). In order to classify elements as CBT or ABA, we grouped elements based on the published definitions (Gosch et al., 2006; National Autism Center, 2015), then employed an iterative backwards stepwise inclusion strategy to achieve acceptable internal consistency. We examined frequencies of individual elements in the literature, as well the presence of ABA and CBT strategies; paired-sample t-tests and χ2 tests were used to compare strategy frequency and associations with age & gender.

Results: The most common elements, appearing in ≥40% of articles, were Graduated Exposure, Cognitive Restructuring, and Homework; conversely, Meditation, Prompt Fading, and Time Out were not present (Figure 1). CBT was comprised of Cognitive Restructuring, Homework, Relaxation/Deep Breathing, and Self-Awareness Bodily Response (Cronbach’s α=.661). ABA was comprised of Prompting, Priming, Games and Activities Involving Social Interaction, Shaping, Stimulus Control, and Video/Audio Modeling (Cronbach’s α=.688). The number of CBT elements used was significantly greater than the number of ABA elements (t=2.06, p=.046; CBT: M=1.4, SD=1.37; ABA: M=0.7, SD=0.46). No significant association was found between gender or age and the presence of CBT or ABA elements (all p>.190).

Conclusions: That Graduated Exposure is the most commonly implemented element suggests that the research literature strongly relies on well-supported approaches to anxiety in ASD. The use of more CBT strategies in this review may reflect a differential preference for representing internal (i.e. cognitive aspects of anxiety) symptoms in CBT, relative to purely external (i.e. visible behavior) symptoms in ABA. Future work is warranted to similarly characterize interventions implemented in the community and to explore variation in treatment techniques on that larger scale as a function of age, gender, or other characteristics.