Abstract
BACKGROUND: There are high attrition rates observed in efficacy studies for social anxiety disorder, and research has not identified consistent nor theoretically meaningful predictors of dropout. Pre-treatment symptom severity and demographic factors, such as age and gender, are sometimes predictive of dropout. The current study examines a theoretically meaningful predictor of attrition based on experiences associated with social group membership rather than differences between social group categories--fear of confirming stereotypes. METHODS: This is a secondary data analysis of a randomized controlled trial comparing two cognitive behavioral treatments for social anxiety disorder: virtual reality exposure therapy and exposure group therapy. Participants (N = 74) with a primary diagnosis of social anxiety disorder who were eligible to participate in the parent study and who self-identified as either "African American" (n = 31) or "Caucasian" (n = 43) completed standardized self-report measures of stereotype confirmation concerns (SCC) and social anxiety symptoms as part of a pre-treatment assessment battery. RESULTS: Hierarchical logistic regression showed that greater stereotype confirmation concerns were associated with higher dropout from therapy--race, age, gender, and pre-treatment symptom severity were not. Group treatment also was associated with higher dropout. CONCLUSIONS: These findings urge further research on theoretically meaningful predictors of attrition and highlight the importance of addressing cultural variables, such as the experience of stereotype confirmation concerns, during treatment of social anxiety to minimize dropout from therapy.