Category: Adult Anxiety - Social

PS10- #A4 - Outcome Expectancy, Working Alliance, and Symptom Reduction in SAD

Saturday, Nov 18
11:00 AM – 12:00 PM
Location: Indigo Ballroom CDGH, Level 2, Indigo Level

Keywords: Adult Anxiety | Change Process / Mechanisms | Therapeutic Alliance

Eighty years ago, Rosenzweig (1936) introduced the concept of common factors- aspects of the therapeutic setting that are common across all types of treatment. These variables include the client’s expectations about the effectiveness of treatment (termed outcome expectancy), and the quality of the therapeutic relationship (termed working alliance). Eight decades of research have demonstrated a robust relation between these variables and treatment response, yet little is known about the mechanisms by which these common factors effect change. Recent investigations suggest that the working alliance mediates the relation between outcome expectancy and treatment response, but this model has not been applied to the treatment of social anxiety disorder, nor has it been applied to virtual reality exposure therapy. Using archival data (Anderson et al., 2013), the present study tests the hypothesis that the relation between outcome expectancy and symptom reduction is mediated by the alliance following treatment for social anxiety disorder. Data were collected in a sample of 65 individuals diagnosed with social phobia with public speaking fears. Participants received 8 sessions of either virtual reality exposure therapy or group exposure therapy conducted according to a treatment manual. Participants completed standardized self-report measures of outcome expectancy at the first session, working alliance measures at all treatment sessions, and questionnaires on symptoms of social phobia at pre- and post-treatment.  Mediation analyses were conducted using Andrew Hayes’ Process Macro (Hayes, 2013). Outcome expectancy served as the independent variable, working alliance as the mediating variable, and two measures of social anxiety symptoms as the outcome variables. None of the mediation analyses were significant. These findings suggest that the mechanisms of common factor variables may vary by disorder. Implications and future directions are discussed.

Amanda A. Benbow

Graduate Student
Georgia State University
Atlanta, Georgia

Page L. Anderson

Associate Professor
Georgia State University