Anxiety sensitivity as a predictor, and non-specific therapeutic factors as predictors and mediators of CBT outcome for obsessive-compulsive disorder in a naturalistic mental health setting.

OBJECTIVE

We examined the unique predictive strength of anxiety sensitivity (AS) and the role of expectancy, credibility, and therapeutic alliance (TA) as predictors and mediators of cognitive-behavioral treatment (CBT) outcomes in obsessive-compulsive disorder (OCD).

METHOD

The current study is a prospective cohort study. Participants (N = 116) were treatment-seeking individuals with a primary diagnosis of OCD. Independent raters assessed patients on Yale-Brown Obsessive-Compulsive Scale (YBOCS) and Anxiety Sensitivity Index-3 at baseline, post-intervention, and three-month follow-up. Participants responded to the Credibility and Expectancy questionnaire and Working Alliance Inventory-Short revised at baseline, first-session, and mid-session.

RESULTS

The individual addition of AS, end-of-first-session credibility and expectancy, mid-session credibility and expectancy, and therapeutic alliance predicted significant CBT outcomes. There was a moderate positive correlation between baseline OCD severity and the global score of AS, but a weak one with AS dimensions. Both expectancy and credibility significantly improved from baseline to end-of-first-session treatment. End-of-first and third-session outcome expectancies, not credibility, have significant, indirect effects on OCD CBT outcomes.

CONCLUSIONS

AS, within-session credibility and expectancies and TA independently predict CBT outcomes. Within-sessions outcome expectancies mediate CBT outcomes in OCD, not credibility. Expectancy and credibility both include state-like elements that can be influenced to enhance the outcomes of CBT. Proposals for reducing treatment barriers in CBT for OCD are offered.