Psychological Medicine

Original Articles

Moderators and predictors of response to cognitive-behavioral therapy augmentation of pharmacotherapy in obsessive–compulsive disorder

M. J. Mahera1, J. D. Hupperta2a3, H. Chena1a4, N. Duana1a4, E. B. Foaa2, M. R. Liebowitza1a4 and H. B. Simpsona1a4 c1

a1 New York State Psychiatric Institute, New York, NY, USA

a2 Department of Psychiatry, University of Pennsylvania School of Medicine, Philadelphia, PA, USA

a3 Department of Psychology, The Hebrew University of Jerusalem, Jerusalem, Israel

a4 Department of Psychiatry, Columbia University, New York, NY, USA


Background Cognitive-behavioral therapy (CBT) consisting of exposure and response prevention (EX/RP) is efficacious as a treatment for obsessive–compulsive disorder (OCD). However, about half of patients have a partial or poor response to EX/RP treatment. This study examined potential predictors and moderators of CBT augmentation of pharmacotherapy, to identify variables associated with a poorer response to OCD treatment.

Method Data were drawn from a large randomized controlled trial that compared the augmenting effects of EX/RP to stress management training (SMT; an active CBT control) among 108 participants receiving a therapeutic dose of a serotonin reuptake inhibitor (SRI). Stepwise regression was used to determine the model specification.

Results Pretreatment OCD severity and gender were significant moderators of outcome: severity affected SMT (but not EX/RP) outcome; and gender affected EX/RP (but not SMT) outcome. Adjusting for treatment type and pretreatment severity, significant predictors included greater co-morbidity, number of past SRI trials, and lower quality of life (QoL). Significant moderators, including their main-effects, and predictors accounted for 37.2% of the total variance in outcome, comparable to the impact of treatment type alone (R2=30.5%). These findings were replicated in the subgroup analysis of EX/RP alone (R2=55.2%).

Conclusions This is the first randomized controlled study to examine moderators and predictors of CBT augmentation of SRI pharmacotherapy. Although effect sizes for individual predictors tended to be small, their combined effect was comparable to that of treatment. Thus, future research should examine whether monitoring for a combination of these risk factors and targeting them with multi-modular strategies can improve EX/RP outcome.

(Received June 01 2009)

(Revised August 27 2009)

(Accepted August 30 2009)

(Online publication April 26 2010)


c1 Address for correspondence: H. B. Simpson, M.D., Ph.D., Associate Professor of Clinical Psychiatry, Columbia University, 1051 Riverside Drive, Unit 69, New York, NY 10032, USA. (Email: