Predictors and Moderators of Cognitive and Behavioral Therapy Outcomes for OCD: A Patient-Level Mega-Analysis of Eight Sites

Cognitive (CT) and behavioral treatments (BT) for OCD are efficacious separately and in combination. Tailoring treatment to patient-level predictors and moderators of outcome has the potential to improve outcomes. The present study combined data from eight treatment clinics to examine the benefits of BT (n = 125), CT (n = 108), and CBT (n = 126), and study predictors across all treatments and moderators of outcome by treatment type. All three methods led to large benefits for OCD and depression symptoms. Residual gain scores for OCD symptoms were marginally smaller for BT compared to treatments containing CT. For depression, significantly more gains were evident for CBT than BT, and CT did not differ from either. Significantly fewer BT participants (36%) achieved clinically significant improvement compared to CT (56%), and this was marginally evident for CBT (48%). For all treatments combined, no predictors were identified in residual gain analyses, but clinically improved patients had lower baseline depression and stronger beliefs about responsibility/threat and importance/control of thoughts. Moderator analyses indicated that higher baseline scores on depression adversely affected outcomes for BT but not CT or CBT, and lower OCD severity and more education were associated with positive outcomes for CT only. A trend was evident for higher responsibility/threat beliefs to moderate clinical improvement outcomes for those receiving cognitive (CT and CBT), but not behavioral (BT) treatment. Medication status and comorbidity did not predict or moderate outcomes. Findings are discussed in light of models underlying behavioral and cognitive treatments for OCD.