Wells' (Wells, A. (1997). Cognitive therapy of anxiety disorders: a practice manual and conceptual guide. Chichester, UK: Wiley) metacognitive model of obsessive-compulsive disorder (OCD) predicts that metacognitions must change in order for psychological treatment to be effective. The aim of this study was to explore: (1) if metacognitions change in patients undergoing exposure treatment for OCD; (2) to determine the extent to which cognitive and metacognitive change predicts symptom improvement and recovery. The sample consisted of 83 outpatients with a diagnosis of OCD who completed exposure and response prevention treatment. The Yale-Brown Obsessive-Compulsive Scale (Y-BOCS), the Metacognitions Questionnaire (MCQ-30) and the Obsessive Beliefs Questionnaire (OBQ-44) were administered before treatment, after treatment, and at 12-month follow-up. Treatment resulted in significant changes in symptoms, metacognition score, responsibility and perfectionism. Regression analysis using post-treatment Y-BOCS as the dependent variable indicated that when the overlap between predictors was controlled for, only changes in metacognition were significant. Changes in metacognitions explained 22% of the variance in symptoms at post-treatment when controlling for pre-treatment symptoms and changes in mood. A further regression revealed that two MCQ-30 subscales made individual contributions. The patients had significantly higher scores compared to community controls on the MCQ-30. Patients who achieved clinical significant change had lower scores on the MCQ-30 compared to patients who did not change. The results did not change significantly from post-treatment to follow-up assessment. These findings provide further support for the importance of metacognitions in treating OCD.