The presence of overvalued ideation (OVI) in patients with obsessive-compulsive disorder (OCD) has been theoretically linked to poorer treatment outcome. OVI has recently been shown to predict treatment outcome in OCD. The purpose of the present study is to determine whether OVI predicts medication treatment response, controlling for initial symptom severity and measurement error. The sample consisted of 34 outpatients diagnosed with OCD who completed an open-label clinical trial of fluvoxamine of 10 weeks' duration. Clinicians administered the Overvalued Ideas Scale (OVIS) at baseline. Symptom severity was rated at baseline and at the end of week 10 using the Yale-Brown Obsessive-Compulsive Scale. Of those who completed the trial, 68% showed a reliable change in obsessions and 62% showed a reliable change in compulsions. Analysis of variance showed that baseline OVIS predicted outcome for obsessions, but not compulsions. A key limitation was the relatively low number of individuals in the upper quartile on the OVIS, thus reducing the predictive power of the measure in relation to treatment outcome. Future research should examine medication treatment outcome with higher scoring patients.