The treatment guidelines for obsessive-compulsive disorder (OCD) propose to switch serotonin reuptake inhibitors (SRIs) in case of refractoriness. However, no controlled research has been published yet that prospectively examined the effects of changing SRIs. This article describes the first double-blind switch study of 2 SRIs in patients with OCD. 150 patients with primary OCD, according to DSM-IV criteria, were randomly assigned in a 12-week, double-blind trial to receive dosages titrated upward to 300 mg/day of venlafaxine (N = 75) or 60 mg/day of paroxetine (N = 75). Primary efficacy was assessed by the change from baseline on the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS), and nonresponse was defined as less than 25% reduction on the Y-BOCS. After a 4-week tapering phase, 43 nonresponders were switched to 12 additional weeks of the alternate antidepressant, of which 16 patients received venlafaxine and 27 received paroxetine. Eighteen of 43 patients benefited from a switch to the alternate SRI with a mean +/- SD decrease of at least 25% on the Y-BOCS. At the end of 12 weeks, responder rates were 56% for paroxetine (15/27) and 19% for venlafaxine (3/16). An intent-to-treat, last-observation-carried-forward analysis demonstrated a mean decrease on the Y-BOCS of 1.8 +/- 3.5 in the venlafaxine group and 6.5 +/- 7.1 in the paroxetine group. After 2 consecutive SRI trials, 109 of 150 patients (73%) achieved a Y-BOCS decrease of at least 25%. The results of the current study show that 42% of the nonresponders benefited from a crossover to the other SRI, and that paroxetine was more efficacious than venlafaxine in the treatment of nonresponders to a previous SRI trial. Switching SRIs in case of refractoriness may be considered a useful strategy for patients with OCD.