Standard criteria for relapse are needed in obsessive-compulsive disorder

Abstract

To assess how different criteria for relapse affect inferences about relapse in obsessive-compulsive disorder (OCD), a post hoc analysis of relapse was conducted using data from a multisite randomized controlled trial comparing clomipramine (CMI), exposure and ritual prevention (EX/RP), and its combination (EX/RP+CMI) in adults with OCD. Different relapse definitions were constructed based on criteria used in prior studies. For each definition, the number of relapsers was computed, and the proportion of relapsers and time to relapse were compared. When applied to this data set, relapse criteria used in prior OCD studies yielded different observed relapse rates (range: 27-63% for CMI; 0-50% for EX/RP; and 7-67% for EX/RP+CMI). Most criteria found that EX/RP responders (with or without CMI) had a significantly lower relapse rate and longer time to relapse after treatment discontinuation than did responders to CMI alone. However, some relapse criteria (e.g., those requiring minimal worsening) found no significant treatment differences in relapse rates or time to relapse, and some generated biases against one treatment or another. Most definitions concurred: in adults with primary OCD, EX/RP treatment (with or without CMI) can produce more durable short-term gains after treatment discontinuation than CMI alone. However, different relapse criteria can lead to very different observed relapse rates and even contradictory inferences about relapse. Standard criteria for relapse are needed in OCD to facilitate comparisons between studies (enabling better treatment guidelines) and to advance research on mechanisms of relapse and relapse prevention.