This work may not be copied, distributed, displayed, published, reproduced, transmitted, modified, posted, sold, licensed, or used for commercial purposes. By downloading this file, you are agreeing to the publisher’s Terms & Conditions.

Original Research

Defining Response in Clinical Trials for Obsessive-Compulsive Disorder: A Signal Detection Analysis of the Yale-Brown Obsessive Compulsive Scale

David F. Tolin, PhD; Jonathan S. Abramowitz, PhD; and Gretchen J. Diefenbach, PhD

Published: December 15, 2005

Article Abstract

Objective: Many studies of the treatment of obsessive-compulsive disorder (OCD) have used percent reduction cutoffs on the Yale-Brown Obsessive Compulsive Scale (YBOCS) to classify patients as treatment responders. However, reduction criteria have varied from 20% to 50%, with studies of cognitive-behavioral therapy (CBT) using a more stringent criterion than studies of pharmacotherapy. The aim of this retrospective investigation was to determine optimal YBOCS reduction criteria for classifying patients as responders.

Method: Data from 87 adult clinic and research outpatients meeting DSM-IV-TR criteria for OCD according to structured interview were examined, comparing the percent YBOCS reduction from pretreatment to posttreatment with 2 “gold standard” criteria from the Clinical Global Impressions (CGI) scale: much or very much improved and mild illness or better. Signal detection analyses were used to determine the sensitivity, specificity, predictive value of a positive test, predictive value of a negative test, and efficiency of various YBOCS reduction cutoffs.

Results: A YBOCS reduction cutoff of 30% was optimal for predicting improvement on the CGI. The 20% cutoff used by many pharmacologic studies resulted in a high number of false positives, whereas the 50% cutoff used by most CBT studies resulted in a high number of false negatives. For predicting mild illness or better at posttreatment, a YBOCS reduction cutoff of 40% to 50% was optimal.

Conclusions: A YBOCS reduction criterion of 30% appears to be optimal for determining clinical improvement, whereas a 40% to 50% reduction criterion is appropriate for predicting mild illness at posttreatment. Future studies should employ a standard definition of treatment response in order to facilitate cross-study comparisons.

Volume: 66

Quick Links:

Continue Reading…

Subscribe to read the entire article


Buy this Article as a PDF