Benchmarking empirical severity for the Yale-Brown Obsessive Compulsive Scale-Second Edition

Abstract

The Yale-Brown Obsessive Compulsive Scale (Y-BOCS) is considered the primary instrument for assessing the presence and severity of obsessive-compulsive disorder (OCD). Conceptual and empirical critiques inspired the development of an updated version of the instrument, the Yale-Brown Obsessive Compulsive Scale-Second Edition (Y-BOCS-II), with a higher ceiling of OCD severity to better differentiate between severe and the most debilitating OCD presentations, among other revisions. The Y-BOCS-II has demonstrated sound psychometric properties across diverse samples. Empirically derived severity benchmarks have been proposed for the original Y-BOCS, yielding somewhat different ranges than what has been commonly used in clinical and research settings, yet severity benchmarks for the Y-BOCS-II have yet to be established. Using a diverse, pooled sample of 2982 children and adults with OCD or obsessive-compulsive and related concerns across 13 countries, receiver-operating characteristic (ROC) analyses yielded severity benchmarks that largely mirrored the original Y-BOCS at the lower range of scores and extended the previously established benchmarks at the higher range of scores, owing to the increased ceiling of the instrument. The optimal benchmark ranges were determined as: non−/sub-clinical (0–14), mild (15–21), moderate (22–34), severe (35–50). Similar benchmarks were present across sex and age groups, and their accuracy was adequate in both a holdout sample and an independent sample of OCD patients from China (n = 78). Limitations and implications for the use of the Y-BOCS-II in clinical and research settings are discussed.