Obsessive-compulsive disorder subgroups: a symptom-based clustering approach.

Although obsessive-compulsive disorder (OCD) is often considered a heterogeneous condition, there is no generally accepted subtype typology. Cluster analysis was used to identify definitive symptom-based groupings of 106 OCD patients. A stable cluster solution was achieved and five patient subgroups were identified based on their pattern of symptoms on the Yale-Brown (Y-BOCS) symptom checklist: harming, hoarding, contamination, certainty and obsessionals. The five subgroups were characterized by dominant symptom patterns and significant secondary concerns reflecting the symptom heterorgenaity often seen in the clinical presentation of obsessional patents. Between cluster differences on multiple symptom measures were evaluated and several meaningful differences were identified. Cluster analytic procedures may prove to be a useful tool for identifying a functional taxonomy of OCD subtypes.