Identification of belief-based subtypes in obsessive-compulsive disorder using latent profile analysis

Abstract

We evaluated dysfunctional belief heterogeneity in a large sample (N = 547) of adults with obsessive-compulsive disorder (OCD) because of the central role of dysfunctional beliefs in cognitive theories of OCD and the prior identification of large OCD patient subgroups who failed to endorse these beliefs. We used a methodological approach not previously applied, latent profile analysis (LPA) because of its utility for objectively testing heterogeneity and avoiding the identification of spurious subgroups. Significant dysfunctional belief heterogeneity was found and congruent with prior studies, a two-subgroup model was supported. A large, low-beliefs subgroup was identified that included over half of our sample and that reported dysfunctional belief levels equivalent to nonclinical groups or individuals with anxiety disorders but not OCD. We tested between subgroups differences on measures of OCD symptom severity and on measures of OCD-related cognition to further elucidate subgroup characteristics. Implications for cognitive theories and cognitive therapy for OCD are discussed.