Back

Profiles of Distress in OCD: A Cluster Analysis Exploring Associations with Global and Facet-Specific Anhedonia

Zaboski, B. A.; Jones, K.; Mattera, E.; Pittenger, C.; Pushkarskaya, H.

2025-12-05 psychiatry and clinical psychology
10.64898/2025.12.01.25341370 medRxiv
Show abstract

BackgroundObsessive-compulsive disorder (OCD) is clinically heterogeneous. Its relationship with anhedonia--a diminished capacity for pleasure--remains unclear, with conflicting findings in the literature. This study identified symptom profiles in OCD to characterize their association with anhedonia. MethodsA k-means cluster analysis was performed on 227 adults with OCD using ten clinical variables, excluding direct measures of anhedonia. The resulting clusters were validated against unidimensional severity measures to confirm their unique construct validity and were then compared on external anhedonia measures. ResultsThree distinct and robust profiles emerged: "High Global Distress and OCD" (HGD-OCD; n = 62), "Intermediate Overall Severity" (IOS; n = 100), and a "Low Symptom Profile" (LSP; n = 65). Validation analyses confirmed these profiles represent a multidimensional construct of distress not reducible to depression or OCD severity alone. Anhedonia was strongly associated with these profiles; its severity was highest in the HGD-OCD cluster, intermediate in the IOS cluster, and virtually absent in the LSP cluster. Item-level analyses revealed unique anhedonia signatures, with the IOS profile showing a complex pattern of both impaired and preserved hedonic domains. ConclusionA data-driven approach effectively stratifies the heterogeneity of distress in OCD. These holistic profiles offer a clinically meaningful alternative to unidimensional measures by revealing distinct, facet-specific manifestations of anhedonia. This framework provides a crucial step toward developing more targeted, personalized interventions for anhedonia in OCD.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.