Back

73 Views

Assessment

Quantifying diagnostic heterogeneity in melancholic and atypical depressive specifiers: A NESARC Analysis

(PS8-B50) Quantifying Diagnostic Heterogeneity in Melancholic and Atypical Depressive Specifiers: A NESARC Analysis

Saturday, November 19, 2022

12:00 PM – 1:00 PM EST

Location: Westside Ballroom, 5th Floor

Author(s)

JB

John F. Buss, B.S.

PhD Student
Indiana University
Bloomington, Indiana

Co-Author(s)

Lorenzo Lorenzo-Luaces, Ph.D.

Assistant Professor
Indiana University
Bloomington, Indiana

Background: Major depressive episodes (MDEs), like other diagnostic categories in the Diagnostic and Statistical Manual (DSM), are highly heterogeneous, which has led to alternative classification schemes like the Hierarchical Taxonomy of Psychopathology (HiTOP). Although DSM-MDE specifiers are supposed to reduce heterogeneity in symptom presentations, recent studies suggest that the atypical and melancholic features specifiers do not reduce heterogeneity and may increase it. We attempt to replicate these findings and extend them to duration-based subtypes using a novel measure of symptom heterogeneity: distance in Euclidean space.

Methods: We used data from the National Epidemiological Survey on Alcohol and Related Conditions (NESARC Waves I; N = 5,749) and Sequenced Treatment Alternatives to Relieve Depression (STAR*D; N = 2,496). We computed Hamming and Manhattan distances, measures from information theory, to operationalize heterogeneity in symptom endorsement. Additionally, we explored if diagnostic heterogeneity differed by the episode's duration.

Results: In neither dataset was there evidence that the melancholic or atypical specifiers reduced heterogeneity. All distance metrics were almost identical when focusing only on the core depression symptoms. Subtyping by duration does not appear to affect the diagnostic heterogeneity either.

Conclusion: Heterogeneity, as operationalized by measures of distance in Euclidean space, does not appear to change based on the subtyping scheme proposed by the DSM. However, if symptom combinations are valid measures of heterogeneity, new proposals to diagnosing should provide evidence that they can reduce heterogeneity.