A Comparison of the Associations of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Section II Personality Disorders and Section III Personality Domains With Clinical Dysfunction in a Psychiatric Patient Sample

• Published in: Personality disorders 2024-09, Vol.15 (5), p.322-331
• Main Authors: Bagby, R. Michael, Lau, Sharlane C. L., Watters, Carolyn A., Quilty, Lena C., Sellbom, Martin
• Format: Article
• Language: English
• Subjects: Adult; Diagnostic and Statistical Manual; Diagnostic and Statistical Manual of Mental Disorders; Female; Human; Humans; Male; Mental Disorders; Middle Aged; Outpatient; Personality Disorders; Personality Disorders - diagnosis; Predictability (Measurement); Psychiatric Patients; Young Adult
• ISSN: 1949-2715; 1949-2723; 1949-2723
• DOI: 10.1037/per0000687

In this study, we compare the incremental predictive capacities of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DSM-5-TR) Section II personality disorders (SII-PDs) with Section III trait domains of the Alternative Model of Personality Disorders (AMPD) in a psychiatric outpatient sample (N = 185). To this end, a series of hierarchical regression analyses was conducted in which the 10 SII-PDs and the five AMPD trait domains served as the predictor variables and five areas of clinical dysfunction as the criterion variables. Two models for each criterion were tested. In Model A, the 10 PDs were entered as a block, followed by the block entry of trait domains; in Model B, the block entry of these predictors was reversed. As the AMPD was designed to address the shortcomings of the SII-PDs, it was hypothesized that the AMPD trait domains would show greater predictive capacity vis-à-vis the latter by (a) explaining more overall variance for each criterion variables when entered first into the model versus when SII-PDs was entered first and (b) explaining more incremental variance than SII-PDs when block was entered second. These hypotheses were partially supported. Overall, the AMPD trait domains predicted more variance than SII-PDs and demonstrated better model fit and more predictive power for three of the criterion variables. Similarly, the AMPD domains predicted a significant but modest incremental increase in variance over that of the SII-PDs for three of the criterion variables. We conclude that more work needs to be done to improve the AMPD, particularly in the assessment of externalizing psychopathology as it relates to clinical dysfunction.