Abstract
Traditional personality disorders (PDs; e.g., Diagnostic and Statistical Manual of Mental Disorders, fifth edition [DSM-5] Section II PDs), as well as dimensional traits (e.g., alternative model for PD [AMPD]), offer unique advantages in personality pathology assessment. However, very little is known about how these systems compare in predicting observable behavior. This study compares self-report ratings of PD symptoms (i.e., Structured Clinical Interview for DSM-IV PD) with self-reports of AMPD traits (i.e., Personality Inventory for DSM-5) in predicting clinical outcomes, 1 year later, via three different methods: (a) naturalistically observed psychosocial functioning (i.e., electronically activated recorder [EAR]), (b) informant-reported interpersonal functioning (i.e., Inventory of Interpersonal Problems-32), and (c) self-reported suicidality (SI), depression, anxiety, and substance use symptoms (i.e., Psychiatric Diagnostic Screening Questionnaire). Data were analyzed from 72 individuals in current or recent psychiatric treatment meeting diagnosis for at least one PD. Results showed that DSM Section II PD and AMPD ratings yielded meaningful and comparable predictions of naturalistically observed EAR variables and informant-rated interpersonal functioning. The AMPD appeared to offer slight advantages in the prediction of EAR-observed negative affect, hostile words, and informant-rated interpersonal functioning, with clearer advantages at the facet level. Overall, these results provide tentative evidence that both DSM Section II PD and AMPD systems show meaningful links with clinical outcomes measured via multiple methods 1 year later, but with clearer advantages for the AMPD at the facet level. Moreover, results show that the EAR is a viable method for capturing naturalistically observed clinically meaningful, in vivo behavior of individuals exhibiting maladaptive personality patterns. (PsycInfo Database Record (c) 2024 APA, all rights reserved).