Comparing the DSM-5 categorical model of personality disorders and the alternative model of personality disorders regarding clinician judgments of risk and outcome.

Abstract

The goal of this study was to compare the predictive validity of the alternative model for personality disorders (AMPD) versus the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) Section II categorical model regarding clinician judgments about mental health outcomes. To do so, we instructed a national sample of 136 mental health professionals to provide clinical judgments on a random subset of four (out of a possible 12) case vignettes. For each case, they made a variety of diagnostic judgments corresponding to each model, as well as clinical outcome judgments (e.g., prognosis). Our analyses included hierarchical and individual regressions to compare the predictive value of each diagnostic system toward these clinical outcome judgments. We found that the AMPD predictors consistently added unique variance beyond the Section II predictors, whereas the Section II predictors were rarely incremental above the AMPD. Further, the AMPD judgments predicted outcome judgments very consistently (98.3% of regressions) compared to the Section II predictors (70% of regressions), and the single Criterion A judgment (level of personality functioning) was the strongest overall predictor. Finally, the categorical borderline personality disorder diagnoses from the two systems performed similarly in predicting clinical outcomes and agreed in 79% of cases. We interpreted our results to suggest that the AMPD is at least as effective, and by some measures more effective, than the DSM-5 categorical model at predicting clinician's judgment of outcomes in clinical cases. We conclude by discussing the value of this evidence in relation to the broader AMPD literature, as well as possible paths forward for the diagnosis of personality disorders.