A comparison of latent class, latent trait, and factor mixture models of DSM-IV borderline personality disorder criteria in a community setting: implications for DSM-5.

With the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) scheduled for publication in 2013, researchers continue to debate the optimal classification of borderline personality disorder (BPD). Much of the discussion has focused on the relative merits of dimensional versus categorical classification schemes for BPD. Advances in statistical technologies have made it possible to adjudicate between continuous and categorical models of BPD using quantitative methods, yet no prior studies have attempted such a comparison. The current study directly compares the fit of dimensional, categorical, and hybrid models of BPD in a large community sample (N = 700) of young adults at risk for psychopathology due to elevated rates of maternal depression. BPD symptoms were assessed using the Structured Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II). Latent class, latent trait, and factor mixture models of SCID-II symptoms were estimated, and a latent trait model provided superior fit to the data, supporting a dimensional conceptualization of borderline pathology. The nosological implications of these results are discussed with respect to a "hybrid" model of BPD diagnosis currently under consideration for DSM-5.