Refining evidence-based treatment for borderline personality disorder: An empirical examination

Background and Purpose: The psychosocial treatment of individuals with borderline personality disorder (BPD) can be exceptionally challenging with such treatment often provided by clinical social workers. Clinically, BPD is heterogeneous in terms of the range of emotional and behavioral difficulties including self-harming behavior, impulsivity, emotion dysregulation and violence. A better understanding of BPD subgroups could assist in developing modifications of existing therapies to better target differential client needs. Our previous work has shown significant co-morbidity of BPD with psychopathy, a condition characterized by lack of empathy, manipulativeness, impulsivity and antisocial behavior. As such, the current study systematically examines the underlying heterogeneity in BPD vis-¨¤-vis psychopathy in an attempt to provide guidance for refining existing evidence-based treatment approaches.

Methods: Participants consisted of a sample of 221 psychiatric patients meeting criteria for BPD recruited from inpatient units in three large urban areas. A well established measure of psychopathy, the 12-item Psychopathy Checklist-Screening Version (PCL-SV), was administered to subjects, and this data were used to identify qualitatively distinct classes of participants. The PCL-SV items are rated on a 3-point scale that assesses interpersonal, emotional and behavioral dimensions reflecting psychopathy. Finite mixture modeling was employed to identify subgroups. The empirical fit of statistical models was based on the Bayesian Information Criterion (BIC) and conditional bootstrapping procedures. A validation analysis of the final model was conducted by testing associations of class characteristics with demographic and clinical measures.

Results: Overall, the four class model provided the best statistical and conceptual fit to the data. Classes consisted of 1) impulsive/antisocial (38%, n =85), 2) low psychopathic (24%, n = 52), 3) interpersonally exploitive/narcissistic (20%, n = 43), and 4) psychopathic/antisocial (18%, n = 41). Classes did differ by gender with females comprising a significantly (69.2%, p< .01) greater proportion of class 2 and males a greater proportion in class 4 (65.9%, p<.01). One-way analysis of variance (ANOVA) with post-hoc comparisons revealed significant differences with strong effect sizes on external measures of agreeableness (¦Ç2 = .15), planning (¦Ç2 = .25), motor skills (¦Ç2 = .16), and verbal skills (¦Ç2 = .23). Classes also differed with respect to an interval measure of socioeconomic status (¦Ç2 = .15).

Implications for practice: Findings possess significant implications for practice and evidence-based treatment development. Identified classes shows distinctive characteristics that suggest specific needs that should be reflected in the treatment provided. For example, one of the most commonly provided treatments for BPD is dialectical behavioral therapy, which focuses on teaching patients skills for managing symptoms. Patients in Class 1 might benefit from an increased emphasis on learning impulse control skills and planning while, in contrast, Class 3 patients might benefit from learning how to be more empathic towards others to achieve self beneficial results. Individuals in Class 4 are most likely to be encountered in the criminal justice system and, thus, modifying existing treatments for provision in this environment is important. This study sheds light on how existing treatments may be refined to better meet the differential needs of the clients served.