A meta-analysis of psychodynamic treatments for borderline and cluster C personality disorders

Personality disorders (PD) carry high psychosocial dysfunction and are associated with treatment resistance in nonspecialized care. Psychodynamic therapies (PDT) are often used to treat PDs, but there has never been a systematic meta-analysis of PDT trials for PD. To evaluate the evidence base for PDTs for PDs across multiple outcome domain, a systematic search for PDT for PD trials was conducted through PubMed and PsycINFO. Sixteen trials were identified, comprising 19 dynamic, 8 active, and 9 control groups predominantly reflecting treatment of borderline and mixed Cluster C PDs, and a random effects meta-analysis was undertaken. PDTs were superior to controls in improving core PD symptoms (g = -0.63; 95% confidence interval [CI; -0.87, -0.41]), suicidality (g = -0.79, p = .02; 95% CI [-1.38, -0.20]), general psychiatric symptoms (g = -0.47; 95% CI [-0.69, -0.25]), and functioning (g = -0.66; 95% CI [-1.01, -0.32]), but not for interpersonal problems due to heterogeneity (g=-1.25; 95% CI [-3.22, 0.71]). Outcomes for PDTs were not different from other active treatments in core PD (g = 0.05; 95% CI [-0.25, 0.35]) or other symptoms. This pattern continued into posttreatment follow-up (average 14 months). Study quality was generally rated as adequate and was unrelated to outcomes. Compared with other treatments, PDTs do not have different acute effects and are superior to controls, although only trials treating BPD employed active controls and non-BPD trials were of lower quality. Underresearched areas include narcissistic PD, specific Cluster C disorders, and personality pathology as a continuous construct.