Major depressive disorder and borderline personality disorder revisited: longitudinal interactions.

BACKGROUND This report investigates the longitudinal association of changes in major depressive disorder (MDD) and borderline personality disorder. METHOD A DSM-IV-diagnosed sample of 161 patients with borderline personality disorder who have been followed with repeated measures at 6, 12, 24, and 36 months are investigated to see whether those with co-occurring MDD differ at baseline and in their course. Proportional hazard regression and cross-lagged panel analyses are used to demonstrate whether changes in the course of either disorder have predictable effects on the course of the other. RESULTS The rate of remissions of borderline personality disorder was not affected by whether patients had co-occurring MDD. The rate of MDD remissions was significantly reduced by co-occurring borderline personality disorder. Both regression analyses and panel analyses indicated that improvements in borderline personality disorder were often followed by improvements in MDD but that improvements in MDD were not followed by improvements in borderline personality disorder. Five of the 9 borderline criteria, including those that most relate to affects, were particularly apt to remit prior to MDD remissions. CONCLUSIONS When borderline personality disorder and MDD co-occur, they can sometimes have independent courses, but more often improvements in MDD are predicted by prior improvements in borderline personality disorder. Clinicians should not ignore borderline personality disorder in hopes that treatment of MDD will be followed by improvement of borderline personality disorder.