Interpersonal problems as predictors of alliance, symptomatic improvement and premature termination in treatment of depression

Background: Previous studies reported inconsistent findings regarding the association of interpersonal problems with therapy outcome. The current study investigates if interpersonal problems predict process and outcome of three different treatments for depression. Methods: The data originate from a randomized clinical trial comparing supportive-expressive psychotherapy, antidepressant medication and pill-placebo for treatment of depression. Interpersonal problems were used as predictors of alliance, symptomatic improvement and premature termination of treatment. Results: Interpersonal problems related to communion predicted better alliances, but slower symptomatic improvement. Low agency predicted slower symptomatic improvement in supportive-expressive psychotherapy, but not in the medication or placebo condition. Lower interpersonal distress was associated with an increased likelihood to terminate treatment prematurely. Limitations: The sample size did not allow the detection of small effects within the treatment groups. Conclusions: Interpersonal problems are influential for the treatment of depression, but parts of their effects depend on the type of treatment. (C) 2013 Elsevier B.V. All rights reserved,