Therapist adherence, patient alliance, and depression change in the NIMH Treatment for Depression Collaborative Research Program.

Using data from the National Institute of Mental Health Treatment of Depression Collaborative Research Program (NIMH TDCRP), this dissertation examined the relationship between the patient alliance (Vanderbilt Therapeutic Alliance Scale patient factor) and therapist adherence to five subscales of the Collaborative Study Psychotherapy Rating Scale: Cognitive Behavior Therapy (CBT), Interpersonal Therapy (IPT), Clinical Management (CM), Facilitative Conditions (FC), and Explicit Directiveness (ED) and their influence on depression change as measured by the Beck Depression Inventory. Analyses were conducted using path models examining the interrelationship between early and later alliance and adherence as predictors of depression change for the full TDCRP sample (n = 239) and for each treatment comprising the TDCRP: CBT (n = 59), IPT (n = 61), imipramine plus clinical management (IMI-CM; n = 57), and placebo plus clinical management (PLA-CM; n = 62). The results indicate that, in each of the treatments, early patient alliance predicted later patient alliance, and later patient alliance predicted depression change. Early therapist adherence rarely predicted later therapist adherence and later therapist adherence rarely predicted depression change with the following exceptions. Later IPT adherence predicted greater reductions in depression in IPT and later CBT adherence predicted greater reductions in depression in both IPT and in PLA-CM. Across all the treatments, there was a positive relationship between both IPT and FC adherence and patient alliance, and a negative relationship between both ED and CM adherence and patient alliance. In each treatment group, the relationship between adherence and the patient alliance was different. In IPT, greater ED predicted reduced patient alliance. In CBT and PLA-CM, greater CM adherence predicted reduced patient alliance. Early patient alliance predicted: (a) greater later FC, IPT, and CBT adherence in CBT; (b) greater IPT adherence in IPT; and (c) greater FC adherence in PLA-CM. These results have important implications for psychotherapy training and clinical practice. Particularly relevant are the findings that non-target techniques had a positive influence on patient alliance and depression change, and that therapist directiveness negatively influenced the patient alliance in IPT. In sum, this study highlights the complex interrelationship of relational and technical dimensions of psychotherapy.