Background: There is a dearth of placebo-controlled studies of cognitive behavior therapy (CBT) of depression and the largest such study, by Elkin et al. (Arch. Gen. Psychiatry 46 (1989) 971-982), failed to find a significant difference between CBT and a clinical management plus placebo condition.
Methods: The outcomes of two consecutive cohorts of out-patients with major depressive disorder, treated with either CBT (n=90) or a nonspecific control condition (support-counseling-placebo; SCP: n=100), were compared. Although the principal comparisons between the CBT and SCP conditions were delimited to the first 4 weeks of treatment, a secondary set of analyses addressed the subset of 16 patients who received 12 additional weeks of supportive therapy.
Results: A consistent pattern of statistically and clinically significant differences favoring CBT over SCP was found in both weeks 4 and 16.
Limitations: Interpretation of these findings are subject to several potential confounds, including the non-randomized nature of the groups and the greater amount of therapeutic contact during the first 4 weeks of CBT.
Conclusions: While these results do not lessen the need for additional prospective studies, our findings do suggest that CBT has therapeutic effects beyond those attributable to placebo-expectancy and other nonspecific factors.