A Systematic Review of Comparative Efficacy of Treatments and Controls for Depression

Arif Khan,Irving Kirsch,Walter A Brown,Pesach Lichtenberg,James Faucett,Christian Holscher

doi:10.1371/journal.pone.0041778

Abstract

BackgroundAlthough previous meta-analyses have examined effects of antidepressants, psychotherapy, and alternative therapies for depression, the efficacy of these treatments alone and in combination has not been systematically compared. We hypothesized that the differences between approved depression treatments and controls would be small.Methods and FindingsThe authors first reviewed data from Food and Drug Administration Summary Basis of Approval reports of 62 pivotal antidepressant trials consisting of data from 13,802 depressed patients. This was followed by a systematic review of data from 115 published trials evaluating efficacy of psychotherapies and alternative therapies for depression. The published depression trials consisted of 10,310 depressed patients. We assessed the percentage symptom reduction experienced by the patients based on treatment assignment. Overall, antidepressants led to greater symptom reduction compared to placebo among both unpublished FDA data and published trials (F = 38.5, df = 239, p<0.001). In the published trials we noted that the magnitude of symptom reduction with active depression treatments compared to controls was significantly larger when raters evaluating treatment effects were un-blinded compared to the trials with blinded raters (F = 2.17, df = 313, p<0.05). In the blinded trials, the combination of antidepressants and psychotherapy provided a slight advantage over antidepressants (p = 0.027) and psychotherapy (p = 0.022) alone. The magnitude of symptom reduction was greater with psychotherapies compared to placebo (p = 0.019), treatment-as-usual (p = 0.012) and waiting-list (p<0.001). Differences were not seen with psychotherapy compared to antidepressants, alternative therapies or active intervention controls.ConclusionsIn conclusion, the combination of psychotherapy and antidepressants for depression may provide a slight advantage whereas antidepressants alone and psychotherapy alone are not significantly different from alternative therapies or active intervention controls. These data suggest that type of treatment offered is less important than getting depressed patients involved in an active therapeutic program. Future research should consider whether certain patient profiles might justify a specific treatment modality.

Highlights

A number of recent articles have emphasized the inability of antidepressant medication to consistently demonstrate superiority to placebo pills [1,2,3,4]
We evaluated if the precursors of Diagnostic and Statistical Manual (DSM)-III, notably Research Diagnostic Criteria (RDC) [24] or Feighner Criteria [25] were used during diagnosis
We began by reviewing several meta-analyses designed to evaluate efficacy outcomes between psychotherapy and other treatments and controls for depression including other psychotherapies, alternative therapies, combination therapies, antidepressants, and placebos or active intervention controls [8,16,26,27,28,29]. During this search we identified a database of 243 psychotherapy trials compiled by Dr Pim Cuijpers and his depression research group at www.evidencebasedpsychotherapies. org [30]

Summary

Introduction

A number of recent articles have emphasized the inability of antidepressant medication to consistently demonstrate superiority to placebo pills [1,2,3,4]. Half of clinical trials fail to differentiate active treatments from controls, and mean differences between drug and placebo on the Hamilton Rating Scale for Depression are small [2,5]. This phenomenon has sparked considerable concern and criticism from the popular media, clinicians and researchers [6,7]. We hypothesized that the differences between approved depression treatments and controls would be small

Objectives

Methods

Results

Discussion

Conclusion