Are antidepressants overrated? A review of methodological problems in antidepressant trials

Abstract

There are no signs that the rapidly escalating use of antidepressants is reducing the burden of depressive disorders. This may be due to the fact that the evidence base for antidepressants is weaker than is commonly assumed. There are a number of methodological problems that may bias the results of clinical trials. Unblinding may inflate the response of people taking an active drug when compared with those taking an inert placebo. Modern measurement techniques may exaggerate the benefit of drug treatment. Excluding some randomized subjects from analysis may inflate the apparent effect of antidepressant drugs and publication bias means that published studies may not represent an accurate picture of the effects of treatment. In trials of long-term treatment discontinuation-related effects may masquerade as clinical efficacy. A brief survey of evidence from controlled trials does not present a consistently positive picture. Two of the largest and most reputable trials found only negligible differences between tricyclic antidepressants and placebo. The evidence on whether antidepressants are specific treatments is also inconclusive. Many other drugs not classed as antidepressants have shown positive effects in depression in controlled clinical trials. It is suggested that the interests of the pharmaceutical industry and the psychiatric profession have helped to establish the notion of the efficacy and specificity of antidepressant drugs.