How should efficacy be evaluated in randomized clinical trials of treatments for depression?

The present system of conducting studies of promising antidepressant therap ies has evolved through the collaborative efforts of government, industry, and academicians and is costly and inefficient. At least one third of the p ublished clinical trials of approved antidepressants are negative for effic acy, which can be partly explained by the clinical and neurobiological hete rogeneity of the depressive disorder and partly because of methodological i nadequacies. Unfortunately, too little attention is given to ensuring the r eliability of diagnoses and dependent measures, sample sizes are seldom lar ge enough to detect modest yet honestly significant differences, and too ma ny trials are pursued before dose-response characteristics are fully unders tood. At present, the only data beyond 1 year of treatment-and the only evi dence about protection against recurrent depression-come during postmarketi ng or phase 4 of the drug development process. Moreover, efficacy data for depressed children and adolescents, bipolar depression, psychotic depressio n, dysthymia, and frail or medically ill elderly patients are rarely availa ble at the time a drug is introduced. Thus, it is remarkable how little cli nicians know about a new antidepressant at the time it is first approved fo r general use. Within a research strategy, tactics that ensure reliability, encourage attention to adherence, and lessen attrition at the outset of a study will increase the power and design sensitivity of a particular trial. Additionally, the issues of research funding-including division of the res earch pie-and the relationship of the Food and Drug Administration and inve stigators to the pharmaceutical industry and the National Institute of Ment al Health need to be revisited. Finally, extension of a compound's patent l ife might be considered to expand the necessary postmarketing research. Thi s article describes the process of conducting the clinical trials that supp ort a New Drug Application, discusses issues in evaluating efficacy, and of fers suggestions for modifying and improving the drug development process s o that clinicians can better judge new drugs.