Study designs and outcomes in antidepressant clinical trials.

Abstract

Clinical trials of antidepressants are difficult to design and conduct. In fact, more than half of all recent clinical trials of commonly used antidepressants failed to show statistical superiority for the drug over placebo. This is not necessarily because of the ineffectiveness of the antidepressant, but rather because of an increased response to placebo. Factors that may contribute to these findings remain elusive. Using data from U.S. Food and Drug Administration (FDA) Summary Basis of Approval (SBA) reports and from studies conducted by our group, we reviewed methodological factors used in clinical trials of antidepressants. The 2 most notable factors affecting positive trials are (1) the inclusion of patients with more severe depression, and (2) the use of a flexible-dose design; these may yield results identifying true antidepressant-placebo differences. Severely ill patients with depression respond well to antidepressants but poorly to placebo. Flexible dosing paradoxically reduces the response to placebo without augmenting the response to the antidepressant. All of these findings suggest that the use of placebo is mandatory when assessing new antidepressants.