Comparative effectiveness clinical trials in psychiatry: superiority, noninferiority, and the role of active comparators

Abstract

The Agency for Healthcare Research and Quality, part of the US Department of Health and Human Services, has issued several Requests for Applications to conduct comparative effectiveness research (CER). Many of the applications will involve randomized controlled clinical trials that include an active comparator. The inclusion of an active comparator has implications for clinical trial design. Despite a common misperception, a clinical trial result of no significant difference between active treatment groups does not imply equivalence or noninferiority. A noninferiority trial, on the other hand, can directly test whether one active treatment group is noninferior to the other. For example, noninferiority of an inexpensive generic could be tested in comparison with a novel, more costly intervention. Although seldom used in psychiatry, noninferiority clinical trials could play a fundamental role in CER. Features of noninferiority and the nearly ubiquitous superiority designs are contrasted. The noninferiority margin is defined and its application and interpretation are discussed.

Conclusions: Evidence of noninferiority can only come from well-designed and conducted noninferiority CER. Sample sizes needed in noninferiority trials and in superiority trials that include an active comparator are substantially larger than those needed in trials that can utilize a placebo control in their scientific design. As a result, trials with active comparators are more costly, require longer recruitment duration, and expose more participants to the risks of an experiment than do trials in which the only comparator is placebo.