Pragmatic problems with clinical equipoise.

It is widely accepted that if one is to follow the ethical tenets of clinical equipoise, phase III controlled clinical trials must be designed pragmatically, to measure effectiveness rather than efficacy. This choice of a pragmatic rather than an explanatory approach to phase III clinical trial design has a number of consequences, some of which may be considered problematic. These include changes in what the trial is expected to accomplish, the way treatments are defined, the selection of subjects, the ways in which treatments are compared, and the assessment of the results. One also may end up challenging the real-world expectation that scientific results will be replicated before they are considered valid. This article discusses the connection between clinical equipoise and pragmatic trials, contrasts explanatory with pragmatic trials, points to the differences in the ways in which trial data are analyzed and interpreted, and discusses the power of replication, one of the defining hallmarks of the scientific method. Viewing clinical equipoise through a consequentialist lens reveals a number of problems, many of which are attributable to equipoise's insistence on a pragmatic approach to trial architecture.