Equipoise should be amended, not abandoned.

BACKGROUND: Benjamin Freedman has argued in 1987 that before a controlled trial is started, there should be 'genuine uncertainty in the expert medical community about the preferred treatment'. Freedman's definition of the concept is widespread in clinical research, but has been controversial since its start. Over the past decade, the equipoise controversy has become increasingly complex.
PURPOSE: This article aims: (1) to identify and clarify the main points of contention in the equipoise controversy, and (2) to reconcile the opposing views by pointing at areas of overlap between proponents and opponents in the equipoise debate.
METHODS: We analyze the positions of the leading authors in the equipoise debate in the past decade.
RESULTS: There is substantial overlap between the opponents and proponents of equipoise. Both should be able to accept the following answers to points of contention in the debate: (1) the therapeutic obligation can remain the basis for equipoise as long as it is conceived as an obligation to provide participants with competent care, (2) equipoise is grounded in a competent care and an epistemological dilemma, (3) equipoise does not as a rule prohibit placebo-controlled trials when proven effective treatment exists, (4) patient equipoise and individual physician equipoise are irrelevant, and (5) having to stop a trial is not always equivalent to disturbing equipoise. Clarification of these points of contention leads to a sharpened definition of equipoise: 'a state of genuine agnosticism or conflict in the expert medical community about the net preferred medically established procedure for the condition under study'. This definition asks of physician-researchers and members of IRBs to meet two conditions: (1) to genuinely evaluate to what extent a proposed randomized clinical trial solves a state of agnosticism or a knowledge conflict in the expert medical community and (2) to respect the standard of competent care, meaning that they consider whether the regular clinical standard from an all-things considered perspective is also the preferred standard in the research context. Equipoise is a specification of two general ethical principles for clinical research: scientific validity and a favorable risk benefit ratio. As a specification equipoise adds substance to these principles since they do not explicitly ask for the two conditions. Equipoise is a prima facie obligation rather than a morally authoritative principle for determining the acceptability of clinical trials. It needs to be balanced against other norms for clinical research. Violation of equipoise is therefore not always unethical. LIMITATIONS: This study is limited to the context of randomized clinical trials.
CONCLUSIONS: There is no decisive reason to give up on the equipoise requirement.