Off-study availability of investigational treatments

Abstract

We agree with Jerry Menikoff (Jan 4, p 63)1Menikoff J. The hidden alternative: getting investigational treatments off-study. Lancet. 361: 63–67.Google Scholar that the off-study availability of a randomised controlled trial's interventions should be disclosed to potential participants. However, we disagree with his harm-benefit analysis of the various treatment choices a patient can make. A patient might choose to receive a trial's experimental intervention off-study, but this decision is not necessarily rational. Patients face the dilemma of a choice between treatments under conditions of uncertainty. In the context of a randomised controlled trial, this uncertainty is captured in the concept of equipoise.2Lilford RJ Jackson J Equipoise and the ethics of randomization.J R Soc Med. 1995; 88: 552-559PubMed Google Scholar Uncertainty about the relative merits of comparator interventions in randomised controlled trials can be considered at three levels: the community of “expert” clinical practitioners and trialists, who propose a trial for its resolution; the individual practitioner, who must decide whether to participate in a trial, and whether to offer enrolment to particular Patients; and the patient, who must decide whether to accept an offer of enrolment.3Sackett DL Uncertainty about clinical equipoise.Can Med Assoc J. 2001; 164: 1831-1832Google Scholar If, after being informed about community uncertainty, the nature of the interventions, and what is resently known about their harms, a patient is also “maximally uncertain” (in equipoise), then enrolment is a rational option. In this situation of uncertainty, enrolment gives the patient the best odds (50:50) of getting the superior intervention, if one is subsequently shown to be better. A patient who discloses that his motivation for enrolling is to exploit the 50% chance of getting the comparator intervention he favours, for whatever reason, is not in equipoise, and does not understand an essential feature of the randomised controlled trial: that which of the comparator interventions is superior is not known. And, more importantly, potential participants need to understand that there is no evidence that the outcomes of randomised trials, in the aggregate, substantially (magnitude of efficacy) and consistently favour innovative over control interventions. Indeed, if this was shown to be the case, it would jeopardise the entire system of randomised trials. Thus, Menikoff's admonition that the consent form should perhaps state that “If you are considering enrolling in this study mainly to get the 50% chance at being assigned treatment X, you should be aware that you could instead be certain to get treatment X by obtaining it directly from a doctor, without participating in this study” is concerning. It will promote—in the minds of members of research ethics committees, physicians, and patients—a serious misconception of the role of uncertainty in the design of randomized trials, and for the ethical enrolment of patients. When clinical equipois4Freedman B Equipoise and the ethics of clinical research.N Engl J Med. 1987; 317: 141-145Crossref PubMed Scopus (1643) Google Scholar is present, randomised trials enable the concurrent achievement of two objectives: the acquisition of valuable knowledge, and an optimised treatment choice under conditions of uncertaintys.5Djulbegovic B Acknowledgment of uncertainty: a fundamental means to ensure scientific and ethical validity in research.Curr Oncol Rep. 2001; 3: 389-395Crossref PubMed Scopus (49) Google Scholar