Alternatives to equipoise

Steve Simon


Stephen Senn wrote an interesting article a couple of years ago that provides an intriguing alternative to the concept of equipoise. I wrote about equipoise in a web page discussing the ethics of placebo controlled trials. The concept of equipoise actually extends well beyond placebo controlled trials, but it is in those trials that the concept is most easily understood.

Dr. Senn argues that society has a vested interest in studying new drugs and therapies in a rigorous fashion, and asks that patients sacrifice some degree of choice in order to conduct these studies efficiently. In particular, some untested or unproven drugs might be available only through randomized trials until enough information is accumulated either to make the drug available to the general public or to reject the drug out of hand and study it no further.

Since society has mandated that certain drugs are only available through randomized trials, the physician only has to ask the question, is the expected benefit from participating in the trial superior to the benefit of the best available therapy outside the trial?

This is a less rigorous standard than equipoise which would mandate that not only would the expected benefit of participating in the trial have to be better, but that the two arms being randomized in the trial had to roughly comparable. Dr. Senn argues that a physician who would refuse to enroll a patient in a study without equipoise, but with the expected benefit being superior to the best available therapy outside the trial is automatically condemning 100% of his patients to an inferior treatment.

Good food for thought. In discussions with researchers planning studies with placebos, I had argued a similar standard, but not quite the same. In order to get volunteers to agree to be in your study (admittedly a more pragmatic concern than the ethical concerns discussed above), you need to make sure that both arms of the study are superior to the standard of care your patients would get outside the trial.

One question still remains. Is it ethical for society to restrict the use of promising new drugs to the confines of a clinical trial? Dr. Senn argues that it is, and cites an approach developed by John Rawls.

An application of the Rawlsian theory of justice to the problem of drug regulation and clinical trials might be as follows. We are to consider in the original position whether we wish to join one of two societies. In the first there is no regulation of medicines. A physician is simply placed under the obligation of providing the patient with that treatment (s)he believes most efficacious. No controlled clinical trials are possible. As a consequence the study of treatment is less efficient, medical progress is slower and less certain and lives are lost. In the second sort of society, treatments are not registered until there is a reasonable consensus that they are efficacious. This consensus is achieved through clinical trials. A side effect of this is that physicians are not always free to give what they consider best treatment. The only opportunity to receive such treatment may be to enter a clinical trial in which its allocation will not be certain. On the other hand, in the second society, standard care is almost certain to be better than in the first. Which society would we choose?

Ethical considerations concerning treatment allocation in drug development trials. S. Senn. Stat Methods Med Res 2002: 11(5); 403-11. [Medline]

You can find an earlier version of this page on my original website.