Letters

Declaration of Helsinki should be strengthened

Equipoise is essential principle of human experimentation

All countries must have common standards for international research ethics

Authors' reply

Equipoise is essential principle of human experimentation

EDITORWe wish to join in the debate about the next revision of the Declaration of Helsinki and to address some of the arguments put forward by Rothman et al.¹

We agree with Rothman and Michels that equipoise ("the uncertainty principle"²) is an essential ingredient of an ethical experiment and that the declaration should be amended to say so. We recently argued that extraordinary care should be given to understand and protect this fundamental principle, on which nearly the entire system of human experimentation stands.³

Baum writes of "tensions between conduct of a trial and the autonomy of the individual."¹ This involves the notion that patients who participate in trials are asked to make a sacrifice for the good of others. This concern, however, is alleviated by explicitly invoking equipoise as the principle on which randomised controlled trials are based. The uncertainty principle states that a patient should be enrolled in a randomised controlled trial only if uncertainty about which of the trial treatments would benefit the patient most is so substantial that they are in equipoise or "indifferent" between treatment options. ^{4 5}

It follows that so long as we are substantially uncertain which treatment is superior, patients do not lose out prospectively and are not required to sacrifice themselves for the benefit of others.⁴ Thus, ethically, randomised controlled trials should be acceptable to both utilitarians (who seek to bring the greatest good to the greatest number of patients by ensuring scientifically robust results) and Kantians (who seek to protect and preserve patients' autonomy).⁴ The same principle applies to any randomised trial, whether it is placebo controlled or not; it is just that in placebo controlled trials we should be particularly vigilant about applying the uncertainty principle.³

In our opinion, the ethical dilemma expressed in Rothman et al's article is falseit has already been resolved. The question is now a technical one: how do we improve communication so that patients can really find out whether or not they are indifferent between treatment options?

R J Lilford, professor of research of public health and epidemiology. 
University of Birmingham, Birmingham B15 2TT helena.smith{at}doh.gsi.gov.uk

Benjamin Djulbegovic, associate professor of oncology and medicine. 
H Lee Moffitt Cancer Centre and Research Institute at the University of South Florida, Division of Blood and Bone Marrow Transplant, Tampa, FL 33612, USA

All countries must have common standards for international research ethics

EDITORThe debate over changes to the Declaration of Helsinki is disappointing.¹ It has in many ways muddied the waters instead of clearing the way to improving international research ethics standards.

What we see at the moment is a clear fault line between the American and British medical associationsboth of which support lower standards of care for people living in developing countriesand continental European, Latin American, and some Asian medical associations, which reject such a double standard. In the United Kingdom the Wellcome Foundation is sponsoring a research project designed to bolster the case of the American and British medical associations but declined to sponsor a project critical of this campaign.

One of the crucial questions that have to be faced is: Do we want to see more clinical research undertaken whose primary objectives are solutions to economic rather than medical problems? If the answer is yes it is only logical to demand lower standards of care, because they are cheaper. If, on the other hand, we are serious about tackling the health problems of people in developing countries (this includes access to affordable drugs) surely we should question the economic frameworks that give rise to the purported necessity to develop cheaper drugs. Anglo-Saxon pragmatism, in this case, readily accepts the economic frameworks and tries to make the best out of the situation.

Is it unreasonable, then, to ask for a bit more honesty in the American and British campaign? These countries ought to state unequivocally that they think it ethically acceptable that people in developing countries die as a consequence of HIV infection acquired during preventive HIV vaccine trials. This is what the lower standards of care that both associations deem ethically acceptable will mean for the impoverished and otherwise vulnerable subjects in these trials.² Reasons for this position can be found, but I find them unconvincing. A first step to improve the debate could surely be to be frank about what one does and doesn't consider acceptable with regard to, for instance, standards of care in preventive vaccine trials.

We should also be concerned about attempts to reach what will be called an international consensus on this matter. International research ethics meetings take place all over the world, but often scholars and treatment access activists based in developing countries can go only if they know a generous Western sponsor who pays for their airfare and accommodation. This itself renders these meetings unrepresentative.

Udo Schuklenk, professor of bioethics. 
University of the Witwatersrand, Faculty of Health Sciences, Parktown 2193, Johannesburg, South Africa bioethic{at}chiron.wits.ac.za

Authors' reply

EDITORLilford and Djulbegovic push the edge of the equipoise envelope by implying that patients rather than their doctors should be the ones who are in equipoise. This is a noble but perhaps impractical goal, because it requires more knowledge by patients than we can expect, even with improved communication. Even informing patients sufficiently for them to give meaningful consent has remained unsatisfactory.¹ Meanwhile, the United States Food and Drug Administration does not recognise equipoise as an ethical requirement. ^{2 3} We would be pleased with any mention of an equipoise requirement in the Declaration of Helsinki, be it of patients or doctors, on the individual or group level.

Defenders of local standards of care instead of a global standard for comparisons in medical experiments will undoubtedly be troubled by Schuklenk's incisive comment. We believe, as he does, that global disparities in the standard of care only become hardened when inferior treatment is accepted as a treatment option in a medical experiment. His letter gives an interesting insight into the premise underlying the use of a local standard.

Kenneth J Rothman, professor. 
Department of Epidemiology and Biostatistics, Boston University School of Public Health, Boston University Medical Center, Boston, MA 02118-2526, USA KRothman{at}bu.edu

Karin B Michels, assistant professor. 
Harvard Medical School Obstetrics and Gynecology Epidemiology Center, Brigham and Women's Hospital, Boston, MA 02115, USA