Ethical research requires ethical researchers

A new report suggests ways of preventing the exploitation of participants in clinical trials in developing countries. Its impact will rest on the determination with which its proposals are put into practice, and the resources allocated to this task.

Five years ago, two leading medical journals — the New England Journal of Medicine and The Lancet — stirred up a heated debate in the medical community by questioning the way in which US researchers tested an anti-HIV drug on pregnant mothers in Africa.

The tests were intended to evaluate the effectiveness of relatively short treatment with the drug zidovudine. What incensed the editors of both journals was that the pregnant mothers being used as a control group were not being offered the 'best available treatment' available anywhere in the world, a standard requirement of such clinical trials if carried out in developed countries, but were merely offered a placebo.

The two journals argued that the researchers were guilty of double standards, as they were adopting a standard of care that would have been ethically unacceptable in their home country. The accusation resonated strongly with those who had, with good reason, become increasingly alarmed at the potential for exploitation embedded in the growing number of clinical trials being conducted in developing countries.

But it also set alarm bells ringing within the research community. Many researchers, from developed and developing countries alike, who had spent their scientific careers seeking new treatments for diseases endemic in the latter, argued that it was often impractical and unrealistic to provide the best care available elsewhere in the world to those recruited to participate in clinical trials.

Justified concern

Three years later, the US National Bioethics Advisory Committee concluded, after a detailed study of the issue, that such concerns were justified, proposing that members of control groups be offered "an established effective treatment". Now Britain's Nuffield Council on Bioethics has come out with a similar conclusion.

In a report published in London last week, an international working party set up by the council states that a universal standard of care should be offered to members of a control group "where appropriate" (see Best treatment 'not always practical'). In other circumstances, however, the working party recommends that "the minimum that should be offered is the best intervention currently available as part of the national public health system".

The report also recommends that all those involved in such trials should be required to provide personal consent to their participation, and discusses the extent to which participants may or may not benefit from the outcome of the research. In all its suggestions, the working party strives to reach a considered balance between idealism and pragmatism.

On the question of consent, for example, the working party firmly states that researchers "must not enrol" prospective participants who have indicated that they, as individuals, do not want to take part — regardless of any indications to the contrary that may have been received from other community or family members. Furthermore, researchers "have a duty to facilitate their non-participation". Indeed, the report eschews the conventional term "informed consent" for the potentially more demanding concept of "genuine consent".

Similarly, the working party disagreed with the view that participants should only be recruited into a trial if they are likely to be the direct beneficiaries of any treatment that results. Its report points out that, in practice, this raises a range of complex issues. For example, sponsors of clinical trials, who are often foreign research funding agencies, are rarely in a position to agree to open-ended commitments once the research is completed. Nevertheless, the report does recommend that such sponsors ensure that the development of local expertise in healthcare "is an integral component of research proposals".

Need for good faith

The danger of proposing a pragmatic approach, of course, is that it relies on both the good will and the good faith of all those involved in both carrying out and monitoring the research. Where either of these is lacking, proposals for action that rely heavily on local input — for example, in communicating decisions about consent, or in providing acceptable treatment for members of control groups — however desirable in theory, remain open to exploitation and misuse.

Some still argue that the way to prevent this is to insist on the application of universal principles relevant to the conduct of research. The Nuffield working party correctly argues against this, insisting that judgements on ethically sound courses of action must include consideration of local cultural values and traditions. It also performs an invaluable service by listing in detail the components that all ethical decisions relating to clinical trials should take into account.

The question that it frequently leaves hanging however, is who, in practice, will judge whether a form of treatment being offered to control groups is "appropriate", whether an individual's consent can be considered "genuine", or even whether efforts to provide participants with subsequent access to any new treatment resulting from the trials can be considered adequate. And who will decide what criteria should be used in reaching such decisions?

Here again there are no easy, or indeed universally applicable, answers. One thing, however, is clear: countries face a higher chance of meeting the norms of ethical behaviour outlined in the Nuffield working party's report if they make a firm political commitment to do so.

Lack of resources

It is also clear that the chances of success will depend on the availability of the necessary resources, both financial and in terms of human capital. This point is made in an opinion article by Abdallah Daar and Peter Singer (see Human capital is key to research ethics). The authors argue that the most useful contribution developed countries can make to raising ethical standards in clinical trials across the world is by assisting in the training of those who will implement national rules and review procedures.

This task is a daunting one. In particular, it will require a major commitment to training researchers in the ethical issues raised by their work. It also raises its own ethical issues that do not receive a significant amount of attention in the Nuffield report. For example, should the members of ethics committes be paid, and if so, who should set the rates, at what level, and using what criteria?

But, as Daar and Singer correctly point out, and the Nuffield working party itself acknowledges, any system of ethical research and ethical monitoring, however all-embracing the rules under which it operates, will only be as effective as the individuals who carry it out.

© SciDev.Net 2002