Ethics

Towards a consensual culture in the ethical review of research

Donald Chalmers and Philip Pettit,
on behalf of the Australian Health Ethics Committee

The Report of the Review of the Role and Functioning of Institutional Ethics Committees was submitted to the Minister for Health and Family Services in March 1996.¹ It recommended, among other things, that the Statement on Human Experimentation, issued under the name of the National Health and Medical Research Council (NHMRC) in 1992, should be revised. A similar recommendation was made in the report by Dr Margaret Allers in 1994 into the collection, manufacture and injection of human growth hormone.² The recommendation for a review of the Statement was approved by the Council in November 1996.

The Australian Health Ethics Committee, a Principal Committee of the NHMRC, had for some time been discussing various aspects of the Statement on Human Experimentation and independently decided that the Statement should now be revised. The Committee's first consideration was the tone that the Statement should set for the ethical review of research. This article expresses the Committee's views on this matter.

MJA 1998; 169: 79-82  

Introduction - The goal of ethical review - The path we should be taking - The path we are actually taking - Recommendations for the reform of ethical review - References - Footnote - Authors' details
Make a comment - Register to be notified of new articles by e-mail - Current contents list - More articles on Ethics

Introduction	The evolution of an ethic among those conducting research on humans has occurred in three stages. First, individual researchers developed a general sense of what was proper behaviour and what was improper behaviour in the pursuit of research goals - in particular, with regard to the treatment of research subjects. Then, professional research bodies (which first appeared in the last century) articulated codes of behaviour for their members. Finally, independent authorities - universities, hospitals, governments - institutionalised the ethical review of research projects that fell within their purview.3,4 The forces driving this evolution are complex; we draw attention to one that may easily escape notice. This is a social mechanism that we describe as the "controversy machine". Some research is pursued in an ethically dubious fashion, or with ethically deplorable results. The whistle is blown by an insider, or a newspaper or community group comes to learn by another means what is happening. There is a public scandal, leading to popular demand that something be done to make sure this never happens again. The professional bodies, or other authorities, respond to the demand by taking appropriate action, or at least action accepted as appropriate. Things quieten down, but then, perhaps years later, another scandal is uncovered and the process runs its course once more. The controversy machine played a major role earlier this century in the development of codes of research conduct; these codes were formulated in the first wave of reaction to various scandals in research on humans. It has also played an important part in the last 20 years or so in facilitating the introduction and then the gradual formalising of arrangements for the review of research proposals. With every subsequent scandal made public, ethical surveillance was made more rigorous (Box 1). The controversy machine has been at the origin of many fine achievements in social life, being associated in various countries with the development of a more or less decent civic and public world.9 However, the controversy machine does not invariably work for good. In some areas the responses that scandals elicit - and that serve best to satisfy public outrage - are not necessarily the responses that deal most effectively with the underlying problem. For example, the controversy machine may be responsible for the widespread failure of governments to deal humanely and effectively with criminal offenders. Scandals in the criminal justice area - the heinous crime, the offence under parole - tend to generate a get-tough reaction that satisfies public outrage without actually being very effective. Because the controversy machine has been responsible for generating our system of ethical review of research, we who are charged, at whatever level, with running that system should be clear about where exactly we want to get to and whether we are on course for getting there. We need to consider, first, the goal of ethical review; second, the path that we should be taking in pursuit of that goal; and, third, the extent to which we are cleaving to that path. Otherwise, we are in danger of being carried along, unthinkingly, by a process that may do more harm than good.
The goal of ethical review	One view of the goal of ethical review of human research would be that it is to reduce, as far as possible, the incidence of unethical research. The best way of realising that goal would be to prevent any research whatsoever occurring: no research, ergo no unethical research. But it is important to see that this cannot possibly be right, and no one in the business of ethical review has ever thought that we would have achieved our goal - as distinct from putting ourselves out of a job - if we got rid of research altogether. Any plausible system for ethical review of human research must have as its goal, not only that research should be ethical, but also that research should continue to be pursued. It must be designed, not as a form of ethical inquisition, but, rather, as a system for facilitating and encouraging ethically informed research on humans. There are two models for facilitating ethically informed research. The first is to identify certain ethical standards - inevitably, certain less than fully determinate standards - and, taking scientific value as given, to encourage those projects that in the committee's judgement satisfy the standards. The second is to make a judgement about how scientifically valuable the research is and to encourage research to the extent that (i) it is judged to be both scientifically worthwhile and ethically satisfactory, and (ii) considering the balance between these two aspects, it scores pretty well in the opinion of the committee. The first of these models is more attractive than the second.2 Ethics committees are not necessarily well equipped to determine the scientific quality of the research projects they consider. In fact, it is often quite hard to make a judgement on whether a project is worthwhile or not, and it is clearly wrong for a non-professional group to take on the burden of this judgement. Besides, it would be a source of reasonable resentment and alienation on the part of researchers if a committee that was meant to judge the ethical acceptability of research projects insisted also on making judgements of a methodological character. Of course, ethics committees may require an assurance that any project submitted is of scientific interest. They might, for example, insist on a prior assessment of the risks attendant on the research, as well as the potential benefits to be gained by it. And in difficult cases where special problems arise (e.g., when the research involves the use of placebos with patients who seem certain to die without the drug under trial) they may reasonably seek an assurance that the project really does hold out a firm promise of achieving the alleged benefits. But such assurance should always be sought outside, from an adequately informed authority; otherwise, the committee is going to look like an arbitrary power. The goal we should be pursuing, then, is an arrangement under which ethics committees obtain judgements of scientific merit from elsewhere and work to facilitate all those meritorious projects, and only those, that meet certain ethical standards. The ethical standards will require, among other things, that the subjects are voluntary and suitably informed participants in the enterprise; that the risk of harm to those subjects, or to others affected, is not inappropriate; and that considerations of confidentiality and privacy are reasonably satisfied. Such standards are often going to be difficult to interpret and apply, which is why we mentioned their not fully determinate character. One way in which they will be difficult to apply is that, in judging whether a risk of harm is appropriate, the committee will have to make an assessment of the importance of the benefits promised by the research, according to the scientific assessment: it will have to make a judgement, not on what the potential benefits are, but on how much significance to attach to them.
The path we should be taking	There are two broadly contrasting approaches that ethics committees might take in pursuing the goal of facilitating ethically informed research. The first is to encourage patterns of ethical self-evaluation among researchers, and to rely on this as the normal means of ensuring ethical conduct. Thus, while routinely overseeing all that happens, and implementing relevant procedures and being available for consultation about them, they would take an interventionist stance only in exceptional cases. The second approach is to adopt an inspectorial and adversarial role in relation to researchers, seeking to articulate an exact code and to make maximum efforts to police that code in every instance. Two facts argue strongly against going the adversarial way.9-11 No system of inspection and policing, no matter how draconian, could hope on its own to achieve a high level of ethical conduct among resistant researchers; there will always be opportunities for researchers to take shortcuts and bend the rules, if they are so inclined. An adversarial system of inspection and policing is likely to make researchers resistant to considerations of ethics; it is likely to put them offside, as they see themselves demeaned and distrusted by the agencies of ethical review. The better prospect for facilitating ethically informed research is the alternative, non-adversarial, approach. This begins from the assumption that most researchers, once they are made aware of ethical concerns, are disposed to take them seriously; only a small minority are likely to ignore them in their enthusiasm for their research goals. The strategy is to build on this strength by generating an awareness of ethical concerns in the research community at large and, without abdicating the position of ultimate authority in ethical matters, by displaying a posture of trust in that community. What would this posture of trust involve? A number of possible initiatives will illustrate the idea; the list offered is meant to be suggestive, not definitive. Invite relevant individuals and bodies to give advice on the ethical problems most prominent in their area, and negotiate about difficulties that individual researchers may find in dealing with their ethics committee. Delegate some degree of responsibility for ethical review to professional departments or agencies; for example, allowing them to review projects coming to the committee, and working towards a situation in which the committee is almost always happy with anything that the professionals approve. Communicate the reservations of the ethics committee to any researcher whose project seems to raise problems and seek to work out ways of getting around those problems, consistent with the aims of the project. Develop a common understanding with the relevant research communities about what is a reasonable period for ethical review and try systematically to complete the review of projects within that period. Seek to devise a system of monitoring that does not communicate a sense of being under surveillance and suspicion to researchers. Encourage the relevant professional bodies to assume responsibility for keeping their members informed about ethical issues and for organising discussion of those issues.
The path we are actually taking	These considerations about the goal of ethical review, and about the best means of pursuing it, are not entirely uncontroversial. Still, they are likely to attract a broad consensus among those concerned with the ethics of human research, in particular among those with some sense of the difficulties associated with regulation, and of the possibility of regulation becoming counterproductive. They lead us now to ask whether the path we are actually taking in ethical review - the path on which the controversy machine has set us - is likely to prove adversarial or non-adversarial in character. Unfortunately, it doesn't require much reflection to see that the controversy machine has left us with a system of ethical review that is in danger of becoming extremely adversarial. The fact that the system orginated in reaction to scandal, and is legitimated as a protection against further scandal, projects two assumptions of an adversarial character. Researchers are regarded as the source of ethical problems: they are cast in the role of potential offenders. And reviewers are regarded as those with the unique responsibility, and the unique power, to prevent those problems arising: they are cast in the role of protectors and police. Under this interpretation of the two sides - and it is not yet a reality, only a threatening scenario - researchers get to be seen as one-sidedly interested in research goals, reviewers as equally one-sidedly interested in ethical standards. The review process is presented as a struggle between those of a single scientific mind, who want to pursue their research ambitions at any cost, and those of a single ethical mind, who have to try to keep the researchers honest. There is no room left in the scenario for the possibility of the two sides coming to a common mind on relevant matters. Not only does the origin of the system in the controversy machine project assumptions of an adversarial character, but also the possibility of the machine being called into action serves to reinforce an adversarial relationship between reviewers and researchers. The system we have at present puts reviewers under threat of being themselves exposed to public censure in the event of approving a research project that gives rise, fairly or unfairly, to a scandal of some sort; that is, it imposes a heavy penalty on a positive mistake: a mistake in approving what perhaps should not have been approved. However, the system in place imposes no such penalty on a negative mistake: a mistake in not approving what should have been approved. Thus, the system creates a conflict of interest between reviewers and researchers; it gives the reviewers a special incentive for taking the cautious line that goes against the interests of the researchers. These effects of the controversy machine on our system of ethical review have the potential to be very damaging. Here is a list of some possible consequences. We stress that these are possible consequences, not results that have actually been documented; what they constitute is a "watch-list" for those who have a responsibility for ethical review. Ethical reviewers see research as the concern of researchers, ethics as their own concern, and so do not worry about the effects of their reviewing on research activity. In particular: they may have no hesitation in refusing to approve research projects that raise any difficult questions; they may delay and obstruct research projects that they approve, through being excessively procedural and legalistic; and they may be willing to see the traditional agenda of research being seriously cut back, if that is where their deliberation leads. Ethical reviewers are unwilling to contemplate a "steady state", in which research generally satisfies the accepted ethical standards and ethics committees play an ever more passive role. In particular: as research projects meet accepted standards, reviewers begin to make those standards more demanding, exploiting the indeterminacy mentioned earlier: they become tougher on questions to do with what is an acceptable risk of harm, on what is adequate information for research subjects to be given, and so on; and ethical reviewers begin to look for further ways of monitoring the extent to which researchers stick to ethical guidelines: they begin to assume a greater surveillance and policing role. Researchers respond to this indifference, and this incremental creep, by adopting a resistant posture. In particular: they become alienated from the process of ethical evaluation, as they come to think of ethics as the reviewers' business, not their own; they see it as perfectly legitimate to do whatever is necessary in order to bypass the standards imposed on them; they may even take positive pleasure in defying and defeating the reviewers; and they develop a collective mentality of this resistant kind and become disposed to close ranks under any scrutiny from outside; whistle-blowers are a thing of the past.
Recommendations for the reform of ethical review	It would be a great tragedy if our system of ethical review became adversarial under the pressure of the controversy machine that helped to spawn it. We do not believe that the system has yet assumed a fully adversarial profile, but there is a danger that it may be headed in that direction. It is therefore important at this juncture to try to reshape it. We conclude with some broad-gauge suggestions about the steps that need to be taken. Before offering those recommendations it may be useful to stress that they are all compatible, in our view, with the fundamental role of the ethics committee in protecting the interests of those affected by research and the interests of the community as a whole. While it is important that ethical review should not have an adversarial cast and should not assume an obstructive role, it is even more important that its role should not be compromised in any way. We believe that our suggestions (Box 2) do not threaten any compromise; on the contrary, we think, for reasons already indicated, that their implementation should help to further the ends of ethical review. These proposals are certainly not exhaustive of the things that might be done to make sure that the course of ethical review, despite its origins in the controversy machine, does not assume an adversarial direction. A widespread discussion among reviewers and researchers would be necessary to generate a full sense of the initiatives available. We hope that such a discussion will materialise and that the adversarial trajectory will not win out by default.
References	Report of the Review of the Role and Functioning of Institutional Ethics Committees. Canberra: AGPS, 1996. Allers M. Inquiry into the use of pituitary drug hormones in Australia and the Creutzfeldt-Jakob Disease. Canberra: AGPS, 1994. Gillespie R. Research on human subjects: an historical overview. Bioethics News 1989; 8 Suppl: 4-15. McNeil P. The ethics and politics of human experimentation. Cambridge: Cambridge University Press, 1993: 15-84. The Nuremberg Code. JAMA 1996; 276: 1691. Beecher H. Ethics and clinical research. N Engl J Med 1996; 274: 1354. Furrow B. Health law. New York: West Publishing, 1995: 839-841. Pettit P. Institutionalising a research ethic: chilling and cautionary tales. Bioethics 1992; 6: 89-112. MacDonagh O. A pattern of government growth 1800-60. London: Macgibbon and Kee, 1961. Ayres I, Braithwaite J. Reponsive regulation. New York: Oxford University Press, 1992. Grabosky P. Counterproductive regulation. Int J Sociol Law 1955; 24: 347-369. Sunstein CR. Paradoxes of the regulatory state. University of Chicago Law Review 1990; 57: 407-441.
Footnote	This paper represents the consensual view of the Australian Health Ethics Committee as that view was developed over the triennium 1994-1996. It was revised in the light of discussions in the Committee itself, and at a number of workshops throughout Australia in 1995, and was approved for publication by the Committee.
Authors' details	Law Faculty, University of Tasmania, Hobart, TAS. Donald Chalmers, LLM, Chair, Australian Health Ethics Committee 1994-. Research School of Social Sciences, Australian National University, Canberra, ACT. Philip Pettit, LPh, PhD, member, Australian Health Ethics Committee, 1994-1996. No reprints will be available from the authors. Correspondence: Professor P Pettit, Research School of Social Sciences, Australian National University, Canberra, ACT 0200. E-mail: pnpATcoombs.anu.edu.au ©MJA 1998 Make a comment Home | Issues | eMJA shop | Terms of use | MJA Careers | More... | Contact | Topics | Search The Medical Journal of Australia    eMJA   Readers may print a single copy for personal use. No further reproduction or distribution of the articles should proceed without the permission of the publisher. For permission, contact the Australasian Medical Publishing Company. Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au>". <URL: http://www.mja.com.au/> © 1998 Medical Journal of Australia. We appreciate your comments.