2004年6月1日 | EN
Establishing research ethics standards involves identifying, promoting, and getting consensus on a coherent set of principles and/or procedures that can guide the actions of researchers, and that ethics review committees can use to judge the ethical merit of a given study.
This policy brief provides an introduction to this field, emphasising the global nature of regulating international human subjects research, describing the different approaches used, and highlighting some of the challenges faced. It aims to sketch out some of the main issues so that the obstacles to successful global regulation can be better appreciated, and creative proposals can be considered with these obstacles in mind.
The debate about the ethics of trials investigating prevention of perinatal HIV transmission in developing countries in the 1990s brought the relationship between research ethics and development into sharp relief, and focused attention on the twin challenges of what should count as standards in research ethics and how these should be applied globally.
The debate revealed different and conflicting views about what ethical and practical standards should govern human subjects research conducted in developing countries. It also gave rise to accusations that one set of standards was being employed for research in developed countries and another for research in developing countries. Although these exchanges were occasionally venomous, they have helped to improve our appreciation of how difficult it can be to achieve consensus on global standards for an enterprise as complex and varied as health research. At the same time they have illustrated how costly it can be when consensus is not achieved.
Several multilateral organisations are internationally recognised for their role in the protection of human subjects in research. Foremost among these are the World Medical Association (WMA), which has 78 member countries and is steward of the Declaration of Helsinki. Other agencies include the Council for International Organisations of Medical Sciences (CIOMS), which has produced international ethical guidelines on both biomedical  and epidemiological research. 
Agencies of the United Nations, such as the Joint Programme on AIDS (UNAIDS) and the World Health Organisation’s Tropical Diseases Research Unit (WHO/TDR) have also been active in promulgating research ethics guidelines, although this is a relatively recent development. These guidelines capitalise on the global status and reputation of these agencies, and on their close working relationships with Ministries of Health in WHO’s 191 member countries, in an attempt to shape the evolution and improvement of research ethics, in particular HIV vaccine research and research ethics review.
This policy brief attempts to provide an introduction to some of the challenges involved in regulating international human subjects research. Given the breadth and complexity of the issue, the policy brief only aims to sketch out some of the main issues so that the obstacles to successful global regulation can be better appreciated, and creative proposals can be considered with these obstacles in mind.
Globalisation of research ethics standards
Despite some important and contentious disagreements about details, principles of research ethics are in fact already embraced globally. Indeed, the debates surrounding the HIV perinatal transmission trials arose from allegations that they violated universal research ethics standards, most notably those embodied in the Declaration of Helsinki. Among other things, this focused the debate on what it means to claim that research ethics standards are “universal” or “global”. Perhaps not surprisingly, the debate has produced no simple answers. The challenge is to determine both what should count as standards in research ethics, and how these standards should then be applied globally.
Establishing research ethics standards involves identifying, promoting, and – to the extent possible – getting consensus on a coherent set of principles and/or procedures that can guide the actions of researchers, and that ethics review committees can use to judge the ethical merit of a given study. Most of the important guidelines for human subjects research have adopted a similar architecture to the Nuremberg Code of 1947, in that they emphasise certain fundamental principles.
Such principle-based guidelines do a good job of laying out the general boundaries of ethical acceptability. But they have been criticised for providing insufficient direction on practical matters, such as whether payment of research subjects is ever acceptable, or who should shoulder the financial responsibility for providing continuing access to beneficial interventions following the completion of a trial.
Questions like these reveal the limitations of relying exclusively on principles as standards. They also illustrate the distinction between substantive and procedural standards, i.e. between substantive goals such as respecting and protecting human subjects, and the procedures that should normally be followed in the course of research involving human subjects. The Declaration of Helsinki is probably the best-known set of substantive international standards, and the Operational Guidelines for Ethics Committees that Review Biomedical Research issued by WHO/TDR is an important example of procedural international standards.
Assuming that it is possible to specify an appropriate and convincing set of standards, their application still presents a complex challenge. For example, to have any reasonable claim to global status, guidelines or rules or regulations must have a credible authority that can compel respect and adherence with some degree of uniformity throughout the world.
Three general strategies have been employed for globalising research ethics standards: voluntary adoption models, conditional voluntary adoption models, and mechanisms under international law. Existing international guidelines are essentially voluntary in nature. And although certain guidelines have some recognition in international law, they do not have formal legal status.
Voluntary adoption and conditional voluntary adoption models
Voluntary models such as the Declaration of Helsinki have moral authority only to the extent that their promulgating agencies are able to compel researchers, sponsors, and ethics review committees to be bound by their aims, principles, and rules. Other examples include the CIOMS guidelines, and the UNAIDS vaccine research guidelines. However, willing parties can have vastly different views about what is required to satisfy any given rule. Voluntary compliance with these guidelines and rules therefore provides little assurance of uniformity in practice or outcomes.
In some cases, medical journals have required researchers to follow the Declaration of Helsinki as a condition of publication, in effect making it a conditional voluntary adoption model. In such models, compliance with the rules is still voluntary, but decisions not to comply carry a significant cost.
For example, US federal government funding for human subject research requires assurance of compliance with US regulations, whether the researcher is in Boston or Bamako. As a result, the US regulations currently function as de facto international standards for US-funded research. Some researchers find this international reach to be inappropriate, and there is even uncertainty as to whether the regulations in question can be applied legally outside the United States. 
It can be argued that criticism of conditional voluntary adoption models is hard to sustain, since they only require researchers and their institutions to follow the regulations if they decide to accept funding. In fact, one of the main criticisms of the current US regulations is that they do not extend to private companies. Yet it is estimated that more than half of the US$70 billion spent annually on global health research comes from private entities, which are now heavily concentrated in the United States. 
It appears that neither the US government nor the pharmaceutical industry is willing to make sufficient resources available to ensure adequate oversight in all countries in which US-sponsored research is being conducted. Thus, both voluntary models raise critical issues with respect to the effective monitoring and oversight of research, which has become something of an Achilles’ heel for US regulatory authorities,  and for the global health research ethics enterprise overall.
It is conceivable that international agencies, such as the WHO, could adopt a leading role in the globalisation of voluntary standards, or even some form of conditional adoption standards. But this is a long-term prospect. And although the leadership of a truly international agency would be welcome, it would not address some of the limitations inherent in the voluntary models themselves.
Mechanisms in international law
The third type of mechanism available for globalising research ethics standards is international law. The Nuremberg Code, the Declaration of Helsinki, and the CIOMS guidelines are not instruments of international law per se, and therefore have no specific legal authority. However,
given their widespread international use – the guidelines have been adopted in whole or in part into the regulatory schemes of various countries – some have argued that they satisfy the requirements to be considered as ‘customary international law’. Yet these requirements also include evidence of general and consistent state practice, and that this is followed out of a sense of legal obligation.  Furthermore, customary international laws are not binding on states that have persistently objected to their use, making this approach vulnerable to becoming merely a voluntary model.
Another potential avenue for conferring enforceability is that international guidelines may have some status in international human rights law. For example, the Nuremberg Code (which, in fact, was not written as a set of research ethics guidelines by an international agency, but rather as a preface to the rulings of a US military tribunal) – in particular its provision that individuals may not be subjected to scientific or medical experiments without their freely given consent – has appeared in human rights treaties such as the International Covenant on Civil and Political Rights. 
Unfortunately, the realities of international relations – particularly given the current US administration, which has demonstrated considerable reluctance to commit to multilateral cooperation – makes any prospect of a meaningful solution to international regulation of human subjects research through mechanisms in international law extremely unlikely. Many observers agree that the economic and political interests in health research — particularly privately-funded pharmaceutical research — are too great, and exert too powerful an influence on the current regulatory practices in human subjects research in the United States to permit the devolution of guidelines to international law, at least in the foreseeable future.
The problem of exceptions
The HIV perinatal transmission trials highighted some important problems about exceptions to research ethics standards, most notably with respect to the standard of care that should be provided to participants in a clinical trial. Guidelines, such as the Declaration of Helsinki and CIOMS, had relied on vague language such as “best proven” and “best available” therapy to communicate their intended standards. This language proved to be too malleable to provide reliable guidance. Thus, even in the poorest countries in sub-Saharan Africa, the notion of a “best proven” standard — which, for the HIV trials many reasonably understood to be the expensive treatment regimen in use at the time in tertiary care hospitals in Boston, Vancouver or London — could be contested on the basis that the standard had not been proven effective in Lusaka or Gabarone.
Although this claim is of dubious biological plausibility, it is an instructive example of how concepts can be too flexible to offer steady guidance.  Even if the Declaration of Helsinki were to state unequivocally that every participant in every clinical trial anywhere in the world must receive the best diagnostic, prophylactic or therapeutic intervention known to medical science, it would still not resolve the question of whether such an intervention would really be the “best” in any given set of circumstances. For example, if an extremely expensive new treatment was demonstrated to be slightly more effective at suppressing HIV replication than the far less expensive existing gold standard drug therapy, it would be hard to imagine that the expensive treatment could be considered the “best” intervention for most developing countries.
But how does this primarily economic dilemma relate to research ethics standards and regulation? In the HIV perinatal transmission trials the justification for using lower standards than would have been permissible in developed countries was based solely on the impoverished conditions in which the research subjects found themselves, not in the established principles of research ethics.  Rather, the justification for selective non-application of the usual research ethics standards, was largely implicit, and had to do with a complex range of assumptions about the scope of researchers’ obligations, the relevance of local practices and entitlements, and the presumed population benefits that would follow the trial. 
The critical insight to be gained from this example, is that any attempt to establish meaningful standards must be able to specify what circumstances might count as exceptions to these standards, and what degree of latitude might be justifiable and for what reasons. And yet it is precisely this challenge that has caused most grief to those drafting revisions to the Declaration of Helsinki, the CIOMS Guidelines, and the UNAIDS vaccine guidelines. It is fair to say that no satisfactory solution has been found.
The continual development of global standards
The process by which international guidelines are developed is critical to their legitimacy and authority, particularly since the main guidelines function under a voluntary adoption model. During recent revisions of the Declaration of Helsinki and the CIOMS Guidelines, the issue of whose perspectives were taken into account emerged as a critical challenge. In particular, questions were raised over whether there was sufficient developing-country representation during the drafting process,  and also whether there was sufficient transparency with respect to the influence of powerful research interests to ensure an appropriate balance between protecting research participants and facilitating important scientific research.
However, concerns about fair representation and inclusive processes are less obvious in the evolution of standards that don’t have a clear history of authority and relevance to developing countries. For example, the Good Clinical Practice guidelines published by the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) explicitly reflect the interests of the pharmaceutical industry. Although they may also represent a positive development in international research practice – in that standardisation is likely to reduce the incidence of unacceptable practices – it is difficult to view them as being global in any meaningful sense, despite the fact that they have become de facto international standards by virtue of their use by many licensing authorities in developed countries.
Although the continual revision of guidelines is acknowledged as a necessary process, many view the most recent revision of the Declaration of Helsinki,  (together with a subsequent clarification,  other proposed clarifications,  and criticisms of the new wording and spirit of the declaration ), as undermining the credibility and authority of the Declaration and raising doubts about its primacy as an international guiding document. This comes at a time when other international guidance documents, particularly the revised CIOMS Guidelines, differ from the Declaration of Helsinki on several critical, substantive issues (such as the use of placebo, and post-trial obligations). Unfortunately, the result has been a dilution of the guidance on these issues, rather than clarification for researchers and sponsors alike.
Of course, it is widely recognised that improved precision in the wording of international guidelines is insufficient, in itself, to ensure ethical conduct in research. Since all the international guidelines require some type of independent review committee to make complex judgements about the ethical acceptability of any proposed protocol – and it is now widely held that review of research must be conducted in both host and sponsor countries – it is essential that the developing country hosting a given research project has the capacity to form such committees, and that members have adequate training (see Building capacity for ethical review in developing countries).
These goals have been enormously enhanced by the work of a number of regional initiatives organised under the umbrella of the Strategic Initiative for the Development of Capacity in Ethics Review (SIDCER). As these activities expand and produce a critical mass of highly trained professionals throughout the developing world, the current domination of US regulations may be challenged and opportunities for innovative new global strategies will emerge.
Applying and enforcing standards
The application and enforcement of global standards depends both on their legitimacy and authority. Although the US regulations have questionable legitimacy when applied in other countries, they do have some authority, since compliance with the regulations is required for the receipt of US federal government research funding. The Declaration of Helsinki and CIOMS guidelines, on the other hand, may have some legitimacy. But they lack the kind of persuasive authority that comes with access to US funding.
Some view the lack of a legitimate and authoritative global agency to promulgate and oversee such standards as one of the central problems with the international regulation of human subjects research. There is, in fact, a great deal of agreement on the substantive standards that should govern research involving human subjects (despite the few important points of disagreement). But there is currently no credible and truly global authority, or appropriate mechanism, that can compel researchers and sponsors to conduct research in a particular way, or that can intervene with appropriate sanctions when actions significantly diverge from the global norms.
One of the strengths of the US regulatory scheme is that it provides a clear, if imperfect, accountability loop: institutions, researchers, and sponsors must follow the regulations, and regulatory authorities are authorised to take corrective action – including withholding funding and even suspending activities – when the rules are not followed. The central challenge for global governance is to duplicate the force of these accountability mechanisms while gradually broadening the base of authority beyond Washington.
Looking beyond the current situation
In reality, there is little likelihood of a single, comprehensive global system for the regulation of human subjects research. A more realistic goal would be greater equity between developed and developing countries in terms of their ability to establish and sustain a strong culture of ethical conduct in research.  To achieve this, research ethics will need to be tied more convincingly into health research systems, as has been proposed in the WHO Health Research Systems initiative. 
Additionally, despite excellent progress in enhancing capacity to conduct ethical review of research proposals in developing countries, there has been insufficient examination of what else is required to protect human subjects and promote ethical conduct in research, such as strong mechanisms of accountability within institutions, and between institutions and regulatory authorities.
Of course, since ensuring the ethical conduct of research is presently left largely to volunteers, meaningful changes will require investment in both people and systems. And although existing training and regional development initiatives have produced remarkable results with little material support, it remains to be seen whether further investment into these activities can be encouraged by developing-country governments and relevant international agencies.
The development of national guidelines and formal mechanisms of research ethics review in developing-country institutions undoubtedly plays an important role in the regulation of human subjects research. But other activities are emerging that also promise to improve global equity in this field. These include voluntary accreditation programmes that aim to enhance the capacity of institutions to develop and sustain strong programmes for the protection of human subjects and promotion of ethical conduct in research, and encourage institutions to strive for excellence in these practices.
There are also calls for host country ethics committees to be given greater latitude to make decisions about the ethical acceptability of research when there is disagreement between host country and sponsoring country committees.  And finally, innovative arrangements for research ethics review are being developed between collaborating institutions in developing and developed countries that focus on creative ways to improve communication and cooperation in research ethics review and oversight.
The author is research scientist at the Centre for Global Health Research and the Centre for Inner City Health Research, both based at St Michael's Hospital in Toronto, Canada. He is also assistant professor in public health sciences at the Department of Public Health Sciences and Joint Centre for Bioethics at the University of Toronto.
 Council for International Organisations of Medical Sciences (2002) International Ethical Guidelines for Biomedical Research Involving Human Subjects
 Council for International Organisations of Medical Sciences (1991) International Guidelines for Ethical Review of Epidemiological Studies
 World Health Organisation (2000) Operational Guidelines for Ethics Committees That Review Biomedical Research
 DuBois W (2003) New drug research, the extraterritorial application of FDA regulations, and the need for international cooperation. Vanderbilt Journal of Transnational Law 36:161-207
 MSF Access to Essential Medicines Campaign (2001) Fatal Imbalance: The Crisis in Research and Development for Drugs for Neglected Diseases p.171-175
 US Department of Health and Human Services (2001) The Globalization of Clinical Trials: A Growing Challenge in Protecting Human Subjects
 Fidler D.P. (2001) “Geographical Morality” revisited: international relations, international law, and the controversy over placebo-controlled HIV clinical trials in developing countries. Harvard International Law Journal 42(2):299-354
 London A.J. (2000) The ambiguity and the exigency: clarifying ‘standard of care’ arguments in international research. Journal of Medicine and Philosophy 25(4):379-97 (abstract only)
 Levine R.J. et al (eds.) (2002) Biomedical Research Ethics: Updating International Guidelines – A Consultation p.27-40
 World Medical Association (2000) Declaration of Helsinki: Ethical Principles for Research Involving Human Subjects
 World Medical Association (2002) Clarification to Paragraph 29, Declaration of Helsinki
 World Medical Association (2003) Secretariat Report, and Documentation for the Preparation of Note of Clarification, on the Revision of Paragraph 30 of the Declaration of Helsinki
 Forster H.P. et al (2001) The 2000 revision of the Declaration of Helsinki: a step forward or more confusion? The Lancet 358:1449-53 (free registration required)
 Lavery J.V. (2001). A Culture of Ethical Conduct in Research: the Proper Goal of Capacity-Building in International Research Ethics. Commission on Macroeconomics and Health Working Paper
 Pang T. et al (2003) Knowledge for better health – a conceptual framework and foundation for health research systems. Bulletin of the World Health Organisation 81(11):815-820
 Mfutso-Bengu J.M. & T.E. (2002) Ethical jurisdictions in biomedical research. Trends in Parasitology 18(5):231-234