In the first of a four-part series on Ethical Issues in Research, Donald Evans, professor and director of the Bioethics Centre  at the University of Otago in New Zealand shares his views on ethical review of innovative treatment. Evans is a research ethics specialist actively engaged in the ethical review of research. He also trains ethics committees in New Zealand and Asia.
Over the past 40 years, medical research has been subjected to growing ethical scrutiny. In most parts of the world, independent ethical review committees have emerged to protect the interests and welfare of research participants. Generally, these committees are not responsible for reviewing clinical practice. The result is that innovative clinical procedures, which are usually research orientated, often evade such scrutiny and patients are denied the protection it provides.
In New Zealand, the system is different due to the origin of ethical review there. Over a period of 30 years, a study of the natural history of carcinoma in situ involved regular examination of women with cell changes in the cervix. These women were never told that they were in a study. The researchers resolutely refused to accept the international opinion, which emerged early on in the study, that such changes were precursors of invasive cancer. The women were not offered treatment. The result was a considerable number of avoidable deaths and other major harms.
One line of defense employed in the national enquiry into the affair was that the women were part of a conservative treatment regime. The enquiry therefore recommended inter alia that independent ethical committees be set up to review both research protocols and innovative procedures in gynaecological malignancy ( 1). The latter was generalized in the implementation of the recommendation to include all innovative treatment. However, review of this kind has been difficult to achieve. There are a number of obstacles that make the process problematic.
What is innovative practice? The question seems innocent but answers to it are not easily found. For example, does the term cover the introduction of an established technique into a new country? Laparoscopic (key-hole) surgery provides us with a good case. Although the procedure was readily used in a number of countries, its introduction to New Zealand called for assurances about its safety. But these assurances were concerned more with the requisite skills of the practitioners than the safety of the technique itself.
The safety of patients was ensured by proper professional standards of application of the technique. This was achieved by the credentialing of surgeons--based on the minimum requirements of numbers of observed and supervised procedures they must have done--by the appropriate professional body before they were allowed to employ the technique. This was, however, a matter for the profession rather than for ethics committees.
On the other hand, the introduction of a proven technique into a country might raise ethical issues if, by its nature, it conflicted with cultural norms in that society. For example, the introduction of xenotransplantation into some societies would raise ethical problems over and above the issues of safety. Ethics committees would have a proper role in reviewing such proposals.
There is a further problem in identifying innovative practice. This is most evidenced in surgery but is to be found in other health care disciplines such as physiotherapy and nursing. Much of surgical practice has developed by means of incremental changes to established procedures over a long period. Each of these changes might be insignificant in themselves but over a period, the procedures might have become something quite different from the original. That begs the question at what stage should we declare it an innovative treatment ( 2)?
An essential element of an innovative procedure calling for review is the fact that it is planned. It would be unwise to rule out the possibility of new interventions being employed in surgery in response to a crisis in theatre, which might affect the rescue of a patient. If the patient recovers, the surgeon might contemplate both proceeding in that way on a future occasion and even modifying the innovation. This would make the subsequent procedure quite different in kind. Given that one case would be insufficient to demonstrate the efficacy of the procedure, the premeditated intervention should be canvassed with potential recipients and their consent obtained. All such innovations should be canvassed with an ethical review committee.
One area of innovative treatment is already covered by ethical review by demand of licensing regulations, and that is the use of new pharmaceuticals and the application of licensed pharmaceuticals to new indications.
Some procedures are undisputedly innovative, such as the insertion of a new medical device in a patient. Whilst in the case of incremental innovations the practitioner might be unclear about the status of the procedure, and thus not report it to an ethics committee, the same cannot be said of these cases. Yet, many such procedures have been introduced without any independent review, or the guarantee of patient consent. Some procedures of this kind have turned out to be very dangerous for patients, sometimes resulting in catastrophic failure, as in the Halifax clamp case ( 3). It took the reporting of these failures to prompt a thorough review of the safety of the innovation, which was too late to protect the patients subjected to the rigors of the procedure.
In other cases, the publication of purported successful results of such procedures alerts ethics committees to the fact that innovative practice has proceeded without independent review. Once again, this makes protection of patients impossible to achieve as it can be reasonably assumed that many innovations, which have not produced allegedly good outcomes, do not see the light of day in learned journals. One control of this unethical practice would be the insistence of journal editors on prospective ethical approval of studies presented for publication.
Sometimes such practices are uncovered by whistle blowing. This is an unfortunate necessity in cases where the practitioner will not heed advice and requests of fellow professionals. It is a reporting method of last resort as it is destructive of collegial relationships and often very costly to the whistle blower. Nevertheless, should all else fail, there is a moral obligation on health care professionals to put the interests of patients first and ensure, as far as they can, that the practice of their fellows is ethical ( 4).
The ideal situation would be where voluntary practitioner disclosure was the order of the day and a collaborative relationship developed between practitioners and ethical review committees. A good deal of education of researchers by ethical review committees is called for in order for this to be achieved. A further general rule should also be considered, vis-à-vis where a practitioner is in doubt as to whether a procedure calls for ethical review, the matter should be canvassed with the chair of the relevant ethics committee.
The foremost function of ethical review is to ensure the safety of participants in research as much as possible. This may require the mechanical testing of devices before use, animal testing and, where it is feasible, testing procedures on cadavers. In the absence of these, theoretical assurances of safety should be provided to the satisfaction of reviewers.
How should the innovative procedure be allowed to proceed? Given the uncertainties with respect to safety, ethical approval for a small number of interventions should be the first step in order to assess the safety of the procedure. The results should be subject to independent evaluation to avoid practitioner bias. Given a satisfactory outcome, a prospective randomized controlled study should be developed and the procedure should be made available to patients only in the context of that study. This should facilitate further safety assurances and an evaluation of the efficacy of the procedure.
There are some theoretical problems in the design of such studies ( 5). Clearly double blinding (where neither the participating individuals nor the study staff knows which participants are receiving the experimental treatment and which are receiving a placebo) of the study is not possible, and on many occasions, single blinding (where either the investigator or participant, is unaware of what treatment the participant is taking) would be impossible. There is also the prima facie problem of clinical equipoise. The inventor of the procedure will often be persuaded that the procedure is an advanced one and might claim to lack clinical equipoise, thereby feeling an obligation to offer it as the best treatment to patients. This would be a misunderstanding of the notion of clinical equipoise, which cannot be removed by the conviction of one practitioner on the basis of a small number of procedures.
In such a situation, the appointment of an independent randomizer and evaluator of pre- and postoperative conditions of the patients should ameliorate the threat of research bias in the study. A study like this would be akin to a Phase II pharmaceutical study (a controlled study to evaluate the effectiveness of a treatment or procedure for the particular indication in patients with the disease or condition under study and to determine the common short-term side effects and risks). Independent evaluation of the outcomes of the study would then decide whether the innovative procedure could be made available to patients on the basis of clinical preference.
The Report of the Committee of Inquiry into Allegations Concerning the Treatment of Cervical Cancer at National Women?s Hospital and into Other Related Matters, Sections 5.a.ii and 5.b.1 (Auckland, New Zealand, Government Printing Office, 1988). See, for example, the ?Sling on a String? case in Donald Evans and Martyn Evans, A Decent Proposal: Ethical Review of Clinical Research, p. 56 (John Wiley and Sons, Chichester, 1996). Donald Evans and Martyn Evans, 'The Halifax clamp case', ibid, p. 57. Such a situation faced Dr. McIndoe in the New Zealand study referred to at the start of this paper. See Sandra Coney, The Unfortunate Experiment, Chapter 1 (Penguin Books, Auckland, 1988). For a discussion of these, see Donald Evans, ?Ethical Review of Clinical Treatment?, Health Care Ethics Forum 14 (1), pp. 53-63 (March 2002).
The Report of the Committee of Inquiry into Allegations Concerning the Treatment of Cervical Cancer at National Women?s Hospital and into Other Related Matters, Sections 5.a.ii and 5.b.1 (Auckland, New Zealand, Government Printing Office, 1988).
See, for example, the ?Sling on a String? case in Donald Evans and Martyn Evans, A Decent Proposal: Ethical Review of Clinical Research, p. 56 (John Wiley and Sons, Chichester, 1996).
Donald Evans and Martyn Evans, 'The Halifax clamp case', ibid, p. 57.
Such a situation faced Dr. McIndoe in the New Zealand study referred to at the start of this paper. See Sandra Coney, The Unfortunate Experiment, Chapter 1 (Penguin Books, Auckland, 1988).
For a discussion of these, see Donald Evans, ?Ethical Review of Clinical Treatment?, Health Care Ethics Forum 14 (1), pp. 53-63 (March 2002).