Canadian Medical Association Journal 1996; 155: 442-444
[en bref]
© 1996 Nicole Baer
What, for example, constitutes informed consent from a child or a mentally incompetent adult? Should the usual procedures for informed consent be waived in emergencies?
A new draft code of conduct for research involving humans aims to bring a fresh vision to the ethics field, where the lines of what is acceptable and unacceptable are often fine and may shift. It lays out the fundamental principles of ethical research and establishes a framework to ensure they are carried out.
"The regulation of research has been very much hit-and-miss in the past in Canada," notes Dr. John Watts, a Hamilton neonatologist and past president of the Canadian Bioethics Society. "There has been a great deal of variability between different parts of the country and between research ethics boards in different institutions."
Once the draft code is revised and approved, likely by the end of the year, investigators will have a clearer understanding of the limits of acceptable research.
The code, a joint endeavour of the Medical Research Council (MRC), the Natural Sciences and Engineering Research Council and the Social Sciences and Humanities Research Council, supersedes all three agencies' earlier ethics statements, including the MRC's 1987 Guidelines on Research Involving Human Subjects. The new code will apply to all research conducted under the auspices of any institution that receives funding from one of these councils. It is intended to set a standard for all Canadian research involving human subjects, no matter where it is done or who is pursuing it.
The single biggest difference between this code and all previous attempts to define the limits of ethically appropriate action lies in its interdisciplinary nature. A code of conduct that applies uniformly to research by sociologists, physicians, engineers, nurses and other investigators acknowledges two important realities:
The code "takes as its premise that there are common criteria -- baselines -- that should govern the conduct of research involving humans," says Derek Jones, then director of the National Council on Bioethics in Human Research. "Does [this] mean that because it's difficult to do that you abandon it? There are, no doubt, some questions about that, but I think the answer is No: you set out a draft -- as they have done -- and you refine it with experience, knowledge, insight and public criticism."
The code suggests that three categorical questions should be posed when assessing the ethical nature of a study. First, is the research scientifically valid? Second, does the research have sufficient overall value? Finally, are research subjects treated with dignity and respect? But while at least one of these conditions is always necessary to build the case for an ethical investigation, none is in and of itself sufficient. For example, a new medication may be very likely to prove effective in patients, thus meeting the first criterion, but if it is the 10th variant in a long line of similar drugs, then it is unethical to subject patients to another clinical trial because it does not meet the second criterion of providing sufficient overall value.
Indeed, the central and unshakable premise of the new code, repeated like a mantra in chapter after chapter, is that the interests of the research subject are paramount. As the document lyrically puts it: "In many cases it is appropriate to think of research subjects as hosts who invite the researcher as a guest."
Watts, who does research on humans at Chedoke-McMaster Hospital in Hamilton, says this approach reflects a growing, and welcome, trend toward patient- or "client"-centred care that is in evidence elsewhere, such as hospitals. And, while the new rules are undeniably tough on researchers, Watts says this is for the overall good of society and the credibility of science.
"It's true that some research won't be done, although my reading of this document is that it would prevent being done only those forms of research that shouldn't be pursued -- that clearly or are likely to carry a risk to the patient, or which didn't have the consent of the patient. The benefits of this form of control far outweigh the risks of being unable to do desirable research."
With the interests of patients foremost in their minds, investigators must then be able to defend their research proposals to the satisfaction of their research ethics board (REB). The REB must be persuaded that the benefits likely to accrue to the subject are at least commensurate with the overall personal costs. Of course, it is up to each subject to distribute entries in the ledger, with anything from enhanced medical care to a rewarding sense of altruism on one side to fear or even out-of-pocket expenses on the other.
The document insists its intent is to educate and guide, not to instruct. In place of recipes for proper behaviour, the code seeks to instil in investigators the capacity to think ethically. "Good ethical reasoning, like good scientific reasoning, must be more than a matter of the mechanical and dogmatic application of rigid rules to fact situations," the draft document says. "Ethical reasoning requires thought, insight and sensitivity. . . . Securing the approval of an REB should be the occasion for more than simply jumping through another bureaucratic hoop. It should be an opportunity for informed ethical reflection and discussion with ethical peers."
Even so, the code is very explicit in some areas, such as the exhaustive description of the minimum composition and procedures of REBs. It also clarifies many concepts that were only implicit in previous ethics guides. For example, it itemizes the type of research on humans that must be submitted for REB scrutiny. (The answer: pretty well everything, even if it involves only the collection of data from existing records; when in doubt, the code warns, consult the REB). [Some researchers argue that increased use of REBs may create operational problems because these are already operating near capacity. Because of system-wide health care cuts, it is unlikely more funding will be available to support additional work. -- Ed.]
On the other hand, the code introduces changes that could make life easier for some researchers. Recognizing that some cultures value oral agreements over written ones, the code permits investigators to obtain oral patient consent under certain circumstances. Also, the new code establishes a framework for parental consent for research involving children, something that was not well developed under the 1987 MRC guidelines. (The National Council on Bioethics in Human Research established more extensive guidelines in 1993.)
For the first time, there is a requirement that reasonable steps be taken to include groups traditionally excluded from research -- the young, the old, pregnant and breast-feeding women, and aboriginal groups. The document argues there are both scientific reasons (different groups may react very differently to the treatments under study) and philosophic ones (members of excluded groups may be unable to share in the benefits derived from the research).
Not surprisingly, the draft code has created plenty of controversy. Some of this was evident during a panel discussion held by the Clinical Epidemiology Unit of the Loeb Medical Research Institute in Ottawa. For example, some physicians are worried about a section that precludes investigators who provide health services from asking their patients to participate in research. Dr. Jean Joly, a Laval University professor and chair of the three-council working group that wrote the document, says an arm's-length relationship between researcher and health care provider is essential to redress the unequal power relationship between patient and doctor.
On the other hand, Joly also admits there are trouble spots that will be improved in the final draft, and underlines that even that will have to be updated regularly in the future.
Some critics are uncomfortable with parts of the sections on clinical trials, while others are baffled by the distinction between research and hospital quality-assurance practices.
In general, areas where science moves most quickly and social sensibilities are most at odds create the biggest challenges. The field of human reproductive technology, for instance, witnessed divisions even among members of the working group.
The draft code does contain an important new concept: continuing ethical review. In the past, once a researcher received "up-front" approval for a study, that was the end of the process. In the draft, there is a recognition that follow-up is needed to ensure that the ethical aspects of the study that was proposed are actually carried out.
Does this suggest you cannot trust the investigator? "No", says Joly, who adds that "the investigator should propose the means of follow-up." As a rule of thumb, a study at low risk for ethical misconduct would have minimal follow-up, whereas a study in which the risk is higher would have more.
Dr. Peter Walker, dean of medicine at the University of Ottawa and one of the authors of the code, says the mere existence of the draft is a testament to Joly's skill at coaxing consensus despite radically disparate interests. "We cannot underestimate the difficulty of the task of writing this document," he says.