Bioethics Forum Essay
As I read a recently published report of an interesting and important placebo-controlled trial of arthroscopic shoulder surgery, one sentence in particular caught my eye: “The study was designed under the ethical supervision of an academic ethicist (JS) with placebo trial experience.” I regularly read scientific reports of clinical trials in the medical literature, and I have never seen a sentence like this before. What does it mean? What does it imply or suggest?
Before taking up a critical examination of this unusual sentence and what it implies about the role of bioethicists, I want to make a caveat. I will not be raising any questions about the ethics of this randomized trial, which employed a sham procedure to afford a rigorous evaluation of the shoulder surgery.
One way to approach the meaning of the sentence regarding “ethical supervision” is to reflect on the role of statisticians in clinical trials. Statisticians routinely serve as collaborators with clinical investigators and often as coauthors. They are charged with such tasks as determining the sample size of a randomized trial that is sufficient to provide an adequate test of the study hypothesis, identifying the appropriate statistical tests to measure study outcomes, and assuring the accuracy of the data presented from a statistical perspective. Scientific reports of studies never contain statements such as, “The study was designed and reported under the statistical supervision of an academic statistician (AB) with experience with placebo-controlled trials.” There is no need for such a sentence. The description of study methods, which typically includes a section under the title “Statistical Analysis,” and the presentation of results should be sufficiently transparent, so that readers interested in a critical evaluation of the study statistics are provided with all the information they need. Similarly, there is no need to include a statement about the study being designed under the “ethical supervision” of an ethicist. The description of the study with respect to its ethically relevant features should be sufficiently transparent to afford ethical evaluation of its design.
I suggest that this sentence essentially functions as advertising. It suggests that the study should be seen as getting a Good Housekeeping seal of approval by virtue of having employed an academic ethicist to provide “ethical supervision.” It may also suggest to readers of the scientific report that they don’t need to give much of any thought to the ethics of the study, since an ethicist supervised its design. Moreover, it is not the proper role of an ethicist to provide ethical supervision. That is the role of the research ethics committee, which is charged with independent prior review and oversight of clinical trials. Scientific reports are (properly) required by medical journals to include a statement about ethical oversight by research ethics committees. Additionally, this sentence suggests a dubious claim of expert authority on the part of ethicists, implying that they can serve as guarantors that a study has been designed so that it passes muster from an ethical perspective.
In the design and conduct of clinical trials, ethics should be the concern and responsibility of all investigators and anyone else on the research team who interacts with trial participants. Investigators can properly defer to statisticians with respect to the statistical dimensions of their study, although they ought to have some familiarity with statistical reasoning. But this sort of deference to the expertise and authority of an ethicist is not appropriate. Ethicists have expertise about ethics; those who specialize in research ethics understand how ethical issues intersect with matters of scientifically valid study design. But this doesn’t give them the sort of expert authority with respect to the ethics of a study that statisticians have with respect to the statistics.
Ethicists with research ethics expertise can play a valuable role as advisors to investigators, especially in the development of study protocols that propose to employ procedures that may be seen by some as ethically controversial. Depending on their contribution to study design and preparing the research manuscript, they can merit authorship for the study report. For example, the landmark 2002 report of a placebo-controlled trial of arthroscopic surgery for osteoarthritis of the knee stated that trial participants were required to write the following in their medical charts: “On entering this study, I realize that I might receive only placebo surgery. I further realize that this means that I will not have surgery on my knee joint. This placebo surgery will not benefit my knee arthritis.” It is my understanding that this approach was recommended by the bioethicist who served as an ethics advisor to the investigators and who was a coauthor of the study report.
The role of ethicists as part of the investigation team is desirable not only to promote ethical reflection on research design and conduct. It is also beneficial to ethicists in giving them concrete experience with clinical trials—experience that can enhance their understanding of the intersection between scientific methodology and protection of research subjects. I have played this role more than once; however, to give ethical advice about research protocols and to collaborate as an ethicist in clinical research should not be seen as performing the function of ethical supervision.
Franklin G. Miller, PhD, a Hastings Center Fellow, is a professor of medical ethics in medicine at Weill Cornell Medical College.