Curricular Objectives

  1. To develop a framework for understanding the ethical issues raised by performing research in human subjects
  2. To understand the concept of clinical equipoise
  3. To understand the procedural requirements for conducting a clinical research study

Summary of Case

You are conducting a study of your new laparoscopic fixation technique for controlling gastroesophageal reflux with the Acidcheck valve prosthesis. You will be presenting a talk at a meeting of your specialty society in a few weeks. You need 6 more patients to round out your series to 100 cases. The safety committee for the study has asked for increased intensity of surveillance because of concerns about side effects of the treatment (two cases of intestinal ischemia from pneumoperitoneum, one intestinal obstruction from migration of the prosthesis).

Mrs. G has severe gastroesophageal reflux unresponsive to treatment. She is interested in a minimally invasive operation, but does not fit the eligibility criteria precisely because she has mild diabetes. She asks that you disregard this minor disqualifying condition, because the treatment is not available outside the trial setting. She would prefer to wait a few weeks before surgery.

Lesson Plan

Question 1:

Should Mrs. G be entered into the trial? Take a pedagogic vote: Yes, No, Don’t know.


Mrs. G should not be entered into the trial because she does not meet the inclusion criteria agreed upon. The design of the protocol is part of a contract among all stakeholders in the research, including patients, investigators, and the funding agency, that certain constraints will be placed on clinical practice in order to assure scientific validity and develop generalizable knowledge. Violating a research protocol brings lasting discredit when it is made public. This was dramatically illustrated by the professional and public response to the actions of Dr. Roger Poisson (who falsified documents and treatment assignments in the NSABP trials of breast cancer treatment) and Dr. William Summerlin (who painted the skin of white mice black in transplantation experiments). Violations of trust and loss of integrity usually induce irremediable distrust.

Question 2:

What is the difference between clinical research and clinical practice?


Clinical research is the application of the scientific method to the practice of medicine in order to produce generalizable knowledge. This change often introduces the requirement for data collection and storage, potentially affecting confidentiality. It may also introduce a conflict of interest for the physician when the goals of the study conflict with the interests of the patient. In clinical practice physicians advise or choose treatments or test based on available information, but do not generate new generalizable knowledge as part of the process.

Question 3:

Is it morally acceptable for physicians to use patients from their practice as subjects in research?


Yes, provided that certain conditions are met. The research should be carefully planned to minimize risk and must be scientifically valid. Patients should understand that they are being asked to participate in research and freely choose to do so. Research protocols should be approved by an independent Research Ethics Board. The patients’ bond of loyalty to the doctor should not coerce them to participate, and a decision to abstain from research should not affect care adversely. Their decision to participate in research should be a voluntary and informed choice. The standard of disclosure is more stringent in the research setting than in clinical practice.

Question 4:

Is it morally acceptable to enter your patients in a randomized trial when you strongly believe that one treatment is superior to the other?


In coming to a satisfactory answer to this question it is important to distinguish between the judgement of the individual physician and the judgment of the community of physicians with the appropriate expertise. If there is consensus within the relevant medical community that one treatment is superior to the alternatives, then it is unethical to enter patients into a trial that includes a known inferior treatment. When the relevant expert medical community has not come to a consensus, it is in a state of clinical equipoise. When clinical equipoise exists within the expert community, it is ethically acceptable to enter patients into the trial, even if you believe that one of the treatments is superior. Entering patients into the trial is an act of intellectual humility on the part of the individual investigator that will help to resolve the uncertainty about which treatment is superior.

Question 5:

Dr. A is a senior surgeon, coauthor, and the inventor of the original Acidcheck valve that you have improved. He is very pleased that you have come up with the modification that makes it safer and more reliable than the original model, which had a tendency to migrate. As the meeting draws nearer, he decides that he should displace you as first author and present the paper in your place. How should you respond? Is this a violation of ethics, or simply a violation of etiquette? What recourse do you have?


The order of authors is not governed by rules but by customs. In English departments, authors are ordered alphabetically; philosophy departments favor single authorship; departments of medicine and science accept multiple authorship. Because the moral community of science has a tradition of granting first place to the author who has contributed most to the work that is being reported, you have a moral claim, based on the principle that justice requires reasonable expectations to be met. This is particularly persuasive when those expectations are explicitly created. The leadership of a department or university may honor an appeal that is based on explicit prior agreement or evidence of scientific misconduct on the part of Dr. A. Disputes about authorship are generally settled by mediation.

Bioethics Bottom Line

Medical research can be defined simply as the application of the scientific method to medicine. The demand for scientific rigour introduces changes from everyday therapeutic practice. Many of these changes create ethical tension. Innovative procedures or therapies may be used, or new drugs administered. Researchers who have no clinical responsibility for patients may gain access to their medical records, raising obvious concerns about confidentiality. The requirements of research design might force physician-researchers to accept limitations on the exercise of their clinical judgment. However, the norms that guide ethical research are the familiar principles of respect for autonomy, beneficence, and justice.

The principle of respect for autonomy dictates that potential subjects be given the opportunity to choose whether to participate in research, and their choice should be an informed one. Patients who may become subjects should understand that they are being asked to participate in research, which of the proposed interventions are part of standard therapy, and which are necessary for research purposes. Patients should also understand that their choice not to participate in research will not jeopardize the physician-patient relationship; those who refuse should still be provided with standard therapy.

It is sometimes argued that the demands of science create insuperable moral barriers to physicians’ participation. The principle of beneficence imposes on the physician an obligation to recommend interventions that, according to the physician’s clinical judgment, are best for a particular patient. But the requirements of good scientific design may take this judgment out of the physician’s hands. This occurs most obviously when the alternative treatments are assigned randomly. The response to this argument lies in recognizing how the physician’s clinical judgment must be tempered by the opinions of professional colleagues. Professional consensus creates boundaries within which individual clinical judgments may operate.

Clinical equipoise exists when the expert medical opinion remains divided over the best choice among treatment options; the medical community is unable to achieve consensus on the preferred treatment. The supporters of each alternative realize that the available evidence is inconclusive. The crucial factor in deciding whether it is morally acceptable to offer patients participation in a trial is whether one has a treatment preference. It is ethical to offer patients participation in a trial only if clinical equipoise exists within the expert medical community. The physician who has a decided treatment preference is not obligated to participate in the trial, but should ensure that his or her patients are informed about the differing opinions.

The existence of clinical equipoise is not the only condition which ethical research must meet. Good design is also a requirement. Ethical research must be built upon a sound foundation. Previous work must give good reason to believe that the proposed interventions will achieve the intended result. There should also be evidence that the research interventions are not likely to cause serious harm to the research subjects.

Justice demands that the benefits that may be enjoyed by research participants should be available to all who would willingly participate. It also demands that the potential burdens of participation should be equally shared. Issues of justice should be dealt with when clinical trials are being designed. Those responsible for the design of protocols should ensure that the inclusion and exclusion criteria are justifiable on moral and scientific grounds. Physicians who believe that some of their patients are unjustly excluded from trials should not attempt to redress this problem by entering these patients into the trial under false pretenses. This would imperil the integrity of the research. Patients who are ineligible to participate in research may gain access to new drugs and procedures through defined protocols for compassionate use.

Women have often been excluded from research studies because of justifiable concerns for the fetus during pregnancy, the future reproductive potential of the woman or her offspring, or effects on future progeny. These concerns must be analyzed carefully and fairly. However, they must be balanced with the need to include women in order for the research to examine how the drugs or procedures in question may improve the health and quality of life of all humans.

Health care institutions, funding agencies such as the Medical Research Council, and even some professional organizations now require that all research conducted by their members receive prior approval by a Research Ethics Board. This arm’s length review provides assurance that the potential risks and benefits of the research are understood and within acceptable limits, and that patients’ rights are protected. It also adds assurance that the research method is scientifically valid. Because REBs have members with a wide diversity of backgrounds, their review helps ensure that no important ethical concerns are ignored.


Weijer C, Dickens B, Meslin EM. Bioethics for clinicians: 10. Research ethics. CMAJ 1997; 156:1153-7.

Freedman B. Equipoise and the ethics of clinical research. N Engl J Med 1987; 317:141-5.

Further Reading

Riesenberg D, Lundberg GD. The order of authorship: Who’s on first? JAMA 1990; 264: 1857. Also [Letters] JAMA 1991; 265:865.

Rennie D, Yank V, Emanuel L. When authorship fails: A proposal to make contributors accountable. JAMA 1997; 278: 579-85.

Conflicts of interest. Science 1992; 257 No. 5070: 616-25.

Beecher HK. Ethics and clinical research. N Engl J Med 1966; 274:1354-60.

Tri-Council Policy on Research Involving Human Subjects. Ottawa, CA, 1998.