In 1996, George Doeschner, a 53-year-old electrician with Parkinson’s disease, flew to the University of Colorado to enroll in a clinical trial. Researchers there were testing an experimental treatment that involved injecting embryonic cells into the brains of patients with Parkinson’s, in the hopes of providing them with some symptom relief. On the day of Doeschner’s procedure, his surgeon put him under anesthesia, drilled several holes into his skull, and then closed him up again — without injecting the cells.
This was no surgical error. Doeschner had been randomly assigned to the control group in the clinical trial. Of the trial’s 40 participants, only 20 received the actual treatment; the rest, including Doeschner, were given a “sham” surgery in which their skulls were opened up but no cells were injected. Going into the operation, Doeschner knew that there was a chance that he would be assigned to the control group, but it was not until a year later — after the clinical trial was over — that he discovered for certain that he had not received the treatment.
Doeschner’s surgery took place more than 15 years ago, but the questions it inspires about the ethics of placebo use in medical research are still pertinent today. When is it acceptable to use placebos in clinical trials? Are doctors, in giving a group of patients a sugar pill or a sham procedure, violating their duty to protect their patients’ interests? Is it ever acceptable to give clinical trial participants a placebo in the process of testing a new treatment for a medical condition?
Dr. Robert J. Levine, a professor at the Yale School of Medicine and internationally renowned expert on bioethics, has written extensively on this subject. The ethics of placebo use, like many issues in human research, are highly complicated and fraught with questions about the responsibilities physicians have to their patients. There are no easy answers, but the medical community, working together with bioethicists, has reached some conclusions about how to deal with these issues.
From Therapy to Research
In the absence of effective medications, placebos were a first-line treatment for many diseases prior to the 20th century. Physicians assigned long, unpronounceable names, such as the “Tincture of Condurango,” to remedies that were of little therapeutic value, relying primarily on the power of positive suggestion to treat their patients.
Despite their widespread use in the 19th century as a medical therapy, it wasn’t until after World War II that placebos became an important tool in clinical research. In his landmark 1955 paper The Powerful Placebo, anesthesiologist Henry K. Beecher became the first doctor to quantify the “placebo effect,” claiming that nearly a third of patients across fifteen trials for different diseases experienced some improvement from placebos alone. Although his data has since been criticized, Beecher’s paper had a revolutionary effect. Scientists realized that, when testing new therapies, they needed to control for the powerful psychological effects of taking a pill — even if that pill contained nothing more than sugar. Whereas the FDA had previously approved new drugs as long as they were shown to be safe, in 1962 it began granting approval only after they were also shown to be effective through controlled clinical trials. Double-blind, placebo-controlled trials became the gold standard in clinical research. In these sorts of trials, neither the patients nor the researchers know who is getting a placebo until the end of the study.
Today, most bioethicists agree that pre-scribing placebos as a therapeutic treatment, as practiced prior to the 20th century, is deceptive and unacceptable. If placebos are used in research, patients should be informed before giving consent that they might receive a placebo treatment. The tricky part is determining when it is acceptable for a clinical trial to use a placebo as a control — a question that has sparked explosive debate within the bioethics com-munity for decades.
The relationship between the physician and the patient is at the heart of the placebo-use debate. Physicians are bound by a legal concept called the fiduciary obligation, which calls for the physician’s undivided loyalty to the interests of the patient. Therefore, is it possible for a physician to give a patient a placebo while still upholding this obligation? If a physician conducts a clinical trial knowing that half of the patients will be given nothing more than a sugar pill, is the physician truly fulfilling his or her duty? The short answer, most bioethicists say, is that it depends on the situation.
According to Levine, the use of a placebo as a control is unquestionably ethical under certain circumstances. One circumstance arises when testing a therapy for a condition for which there is no proven treatment. “If there’s no medication out there that anyone thinks is any good, then it is considered ethically acceptable to give half of the patients a placebo,” Levine said.
On the other hand, there are some circumstances under which it is clearly unacceptable to use placebos. If researchers are testing a new therapy for a serious condition that has a known effective treatment, it is not considered ethical to give placebos to any of the patients. “This would be considered a serious violation of the fiduciary obligation,” Levine said. If, for example, a researcher were conducting a clinical trial on a new medication that might prevent heart attacks, they should not give any of the trial participants a placebo if there already existed a reasonably effective medication for that purpose. Because use of a placebo in such a situation could cause death or permanent harm to participants, the researcher would be obligated to test the performance of the new drug by comparing it to that of the known treatment.
Other examples, however, are not so straightforward. Levine points to hypertension, commonly known as high blood pressure, as an example of a condition for which the acceptability of placebo use is less clear. Over time, untreated hypertension can be deadly, increasing the risk for conditions such as heart attack, stroke, and kidney failure. Yet in the short-term, it is unlikely to be fatal for individuals who are otherwise healthy and are under the observation of experienced doctors. Therefore, is it ethical to use placebo controls when testing new medications for hypertension?
Levine claims that, in these situations, it is ethical to use placebos, provided that a few conditions are met. First, the placebos must only be used in individuals with a conditions so mild that they are unlikely to develop serious complications. Second, all trial participants must be monitored carefully, and withdrawn immediately from the trial at the first sign of adverse effects. Finally, the trial must only be conducted over a short period of time — on the order of months, rather than years. “We do these sorts of trials under circumstances where nobody’s going to get into serious trouble,” Levine said.
Above all, Levine stresses the importance of acknowledging the complexities of human research ethics. “Almost everything you read in the papers is sound-byte ethics,” he says. “Things are virtually never as simple as they appear to be in the media.” The complicated issues raised by the ethics of placebo use force physicians to tackle difficult questions about their responsibilities to their patients, the place of scientific research in society, and the moral duties that bind us together as human beings.
About the Author
Bridget Kiely is a freshman in Calhoun College planning to major in Molecular, Cellular, and Developmental Biology.
The author would like to thank Dr. Levine for sharing his expertise concerning placebo use in bioethics.
Levine, Robert. “Placebo controls in clinical trials of new therapies for conditions for which there are known effective treatments.” The Science of the Placebo: Toward an Interdisciplinary Research Agenda. Ed. H. Guess, A. Kleinman, J. Kusek and L. Engel. London: BMJ Books, 2002.