Placebos in Research

To the Editor: In a letter to the editor in the May 1998 issue, McCarthy (1) posed important and potentially disturbing questions about the ability of patients with schizophrenia to fully understand the implications of receiving a placebo in a clinical trial and to provide genuine informed consent.

We fully agree with McCarthy that all necessary measures should be taken to assess and facilitate patients' understanding of the research process and to safeguard their rights. Investigators and other parties, such as institutional review boards, must be vigilant in this regard. Clearly, the good of society—in particular the interests of other patients with the same disorder—cannot be used to justify encroaching on the principles of beneficence and autonomy as they apply to the individual patient.

However, in considering the use of placebos in research, several other issues need to be examined, which we address briefly below.

Are there viable alternatives to placebo controls for research? Alternatives to placebo controls include active controls, who receive the standard drug for the condition, and historical controls, which are based on data from previous placebo-controlled studies with similar patients. Several writers have argued that placebo controls are essential to clinical trials in certain circumstances and that alternative control groups are inadequate. Given the need for placebo controls to permit meaningful conclusions to be drawn from clinical studies, it may be argued that not using placebo controls would be unethical because patients receiving the experimental drug would be exposed to potential risks in a study of questionable scientific benefit.

What is the potential harm to patients receiving placebo? Conventional antipsychotics are not uniformly effective in schizophrenia, in particular against negative symptoms. Between 30 and 50 percent of patients are exposed to the risk of troublesome adverse effects for minimal or no benefit (2). Also, contrary to popular belief, as many as 55 percent of patients who receive placebo in the acute phase do not worsen, and up to 25 percent are much improved (2,3).

In addition, some studies have found that patients who relapse while on placebo do not suffer any long-term consequences. For example, Curson and associates (4) followed patients who had participated in a short-term trial of relapse prevention. They found that although in the original trial 66 percent of patients on placebo had relapsed compared with 8 percent of those on the active drug, there were no differences between the two groups in the number of relapses after the end of the trial or in any clinical or social variable over a seven-year follow-up period.

How well informed is the informed consent of patients with schizophrenia who participate in clinical trials? Carpenter and associates (3) have cited recent research suggesting that approximately 75 percent of patients with a diagnosis of schizophrenia incorporate information and make decisions much like comparison groups do when dealing with consent issues.

What can be done to protect the individual patient? Explicit guidelines, such as those adopted by Loma Linda University (5) or proposed by Carpenter and associates (3) may aid in protecting psychiatric patients, particularly those with a psychotic disorder, who are enrolled in placebo-controlled trials. Although researchers must continue to be very concerned and vigilant about protecting the rights of subjects participating in research, the use of placebos in research has unique scientific benefits. With proper care, it need not be incompatible with concerns for the safety of the subjects.

Dr. Mago is associated with the department of psychiatry at the University of Pennsylvania in Philadelphia. Dr. Petty is clinical director of the neuropsychiatry program and Dr. Wolpe is associated with the Center for Bioethics at the University of Pennsylvania.