When designing a controlled clinical trial, it is essential that the diagnostic, therapeutic and preventive interventions in each arm of the clinical trial (control and experimental) be in a state of clinical equipoise. A state of clinical equipoise exists if there is genuine uncertainty within the expert medical community (not necessarily on the part of the individual investigator) about which arm of the study contains the more effective intervention. There must be an honest professional disagreement among experts regarding the preferred diagnostic, therapeutic, or preventive intervention.
As a general rule, research subjects in the control group of a trial of a diagnostic, therapeutic, or preventive intervention should receive an established effective intervention. In a placebo-controlled trial, a percentage of the subjects will be assigned to a group being treated with a placebo. The placebo has no known treatment value. Requiring subjects to forego an intervention when assigned to the placebo-treated group is morally acceptable only under very specific circumstances. When a protocol is designed using a placebo (or no treatment control arm), the default position is that the placebo must satisfy the requirement for clinical equipoise. In most cases this requirement can only be satisfied if there is no standard treatment for the subject’s condition or if the subjects enrolled in the study are not responsive to the standard treatment. The use of a placebo control may also be acceptable in studies of relatively minor medical conditions, such as mild hypertension, where a competent practitioner could reasonably forego prescribing treatment. In all studies in which placebo controls are incorporated into a protocol’s design, the absence of treatment must be consistent with competent medical care. When considering the justification of a placebo control, IRBs will often take into consideration the following issues: the safety and efficacy of other therapies (if available), the maximum total length of time of placebo exposure, the greatest potential harm that may come to someone as a result of not receiving other effective treatment, and the procedures in place to safeguard participants receiving a placebo.
The following list identifies conditions where a placebo-controlled trial would be considered ethically acceptable:
- There is no therapy known to be effective for the treatment that will be evaluated.
- The new therapy targets symptoms (e.g., a headache or a runny nose) which present few risks and are short-lived.
- The clinical trial design is such that withholding a known effective therapy is extremely unlikely to result in death or disability (e.g., a typical study of a new antihypertensive or oral antidiabetic drug).
- Studies in which both of the following conditions are met: an active comparator is likely to yield an inconclusive result and in which withholding the known effective therapy is unlikely to result in death or permanent disability (e.g., a trial of new antipsychotic drug for schizophrenia).