In Reply:—

We emphatically agree with Drs. Kyle and Connelly that Beecher’s legacy 1will be for naught if we allow our commitment to informed consent for research to wane. Indeed, we see our proposal as reinvigorating this commitment by emphasizing its spirit and refocusing our efforts toward where they are most needed.

Our proposal questions the orthodoxy that automatically assumes and requires specific informed consent for all types of research. In our editorial, we mention a hypothetical randomized controlled trial comparing two brands of disinfectant soap for preparing patients before surgical skin incision, in which half of the operating rooms would use one brand and the other half the other brand. We assume that both brands have been in standard use and that the purpose of the trial is simply to assess which brand is associated with the lowest rate of postoperative wound infection. In this hypothetical example, consider the following:

  • 1. The patients have given a general consent for treatment (although not specific consent to be part of this project).

  • 2. There is no a priori  reason for believing one of the soaps to be superior to the other.

  • 3. The patients could, and probably would, receive either one soap or the other (in an unpredictable and unsystematic random fashion), but without systematic observation of the results.

  • 4. We have no reason to believe that the nature or the severity of the side effects of the two treatments differ in any significant way.

  • 5. A “reasonable person” would have no reason to choose one soap over the other. 2 

Now it is true, of course, that a patient could have a known allergy to an ingredient in one of the soaps but not the other, and that this would be an excellent reason for not randomizing that patient. It would be important to ask patients about this possibility. But note that this is true whether or not the patient is enrolled in the trial and has nothing to do with whether it is ethically mandatory to obtain specific informed consent for the research per se .

Based on this analysis, we conclude that the value of specific informed consent in this case is nothing more than symbolic. By requiring physicians to engage in a process of informed consent when it is only symbolic and without substantive value, we undermine our commitment to the process when it really does matter (which is most of the time).

Furthermore, in this case, not only does the process of obtaining informed consent lack value, it unnecessarily impedes the acquisition of useful knowledge. If, for example, the project calls for stocking half the operating rooms with one soap and the other half with the other soap, then there are real logistical hurdles if, for purely capricious reasons, a patient does not want to be included in the trial. Furthermore, it raises questions as to the treatment that the patient should receive if not enrolled in the study. Should the alternative soap be obtained specially just to demonstrate that the patient is not in the trial? Or would it be acceptable to use the “assigned soap” on this patient but just not collect data on the outcome? The point is that when we blindly follow the rules without regard for the underlying value and spirit of the concept, we end up having to address questions that are devoid of meaning.

This example may seem trivial. Indeed, that is a part of the point we are trying to make. In most circumstances the value of obtaining specific informed consent for research is not trivial and should be given serious attention. When we insist on rules and procedures solely for their own sake, however, we risk eroding our commitment to the very important principles on which they are based. This is the risk to Beecher’s legacy that we must seek to avoid.

Kopp VJ: Henry Knowles Beecher and the development of informed consent in anesthesia research. A NESTHESIOLOGY 1999; 90:1756–65
Truog RD, Robinson W, Randolph A, Morris A: Is informed consent always necessary for randomized, controlled trials? N Engl J Med 1999; 340:804–7