Reprinted from BMJ 1994;309:667

Vernon M S Oh glosses over ethical questions about the use of placebos and ignores problems in defining the placebo concept.¹ Because he uses Shapiro's definition of placebo,² which is what Grunbaum calls an intentional placebo,³ he does not make a clear differentiation from the inadvertent placebo. Placebos are not always given wittingly, as Oh acknowledges. He is prepared to accept that homeopathy or even some conventional treatments may work through the placebo effect and that patients with cancer gain in optimism because of complementary treatments. Yet exhorting doctors to use placebos in certain circumstances in appropriate patients implies that doctors should deliberately use simulated effects. Oh is apparently encouraging doctors to deceive their patients.

The action of placebos emphasises the power of suggestion. A history of Freudian psychoanalysis is the rejection of the suggestion of hypnosis for the transference of the analysis. Frank postulated that all treatments comprised a specific element and a non-specific placebo element aimed at counteracting demoralisation.⁴ Indeed, Frank suggested that the success of the many forms of psychotherapy was due to a placebo effect rather than the distinctive features claimed by the different therapists. Like Oh, Frank encouraged us not to minimise the value of the placebo effect. Frank, however, proposed taking advantage of the dynamic relationship between doctor and patient, whereas Oh suggests that doctors should give dummy drugs or other phoney interventions.

Therapeutics generally does not pay adequate attention to psychological and social variables. Classic papers such as Beecher's have a sobering implication.⁵ The recognised surgical procedure of ligating the internal mammary artery for angina was shown to be ineffective by well controlled studies that used sham bilateral skin incision as a control. The ethics of these studies has been questioned because of the deception entailed in the sham surgery. The advantages to patients who receive inadvertent placebos may, however, justify more studies of this sort, instead of subjects being asked to join in the deception of a double-blind trial through informed consent, as is more usual in current research. Placebos could be used better in research and must not be used to exploit patients.

1. Oh VMS. The placebo effect: can we use it better? BMJ 1994;309:69-70 (9 July).

2. Shapiro AK. A historic and heuristic definition of the placebo. Psychiatry 1964;27:52-8.

3. Grunbaum A. The placebo concept in medicine and psychiatry. Psychol Med 1986;16:19-38.

4. Frank JD. Persuasion and healing. Revised ed. Baltimore: John Hopkins University Press, 1983.

5. Beecher HK. Surgery as placebo. JAMA 1961;176:1102-7.