Introduction

Every medical man commits that act of treachery, Mr Blake, in the course of his practice…. Every doctor in large practice finds himself, every now and then, obliged to deceive his patients.—Wilkie Collins, The Moonstone

The placebo, it is agreed, is an excellent aid in determining the specific pharmacological effects of pharmaceutical agents. Any criticisms of placebos have usually been directed at the morality of giving an inert substance to patients hoping for effective treatment.1 2 Indeed, it might be argued that the object of drug development is to make the use of placebos unethical: to find a treatment that is so effective that it will be unacceptable to withhold it in the future. The debate has focused on acute conditions (especially if serious), where the ethical stakes are highest. For chronic conditions (especially if they are less severe) the situation is different. Here the patient may have to live with the condition for many years and, in his or her own interest, be prepared to try a number of treatments–including, from time to time, no treatment at all. With informed consent, the use of placebos in randomised clinical trials may then be uncontroversial.

Ethical objections

It is often overlooked, however, that fully informed consent and a placebo can go together only if the clinical trial is randomised.3 Where this happens, the patient's doctor may say: “If you come on …