Intended for healthcare professionals


The power of placebo

BMJ 1996; 313 doi: (Published 21 December 1996) Cite this as: BMJ 1996;313:1569
  1. E Ernst,
  2. A Herxheimer
  1. Professor Department of Complementary Medicine, Postgraduate Medical School, University of Exeter, Exeter EX2 4NT
  2. Emeritus fellow, UK Cochrane Centre, 9 Park Crescent, London N3 2NL

    Let's use it to help as much as possible

    Links are revealed this week between the colour of a pill, its name, and its pharmacological action (pp 1624, 1627).1 2 This news will come as no surprise to many. Pink pills and tonics were the mainstay of many physicians—perhaps their main resource—before the era of antibiotics. But what are the active ingredients of the placebo effect and how can we make the best use of it?

    Many non-specific concomitants of treatments help to determine the direction and size of the placebo effect. These can be placed on a continuum ranging from the tangible to the intangible.3 The form of medications, touch, words, gestures, and the ambience of the consultation can all play a part in conveying a doctor's confidence in a treatment, empathy with the patient, and professional status.4 5 6 Non-specific aspects of the remedy itself can also have a powerful influence; the more invasive it is, or the more actively it involves the patient, the larger the placebo effect.7 8

    All of these determinants relate to the fact that the mind can influence the body. This notion has always been accepted in good medical practice, and much evidence exists to show that the effect is clinically relevant. It would be desirable to know how the use of placebo effects differs between mainstream and complementary practices. Preliminary survey data suggest that patients who use both forms of treatments are more impressed by the therapeutic encounter in complementary rather than mainstream medicine (Ernst E, unpublished data).

    We know far too little about the importance of the non-specific effects and their interactions with specific treatments. We know that patients who receive a reasonable explanation from a member of the surgical team about an intervention will fare better than those who get no such information.9 One can argue about whether this information is part of the treatment or whether it is a non-specific effect. But the effect itself is important, must be studied, and should be optimised.

    Systematic research on placebo effects has been neglected for the past 30 years; placebos have been used largely as a tool for reducing bias in clinical trials.10 Nurses routinely apply placebos in clinical practice,11 but most doctors still feel uncomfortable about the subject because using a placebo seems to imply deception.12

    A multidisciplinary research programme is needed to define and examine the most important questions about non-specific factors and their effects. Further studies on the “best” colours and other properties of tablets, capsules, and patches would be interesting but do not perhaps offer much scope for improved effectiveness over the many purposefully distinctive products now available. The issues are complex. For example, heart shaped patches worn over the heart releasing transdermal glyceryl trinitrate are probably marvellous placebos as well as having a pharmacological effect, but they cost much more than other more versatile forms of the drug. Is the balance between cost-benefit acceptable?

    Holm and Evans (p 1627) raise the possibility that the names of drugs could influence their actions.2 This requires experimental investigation. An effort to control names seems necessary on ethical grounds to prevent implied claims that are unjustified or exaggerated. The five yearly review of product licences by Britain's Medicines Control Agency could be used to give an expiry date to the worst names, and the licensing rules could be used now to prevent names that make unjustifiable claims. Perhaps there should be an international prize each year to be awarded for the most honest, optimistic, and pleasant product name, and an award for the worst name of the year.

    Meanwhile, all doctors should be encouraged to look at their own practice, to examine the non-specific ingredients that they use daily, and those that they do not use. Giving greater attention in daily practice to “adjuvants” (specific as well as non-specific) could considerably increase effectiveness and efficiency—for example, by saying more useful things to patients in better ways. Methods will be needed for implementing such approaches. Until they are available, good common sense and old fashioned bedside manners might already take us far—as they say, when all else fails talk to your patient.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    View Abstract