An independent charity for science and integrity in healthcare

Read the latest HealthWatch newsletter:  Issue 106, Autumn 2017

The placebo effect: let's unwrap it and then tap it

The word placebo comes from the Latin word to please, implying that after treatment the patient is pleased, grateful and convinced that what was done has helped; but that, in fact, there are reasons to think that either

  1. the improvement would have occurred anyway, even without any treatment, or
  2. that  if there really was benefit  the mechanism of this was purely psychological and subjective.


No definition is completely satisfactory. Some doctors use the term placebo only when the person giving a remedy knows that it is either certainly, or probably, going to have no more than a placebo effect.

One example of this would be when research is being done  perhaps to test the benefits (and side effects) of a new drug on volunteer patients by comparing it with an inert placebo tablet that looks the same. Whether studying benefits or side effects, unless a comparison of this type is made, the findings are difficult to assess. Only thus will non-specific psychological effects be clearly separated from physical effects due to the active principle in the drug. And it is best that the doctors and nurses involved are also unaware which is the real drug and which the dummy one, for in that way (1) their assessments of both benefits and side effects will be free from bias, and (2) all patients will get equally good care, observation and support.

Another example of the variety of definition is when, in ordinary clinical practice, an active drug rather than an inert one is given to patients. This might be done partly so that the patients know that at least something is being tried. If they later say that it helped them it may be felt that this is likely to be due mainly, if not entirely, to the placebo effect.

Some doctors would also include under the heading of placebo therapy situations where the doctor (or alternative medicine healer) does not consider that he is giving placebo therapy, although comparative studies suggest that the placebo effect is really all it is. He or she may have as much faith as the patient that any apparent benefit is due to the active agent in the drug.

Suppose, for example, that many patients consider that a particular cough remedy helps them. Careful comparisons  when volunteers do not know whether they are having the real treatment or the sham one  may reveal that the proportion of patients reporting improvement in their cough is as high after the sham treatment as after the real one.

Finally, no matter how effective a remedy may be, it is likely that some of the benefit is psychological. This is certainly not placebo treatment in the ordinary sense, but the placebo effect has played a part. How big a part? HealthWatch recognises the extra benefit to the patient that may come from such placebo effects but believes it is important to try and identify how much comes from this effect — and how much from a 'real' (ie physical) effect.

Faith healing

Some patients have such faith in the doctor (or healer)  perhaps in his or her air of authority, warm smile, confidence or touch  that nothing will ever convince them that the treatment given was not of benefit. Or it may be that what impressed them was the dramatic nature of the surgical operation that was done. Or the marvellous technology evident in the machine that treated them. Or perhaps the unusual shape and colour of the tablets taken.

Sometimes what is done is further boosted by a strong verbal dose of what psychologists call 'suggestion'. The patient is told 'within 24 hours you will feel better'  or even 'when you wake up to-morrow your backache will be gone'. There is usually an element of deception here because the therapists  if they are not deceiving themselves - know that nothing in the real world ever works quite so well every time.

It is likely that the proportion of grateful patients  at least in the short term -will be increased by the apparent faith, confidence and enthusiasm of the practitioner. Nevertheless, nearly everyone dislikes the thought of any deception. The ethical dilemma here is that there may be more benefit, fewer side effects (for these, too, can be partly due to suggestion') and perhaps less cost, if there is an element of what might be called paternalistic or benevolent deception.

Subjective improvement

In conditions that fluctuate in severity it is not uncommon for as many as two out of every three patients to report subjective improvement after a treatment that, unknown to them, contains no active agent. This has been observed, for example, in relation to the discomfort of varicose veins or the irritation of a skin rash.

It has also been found that, even when it is carefully and fully explained to the patient that sham (placebo) treatment is given, some will still be convinced that it has helped them.

Even in cases of severe pain, such as might be caused by injury, a sizeable minority of patients will report marked relief after, for example, an injection of sterile water. This happens to as many as a third of patients in most studies. The duration of benefit, however, may not be as great as it is when a real painkiller is given.

Objective benefit

Much less impressive is the evidence that placebo therapy can produce measurable objective benefit in a disease, such as quicker healing than would otherwise occur, or, in serious disease, prolongation of life. This, of course, is by no means impossible. We know that a person's state of mind can affect the body in many ways. This is certainly an idea that many people are rather in love with, so to speak, and badly want to believe, for example in the case of cancer. But so far the evidence from comparative studies suggests that such an effect, if there is any, is slight.

Once again, HealthWatch urges that in both mainstream and alternative medicine there should be more comparisons between different remedies, making as sure as possible that the groups being compared are alike apart from the way in which they are treated. This has been described in another of our Position Papers (Newsletter 10, Summer 1992). There will always be some patients who do far better than average. What counts is whether there are significantly more of such patients in one group than the other.


One of the many worldwide problems now facing health workers and the populations they serve is that whereas most placebo therapy used to be cheap and harmless (often consisting of just a bottle of medicine) it is now more likely to be expensive - and more likely to carry a risk of side effects. HealthWatch urges more studies so that decisions can be based on reliable information as to the extent of the placebo effect. More studies would help us to know whether equal benefit could be achieved in cheaper ways - thus releasing both money and resources for other needs.