If patients complaining of back pain are administered only a placebo, for example, typically about a quarter of them will report a relief or diminution of the pain. Remarkably, not only do patients report improvement, the improvements often are objectively measurable, and the same improvements typically are not observed in patients who did not receive the placebo.
Because of this effect, government regulatory agencies approve new drugs only after tests establish not only that patients respond to them, but also that their effect is greater than that of a placebo (by way of affecting more patients, by affecting responders more strongly or both). Such a test or clinical trial is called a placebo-controlled study. Because a doctor's belief in the value of a treatment can affect what his or her patient believes, such trials are usually conducted in "double-blind" fashion: that is, not only are the patients made unaware when they are receiving a placebo, the doctors are made unaware too. Recently, it has even been shown that "mock" surgery can have similar effects, and so some surgical techniques must be studied with placebo controls (rarely double blind, for obvious reasons).
The size of the placebo effect is controversial. A part of the controversy may be due to the fact that patients who have been given a drug (or a placebo for that matter) will oftentimes report improvement earlier and more eagerly in order to please and thank the care giver. These patients may even do this when there is no real physical improvement attained. Hence, the uncertainty over the real size of the placebo effect. An often quoted number is that about one third of patients improve on a placebo, but a recent study has called that number into question, claiming that the effect is much smaller, if it exists at all.
It has been argued that the placebo effect may be simply an experimental artefact that arises because the patients in any study are selected to show a particular characteristic (symptom of illness). Since the sample is thus initially skewed away from the mean, it is likely that during the course of the study there will be a natural tendency for the sample to revert towards the mean - i.e. show better than average levels of recovery.
The reverse effect, sometimes called the Nocebo Effect, is the result of negative expectations by the patient. Some patients who believe that they should be ill can exhibit genuine symptoms.
The Nocebo Effect may or may not be distinguished from older concepts such sociogenic illness or psychosomatic illness. In any case, distinguishing between these effects can be a difficult problem for medical semiotics, or the study of how to interpret the patient's description of their ailments.
The strictest definition would be the negative effect produced by an expectation that a medicine or treatment does not work in general or for the patient in particular. Sociogenic illness and psychosomatic illness might therefore be broader terms where a known or unknown cause with cultural or psychological factors is involved.
Example: I believe that cough medicine does not work. I suffer, therefore, from the Nocebo effect--the medicine is less effective for me than for a patient who believes the medicine actually prevents coughs.