The standard explanation for why people respond to placebos is a psychological one: people are suggestible, or they are neurotic, or something like that. But consider an interesting experiment described by Richard Gracely, one of the leading pain researchers in the United States. It shows clearly that clinicians - doctors, dentists, nurses, and so on - play a very important role in this response.

What the doctor knows makes a difference

Sixty people who were having their wisdom teeth removed participated in an experiment designed by Dr. Gracely. They were told they would receive either placebo (which might reduce the pain of having the tooth removed, or might do nothing), naloxone (which might increase their pain, or do nothing), fentanyl (which might reduce their pain, or do nothing), or no treatment at all. Subjects were all recruited from the same patient stream, with consistent selection criteria by the same staff. The tricky part of the experiment is this: What Gracely was actually studying was not so much patients as clinicians. In the first phase of the study, the clinicians (the dentists and nurses) - but not the patients - were told fentanyl was not yet a possibility because of administrative problems with the study protocol, yielding the PN group. In the second phase, a week later, clinicians were told that the problems had been resolved, and now patients might indeed receive fentanyl, yielding the PNF group. Figure 4.1 shows the effects of placebo treatment in the two groups.