The placebo effect is amplified when doctors appear likeable and competent

A lot of clinical research tries its best to find the true effects of a treatment above and beyond the placebo effect – that is, the benefits that can arise purely from a person's expectations that an intervention will be helpful. A new study in Health Psychology takes a different approach: instead of always seeing the placebo effect as "a nuisance variable with mysterious impact", argue Lauren Howe and her colleagues at Stanford University, we should try to find out more about how to enhance it, and how to diminish its harmful twin, the nocebo effect (when negative expectations can lead to harmful effects). To this end, the researchers examined the possible moderating influence of a physician's demeanour, finding that the placebo effect is enhanced by the impression of warmth and competence.

Howe and her team write: "Among the many demands of a career in medicine, physicians have been increasingly directed to build rapport with their patients (e.g., by exhibiting empathy). This research suggests a compelling reason for why physicians should pay attention to these psychological and social forces: They can impact physiological health outcomes."

The study involved 164 healthy participants who thought they were taking part in an investigation into food preferences. As part of this, they were told that they would first need to undergo an allergy test. A female doctor gave them an injection of histamine, which reliably induces an allergic reaction in the skin, including swelling and redness. After the injection, the doctor applied some inert cream to the injection site, which she told some participants would reduce any allergic reaction, while she told others it would increase their allergic reaction.

Crucially, the participants were divided into four further groups, for which the doctor adapted her demeanour and behaviour to be either: highly warm and competent; low on warmth and competence; high competence but low warmth; or low competence but high warmth.

For instance, to appear warm, the doctor might ask the participant's name and give her own, but in the low warmth condition she would not bother with these introductions. To appear competent, the doctor made eye contact and spoke eloquently (but did the opposite to appear less competent). Environmental cues added to the effect – for instance, in the high competence condition, the doctor's office was clean and tidy whereas it was messy and disorganised in the low competence condition.

After a histamine injection, the physiological effects appear on the skin rapidly over a course of minutes. The doctor traced the size of the unfolding allergic reaction several times for each participant and these traces were then sent for measuring by a research assistant who was unaware of which participant was in which trial condition.

The results showed that the placebo effect was enhanced when the doctor appeared both warm and competent: that is, the allergic reaction was smaller when the doctor said the cream would have this effect, and when she was more likeable and credible (when she was warm or competent, but not both, the placebo effect was of intermediate size).

In real life, of course, doctors don't only raise positive expectations about treatment (which can induce a placebo effect), but they also warn about side-effects and adverse consequences, which may provoke self-fulfilling negative expectations in the patient. The researchers had thought that physician warmth and competence might also amplify this nocebo effect: that is, increase the allergic reaction when the doctor said the cream would have this effect, but there was no evidence for this happening.

It's important not to over-generalise these findings: for instance, the participants were all healthy; the observed placebo-enhancing effects occurred over a very short time-scale; and it's not clear if the findings would be the same with a male doctor. Nonetheless, the study introduces an elegantly simple procedure for investigating factors that may interact with the placebo effect, and the fascinating findings are ripe for further investigation.

"Future research could elaborate on the precise mechanisms through which expectations impact physiology," the researchers said. "For example, the manipulation of positive and negative expectations in this study could have served to decrease or increase patient anxiety, prompted cognitive reappraisals of the treatment, and/or changed where participants directed their attention during the medical procedure. We look forward to studies that explore these and other intriguing possibilities."