Mind triumphs over matter

Placebo responses can be really quite unhealthy. The problem isn't the placebo effect itself, though. It lies mainly in the response to the idea that a medical treatment may work simply because the person treated believes it will.

For those who mistrust orthodox medicine and resist the principles on which it rests, the placebo effect nourishes the feeling that mind is above matter. For orthodox practitioners, it is the woodworm in the desk of reason: it reminds them that they are faith healers too. The placebo effect is a profoundly subversive one.

That, however, is largely because it is known widely but not well. Many people now assume that it can work on almost any disease, from colds to cancers; in response, some revisionist researchers have dismissed it as a myth. It certainly had a founding moment fit for a legend, in a field hospital during the Second World War. Lacking morphine to treat a soldier who needed surgery, a nurse injected him with salt water: it had the necessary analgesic effect.

The anaesthetist for the operation, Henry Beecher, was profoundly impressed, and set out to make his colleagues share his

awe. Although doctors had long been aware that pills have power even without drugs in them - indeed, the understanding may be as old as the practice of healing - Beecher wanted them to see placebos as much more than a sop for dissatisfied patients.

To do so, he had to stretch the available evidence. Dylan Evans, a psychologist currently researching evolutionary robotics at the University of Bath, depicts this as the start of a tradition of inadequate analysis which has preserved the mystery of the placebo effect ever since.

Two particular shortcomings are the failure to make comparisons with patients left untreated, and the failure to apply "blind" procedures in which neither practitioners nor patients know who is receiving a treatment and who is getting the placebo.

These considerations are difficult to resolve when the treatment does not come in a pill, but in a complex procedure like acupuncture or psychotherapy.

One group of researchers worked around the problem by getting university professors to conduct psychotherapeutic sessions. The professors' patients did as well as those treated by trained therapists.

EVANS himself is a firm believer in what he calls the "belief effect", but within firm limits. The evidence is good, he judges, that placebos may relieve pain and inflammation; there is also some evidence that they may act against ulcers, depression and anxiety. These conditions, he argues, are all related to a physiological process called the "acute phase response", which comprises swelling, heat, pain, inactivity and loss of appetite.