Expectation can be an effective drug. A placebo stimulates the brain in the same way as drug treatment in Parkinson's disease, shows a Canadian study. Both increase the release of the brain chemical dopamine, fuelling recent controversy over whether the placebo effect exists at all.

Thought to affect around 30% of patients, the placebo effect, in which patients benefit from treatment because of expectation alone, is a long-standing medical conundrum. Drugs are generally approved on the basis of their effectiveness over placebos.

Comparable levels of dopamine, the brain chemical lacking in patients with the neurodegenerative disease Parkinson's, are released after injection of a drug or a placebo, researchers at the University of British Columbia in Vancouver, Canada have found1.

"The magnitude of the response is striking," says Jon Stoessl, who led the brain-imaging study. "A lot of effect may be coming from the placebo," he says, highlighting the importance of placebo-controlled trials.

Though detected in the movement control circuits that degenerate in Parkinson's, the dopamine boost is triggered by expectation of a therapeutic 'reward', the researchers suggest. Such reward-based release may occur in other medical conditions susceptible to placebo effect, they speculate.

Keep taking the pills though, is the message: higher doses of the drug outstrip the effects of the placebo. And placebo effects often do not last as long as treatment, warns Parkinson's expert Joseph Friedman of Brown University School of Medicine in Providence, Rhode Island. "We don't know how sustained the benefit is," he says.

Parkinson's pioneer Curt Freed of the University of Colorado in Denver also cautions that brain imaging does not reflect clinical outcome. Movement problems in Parkinson's may be influenced by psychological factors, he says. Unable to walk, some sufferers can still cycle, Freed points out.

A controversial review of clinical trials recently called the very existence of the placebo effect into question2. "If there is an effect it's probably quite small," concludes one of its authors, Asbjorn Hrobjartsson, a medical philosopher at the University of Copenhagen in Denmark.

The problem is that many studies claim a placebo effect only by comparison with patients who receive a drug, Hrobjartsson explains, omitting those who receive nothing. Such studies may ignore health improvements in the natural course of the disease, he points out.

In trials that offer only three different treatments - drug, placebo or nothing - the placebo effect may be slight, concedes Stoessl, because of patients' low expectations. He offered patients a one-in-four chance of receiving a placebo. Study design can have "a profound impact" on the ability to detect the placebo effect, he says.

Although some argue that administering placebos is unethical when a potentially effective treatment is available, Stoessl makes the case for their continued use. "I suspect the controversy will go for a while," he says.