Most people think of placebos as harmless "sugar pills" given in clinic trials to some participants so that medical researchers can gauge the effects of the real drug on others. But in some trials, the "placebo effect" proves to be as strong as that of the drug. Consistently 30 percent or more of the subjects given placebos will show some improvement by taking the dummy pills.
So over the decades a small band of researchers has taken a hard look at those pills. Are they really effective? Should they play a role in medical therapy?
A landmark study in 2001 concluded that they weren't useful. It "found little evidence in general that placebos had powerful clinical effects," the New England Journal of Medicine (NEJM) reported.
But that hardly put the matter to rest as new studies emerged. A March article in The New Scientist summed up the problem: It listed the placebo effect as one of "13 things that do not make sense" to science.
Today the definition of placebo effect has broadened beyond dosing with inert pills to include questions about whether healing is still in part an "art" and issues such as how the relationship between doctors and patients affects treatment outcomes. In its last fiscal year, the National Institutes of Health approved 14 clinical studies that aim to better measure and understand the placebo effect. Some of the most exciting work for scientists has come when they have scanned the brain and measured actual biochemical effects of placebos at work.
Among the placebo studies that have been reported since the NEJM article:
• A report in the journal Psychosomatic Medicine earlier this year showed that patients with problem coughs who were given a placebo (a dose of Vitamin E) significantly reduced their coughing compared with a group of similar coughing patients who received no treatment.
• Heart patients who faithfully took either a heart medication or a placebo fared equally well, according to an article in the journal Nature last November. The authors theorized that the faithful taker of a drug or placebo might also be faithful in following other recommended routines, such as dieting or exercising, which were creating the real benefit. Or, more controversially, they said, the study might also suggest that a patient could be as well off taking a less-expensive placebo as taking a drug.
• In an April 2004 study reported in the Archives of General Psychiatry, 20 Parkinson's patients received surgery in which human dopamine neurons were transplanted into their brains. Twenty other patients received "sham" surgery, in which nothing was transplanted. Neither the patients nor the medical staff knew what each patient received (the "double blind" standard). After a year, those who declared themselves better, and those whom the staff concluded were improved, came from both the treated and nontreated groups.
• In an experiment published in the journal Science last year volunteers were subjected to a harmless but painful electrical shock. When they were told that they had had an antipain cream applied to the site (really an inert placebo), they rated the pain as less intense. What's more, brain scans showed that areas of the brain associated with pain indeed were less active.
The study suggested a practical use of the placebo effect by doctors, said one of the researchers. If you share your expectation that a drug will work with your patient, "you enhance the effect," he concluded. But if a doctor said, or even hinted, that his treatment might not be effective, "it would be much less likely to be effective," the researcher said.
In Israel, 3 of 5 doctors surveyed reported using placebos in their practice at least once a month. And 94 percent said the inert pills were effective to some degree, according to a study reported in the British Medical Journal last year. "Clearly wider recognition of [placebo use], and debate about its implications, are needed," the Israeli researchers said.
But is that honest? To most practitioners of Western medicine, dispensing sugar pills or using "sham" surgery in actual practice teeters on the edge of unethical practice. They view it as withholding real treatment and as establishing a deceptive relationship between doctor and patient.
Others aren't so sure. If "we look at the first rule of a medical practitioner of 'first, do no harm' there could be an argument made for saying that using a placebo, if it does no harm and if it does help, could be quite a reasonable solution to problems," says Rory Coughlan, a health psychology professor at Trent University in Peterborough, Ontario.
A fuller understanding of the placebo effect would embrace the entire doctor-patient relationship, says Anne Harrington, a professor of the history of science at Harvard University and author of the book "The Placebo Effect." A better way to think about the placebo effect would be as "the sum total of effects," she adds. Part of that experience might be going into a waiting room where the patient sees lots of certificates on the wall, and where a confident doctor walks in with a white coat that designates him as a professional expert. "All of these things produce the placebo effect," she says. "There's a placebo effect probably that's involved in every treatment, not just treatments using sugar pills."
"In a sense, placebo is the clinician's best friend," Mr. Coughlan says. "But in the scientific view of medicine, it's the enemy because we're always trying to prove that something is better than placebo."
Even those who study placebos acknowledge that the jury is still out.
The 2001 NEJM study "still has not been refuted," points out Ted Kaptchuk, a professor of medicine at Harvard University who has been involved in a number of clinical trials studying the placebo effect.
That study - titled "Is the Placebo Powerless?" - reviewed 130 clinical trials. While inert pills "had possible small benefits" for the treatment of pain, the study concluded, "outside the setting of clinical trials, there is no justification for the use of placebos."
"There's still a need for rigorous research in terms of the power - or the lack of power - of the placebo in conditions that have to do with real illnesses over time," Professor Kaptchuk says.
To be effective, these studies need to be three-tiered, comparing placebos not only to active drugs but also to doing nothing - no treatment at all, researchers say.
Despite the 2001 NEJM report, "there has been a change in the last couple of years" in recognizing the importance of the placebo effect, argues Daniel Moerman, a medical anthropologist at the University of Michigan at Dearborn, who's studied the placebo effect for nearly three decades.
He and others in the field also urge more research into why placebos work.
"That's a big, tough, wide-open question," says Professor Moerman, author of the book "Meaning, Medicine and the 'Placebo Effect.' " "My own feeling is that there's a lot of different things going on, and if you think it's just one thing, you get lost," he says. "The one thing we can know about 'placebo effect' is that it isn't caused by placebo [pills], because they're inert."
Can placebos play a significant role in healing? "I don't know," Kaptchuk says. "I think so. But I don't think the clear evidence is out there.
"I think that the way medicine has constructed itself or parsed itself, there's no place for placebo," he adds. "Placebo is really about the patient-doctor relationship; it's about how people respond to symbols; it's about the intangible parts of medicine," such as the inherent capacity of human beings for self-healing.
Most of the drugs that people take probably are not doing much more than placebos or "natural history" - people getting better without treatment, Kaptchuk says. "And that's important to know. We have a society that's taking lots of drugs."
Perhaps 50 percent of the effect of seeing a physician is "based on the doctor-patient relationship, the art of medicine, and it's not acknowledged," he adds. "And I think that's what people want more of. We want high-tech, and we want sophisticated interventions. But we also want the art."