The Placebo Effect: What It Shows About Mind Over Body, and What It Does Not

Few scientific ideas are quoted as often in spiritual circles as the placebo effect. If a sugar pill can make people feel better, the argument goes, then the mind must have power over the body, and teachings that say thought shapes health must be onto something. The real picture is more interesting and more limited. Placebos do something, and what they do tells us about expectation, attention and the brain. Careful research has also shown where the effect stops. This article covers the history, the strongest modern studies and the honest limits, and then asks what it all means for New Thought claims about mental healing.

Where the Word Comes From

Placebo is Latin for "I shall please." By the late eighteenth century the word was used in medicine for a commonplace remedy, and in 1811 it was defined as "any medicine adapted more to please than to benefit the patient." It described a kindness or a convenience more than a treatment. Its opposite, nocebo, from the Latin for "I shall harm," was coined by Walter Kennedy in 1961.

Researchers also separate two ideas that are often merged. The placebo response is everything that happens to people who receive a placebo in a trial. That includes natural healing, the normal ups and downs of symptoms, and errors in measurement. The placebo effect is only the part left over when you compare that response with receiving no treatment at all.

Beecher and the Powerful Placebo

The modern story begins with Henry K. Beecher, a Harvard anesthesiologist whose wartime service in North Africa and Italy stirred his interest in placebo-like effects. In 1955 he published "The Powerful Placebo" in the Journal of the American Medical Association. Looking across 15 trials of different conditions, he reported that 35 percent of 1,082 patients were satisfactorily relieved by a placebo alone. For decades the paper was the most frequently cited reference on the subject, and it stressed the need for double blind, placebo controlled trials.

The trouble was that Beecher's figures tracked change in patients given a placebo, not the difference between placebo and nothing. In 1997 Gunver Kienle and Helmut Kiene reanalyzed his sources and found no evidence of a placebo effect in any of the studies he cited. The improvements he counted could be explained by other things: spontaneous recovery, symptoms that rise and fall, regression to the mean (an unusually bad day tends to be followed by a more ordinary one), additional treatment, and patients giving polite answers. That does not mean placebo effects are fictional. It means Beecher's famous number measured something much broader than the power of belief.

Testing Placebo Against Nothing

The cleaner test is to randomize people to a placebo or to no treatment. Asbjørn Hróbjartsson and Peter Gøtzsche of the Nordic Cochrane Centre in Copenhagen gathered every such trial they could find. Their 2001 analysis in the New England Journal of Medicine, titled "Is the Placebo Powerless?", covered 130 trials. Placebo had no significant effect on objective outcomes or on yes or no outcomes. It showed possible small benefits on continuous subjective measures, especially pain, where the average effect equaled a drop of about 6.5 millimeters on a 100 millimeter pain scale.

Their Cochrane review, updated in 2010, widened the net to 234 trials across 60 clinical conditions. The conclusion was blunt: "We did not find that placebo interventions have important clinical effects in general." Placebos did influence outcomes that patients report themselves, especially pain and nausea, though the authors cautioned that these were hard to separate from biased reporting, since patients in such trials know whether they got something or nothing. For smoking, dementia, depression, obesity, high blood pressure, insomnia and anxiety they found no statistically significant effect, though some estimates were imprecise. Other researchers have criticized parts of the method, and the debate continues. Mainstream summaries still say much the same thing: placebos can affect outcomes such as pain and nausea but otherwise do not generally have important clinical effects.

Feeling Better Versus Getting Better

A 2011 asthma study in the New England Journal of Medicine, with Ted Kaptchuk among its authors, shows the gap clearly. Researchers gave 46 patients, in random order over repeated visits, a real albuterol inhaler, a placebo inhaler, sham acupuncture or no treatment. Among the 39 who finished, the real drug improved a standard measure of lung function by about 20 percent, compared with about 7 percent for each of the other three. Yet patients rated their improvement almost the same for the real drug (50 percent), the placebo inhaler (45 percent) and sham acupuncture (46 percent), all well above no treatment (21 percent). They felt better on placebo. Their airways did not open.

That is the core distinction. There is no good evidence that a placebo shrinks a tumor, clears an infection or changes the course of the disease itself. What placebos change is mostly the experience of illness: pain, nausea, discomfort and how people rate their own condition. That matters, because suffering is real. It is not the same as cure.

The bodily side of this is not imaginary either. Since 1978 researchers have known that placebo pain relief depends on the release of the brain's own opioids. Conditioning plays a role too: in experiments where a neutral taste was repeatedly paired with an immune suppressing drug, the taste alone came to produce some immune suppression. These findings belong to the study of consciousness and the brain, and they show expectation working through ordinary biology rather than around it.

Open-Label Placebos

For decades it was assumed that placebos work only if patients are deceived. Ted Kaptchuk, who became director of the Harvard Program in Placebo Studies and the Therapeutic Encounter in 2011, tested that assumption. In a 2010 trial in irritable bowel syndrome, 80 patients were randomized either to no treatment or to pills openly described as placebos. After three weeks the placebo group reported significantly greater global improvement and lower symptom severity. A 2021 meta-analysis of 11 open-label studies found a significant, if somewhat smaller, overall effect, while noting that this research is "still in its infancy."

Two cautions apply. The trials are small, and the outcomes are again self-reported symptoms. And open label did not mean neutral. In the 2010 study, patients were told the pills had been shown in clinical studies to produce significant improvement in IBS symptoms through mind-body self-healing processes. Expectation was still being built, just honestly.

The Nocebo Side

Expectation can harm as well as help. In the nocebo effect, people who expect side effects can experience them even from an inert pill. One review found that nearly 1 in 20 patients taking placebo in depression trials dropped out because of adverse events thought to come from this effect. In the UK, researchers concluded that a large share of reported statin side effects were nocebo responses rather than effects of the drug, and publicity about those side effects led many patients to stop taking statins. Nocebo deserves equal attention, because it shows that the same mechanism that eases symptoms can also create them.

What This Means for New Thought

The New Thought movement taught that sickness originates in the mind and that right thinking heals. William James, who called it the mind-cure movement in The Varieties of Religious Experience, took the healings its followers reported seriously as psychological facts. Placebo research partly supports a modest version of that view. Expectation, trust, attention and meaning do change how people feel, and those changes can be measured.

The research does not support the strong version. Nothing here shows that thought alone cures cancer, infection or other disease, and the largest reviews found no important clinical effect of placebos in general. Claims built on the Principle of Mentalism or the law of attraction, that the mind directly creates bodily conditions, go far beyond the evidence. There is also a moral risk. If illness is blamed on wrong thinking, people can end up blaming themselves, or delaying treatment that works.

The honest summary is this. Belief changes experience, sometimes a great deal, and experience matters. Belief has not been shown to change the course of disease. Holding both facts at once is harder than choosing one, and it is much closer to the truth.

Sources and further reading

• Placebo (Wikipedia)
• Nocebo (Wikipedia)
• Henry K. Beecher (Wikipedia)