When a placebo might be the best 'drug'

Inactive pills and injections are controversial but widely used

Published: April 2012

Would your doctor offer you a pill, knowing it had no proven medicinal value for your ailment? A study in 2010 found that 40 percent of physicians had used an antibiotic as a placebo, and 11 percent had prescribed inactive pills or injections. But doctors who use placebos rarely describe them as such to their patients, according to the research. A 2008 study had similar results: about half of U.S. physicians reported using placebos, such as vitamins, over-the-counter pain relievers, and sedatives, solely to boost patient expectations of getting better.

The use of placebos in everyday health care is ethically controversial. Some doctors prescribe them to quiet a demanding patient. But the drugs might not be benign. In fact, the most common placebo used in the 2010 study—antibiotics for nonbacterial diagnoses—can cause serious harm. Yet clinical trials have found  that even nondrug, dummy treatments like sham acupuncture for back pain and hot flashes can be just as effective as actual treatments in relieving symptoms.

So researchers are examining whether placebos can be used in honest ways—without deception—to benefit patients. "The advantages are that people wouldn't have the side effects and expense of medication, and it would make them feel empowered," says Ted J. Kaptchuk, an associate professor of medicine and director of the program in placebo studies and therapeutic encounter at Beth Israel Deaconess Medical Center and Harvard Medical School. "But more research has to be done before placebos can be considered a standard of care."

In the meantime, much has been learned about the placebo effect and its potential. Here's a look at the evidence.

What pushes the placebo button?

“It’s unreasonable to think a placebo treatment will shrink a tumor, but we know it can relieve nausea and pain in cancer patients.”—Ted J. Kaptchuk, associate professor, medicine

The following factors appear to work together to promote a placebo response:

The medical encounter. A positive doctor-patient relationship can produce its own placebo effect. "People are wired to feel better when doctors listen to them, explain the cause of their problem and convey that they care," says Howard Brody, M.D., Ph.D., director of the Institute for the Medical Humanities at the University of Texas Medical Branch in Galveston. In a 2008 study, sham acupuncture helped relieve irritable bowel syndrome, but the same procedure done by a practitioner who was warm, attentive, and confident yielded better benefits. In an older study, published in 1964 but still cited by researchers, patients about to undergo surgery were divided into two groups. One received no information about postoperative pain; the other was told what to expect and instructed on relaxation, breathing, and movement exercises following surgery. The patients who received special care required half the dosage of narcotics for postoperative pain and went home two days earlier than those in the other group.

Expectations. The doctor's verbal cues may trigger patient expectations that can alter symptoms for better or worse. For instance, in a study from 1978 that is still relevant to experts, dental patients who were told they would receive a very effective painkiller reported less pain than patients whose expectations about the drug hadn't been raised. The effect was particularly strong when a warm, friendly practitioner delivered the message. A 2008 analysis showed that for patients with osteoarthritis, the degree to which they experienced the placebo effect was strongly influenced by the expected strength of the treatment. For example, the placebo effect for drugs was greater than that for nondrug therapies, and it became even larger when the drug was given by injection. And in a study published in 2011, the benefits of a powerful narcotic painkiller were doubled for those given positive expectations. Those who were given negative expectations didn't experience a benefit at all from the drug.

Negative suggestions can even induce unpleasant symptoms known as the "nocebo effect"—"the evil twin of the placebo effect," Kaptchuk says. In a 2006 trial, people given a placebo pill for chronic arm pain were told they might have the side effects of a medication—including sleepiness and dry mouth—and those treated with sham acupuncture were informed about the side effects of acupuncture. Patients in both placebo groups experienced side effects that mimicked the information provided, with several pill-takers withdrawing due to fatigue or dry mouth.

Conditioning. Effective treatments can lead to positive associations and results. In one study, patients who were switched from an active painkiller to a placebo experienced longer pain relief than if the painkiller was simply stopped. Similarly, after treatment with an antihypertensive drug, patients' blood pressure remained lower on a placebo than if no pills were substituted. "And if you're used to going to the doctor and getting better, then seeing the white coat, stethoscope, and diploma on the wall will turn on that memory track and stimulate a placebo response," Brody says.

Diagnosis. The identification of a patient's problem along with an upbeat prognosis can be enough to prompt a placebo effect. For instance, in a study published in 1987 but still referred to by experts, people with nondescript symptoms such as cough or fatigue who were given a firm diagnosis and told confidently that they would be better in a few days were more likely to feel better within two weeks compared with those who were told, "I cannot be certain what is the matter with you."

Inner pharmacy. Brain scans of people given placebos show brain changes and the release of a chemical that closely mirrors the effects of painkillers. "These inborn painkillers act similar to morphine and have a secondary anti-anxiety effect," Brody says. The placebo response may also lead to a boost in other brain chemicals that help reduce stress or ease symptoms of Parkinson's disease.

Conditions that respond to placebos

Placebos have been shown to be effective for subjective complaints such as pain, chronic fatigue, depression, and gastrointestinal disorders that don't stem from an underlying disease. "It's unreasonable to think a placebo treatment will shrink a tumor, but we know it can relieve nausea and pain in cancer patients," Kaptchuk says. In fact, a 2010 Cochrane review covering 60 conditions found that placebos produced no major health benefits when compared with no treatment, but they modestly influenced complaints reported by patients, particularly pain and nausea.

Despite some promise, placebo treatment in patient care raises concerns. The American Medical Association states that the use of a placebo without a patient's knowledge may undermine trust in the doctor, compromise the patient-physician relationship, and even result in medical harm to the patient.

Moreover, doctors appear willing to prescribe active medications with potentially harmful side effects as placebos. And even sugar pills can trigger nocebo effects or, if used inappropriately, may mask and possibly delay treatment of underlying medical problems. Placebos are especially likely to harm more than help when used for the doctor's convenience, for instance, to placate a patient who is difficult or has a problem that has become frustrating to the physician.

How the placebo effect can help

“People are wired to feel better when doctors listen to them, explain the cause of their problem and convey that they care." —Howard Brody, M.D., Ph.D.

"Physician behaviors that clearly increase the placebo effect are attention, empathy, touch, and a certain amount of confidence," Kaptchuk says. Brody suggests that doctors also capitalize on the power of medical rituals—for instance, by writing a recommendation for exercise on a prescription pad to increase patient expectations and adherence.

In addition, doctors should present treatment information in a positive manner to maximize benefits and minimize side effects (or the nocebo response). For example, women receiving an epidural injection for labor pain who were told, "We are going to give you a local anesthetic that will numb the area and you will be comfortable during the procedure" had significantly less pain than those who were warned, "You are going to feel a big bee sting; this is the worst part of the procedure."

The use of placebo treatments is a grayer area. Some researchers argue that prescribing sugar pills may reinforce habits of overmedication. But Kaptchuk thinks the medical ritual of taking a pill that a doctor says may work is a positive experience for patients. "If an individual wants to try a placebo for depression or pain, there's no reason a physician can't cooperate with that—as long as the substance is inert and given with transparency," he says.

Case in point: In a study published in 2010, Kaptchuk and colleagues randomly assigned people with irritable bowel syndrome to receive either no treatment or placebo pills that were clearly presented as such. The placebo group was also told that the placebo effect was powerful, and that taking the inactive pills faithfully was critical. Improvements among people in the pill group were significant and remarkably similar to those achieved with established IBS treatments. Although further studies are needed, the results suggest that placebos just might work—even when they're offered honestly.

When to consider using a placebo

While more research is needed before physicians prescribe placebos as routine care, it may make sense to try a sugar pill before taking medication—or if medication hasn't helped—provided you do so under the following circumstances:

  • The placebo used for a pain condition, mood disorder, insomnia, fatigue, irritable bowel syndrome, or other symptom-based complaint.
  • The decision is made with your physician with your informed consent. The doctor's positive participation and interest will likely boost the chance of a successful outcome.
  • Your doctor first ensures that use of a placebo won't mask or delay the treatment of a medical disorder.
  • The placebo is an inactive or benign substance.


Editor's Note:

This article and related materials are made possible by a grant from the state Attorney General Consumer and Prescriber Education Grant Program, which is funded by the multi-state settlement of consumer-fraud claims regarding the marketing of the prescription drug Neurontin (gabapentin).

E-mail Newsletters

FREE e-mail Newsletters!
Choose from cars, safety, health, and more!
Already signed-up?
Manage your newsletters here too.

More From Consumer Reports

The Best Matching Washers and Dryers These washer-dryer pairs cleaned up in Consumer Reports' tests.
Best 4K TVs to Buy Right Now The top picks from the hundreds of 4K TVs we've tested.
Best New Car Deals Save money on the cars that Consumer Reports recommends.
How to Pick the Right Size Generator for Your House Add up the items you need to power before making your choice.


Cars Build & Buy Car Buying Service
Save thousands off MSRP with upfront dealer pricing information and a transparent car buying experience.

See your savings


Mobile Get Ratings on the go and compare
while you shop

Learn more