Why a placebo can work—even when you know it's fake

A placebo can trigger pain relief and other benefits even when patients are told the pills they are taking lack therapeutically active ingredients.

By Meryl Davids Landau
Published 25 Jun 2022, 06:03 BST
A psychiatrist holds two pills—one containing medication, the other a placebo.
A psychiatrist holds two pills—one containing medication, the other a placebo.
Photograph by Spencer Weiner, Los Angeles Times via Getty Images

When Betty Durkin stepped onto her deck last June, she slipped on a loose board and fell on the floor. Durkin broke her neck, seriously bruised her wrists and knees, injured the top of her cervical spine, and got splinters lodged in her face. The pain was instantly unbearable.

After several days in a hospital, Durkin, a 73-year-old security clearance administrator in Marion, Massachusetts, was transferred to Spaulding Rehabilitation Hospital in Boston. Her pain was still intense, so Durkin was prescribed a round-the-clock regimen of opioids, something that worried her since a close friend had become addicted to these painkillers following two hospital stays. “I saw what opioids could do to a person. I never wanted to get close to that point,” Durkin says.

That’s why she was thrilled to learn of an unusual clinical trial taking place during her stay; physicians told her they would test whether her pain improved after receiving a placebo pill filled with soybean oil instead of a medicinal ingredient.

In the past dozen years, scientists have published multiple studies testing the concept of an honest or open-label placebo, in which subjects are told in advance that the pill or capsule they are taking does not contain therapeutically active ingredients. In Durkin’s case, she was not only told, but the capsule’s bottle also was clearly marked “open-label placebo.” Based on everything scientists once believed about the importance of concealment for placebos to be effective, these honest placebo pills should not reduce pain, fatigue, migraines, or other symptoms.

But in a significant number of cases, they do.

For three days, as part of the trial, Durkin was asked to smell a whiff of cardamom spice and swallow the capsule before taking her opioids. The goal was to train the brain to associate the experience of taking the placebo with the pain relief from the opioids. After the third day, she was given the scent and the capsule—but no opioids. She was told she could request painkillers whenever she needed them, but she never did.

“I did not expect it to work. I knew it was a fake pill, not something active,” Durkin says. “But somehow my brain didn’t know the difference.”

Most of the trials done so far on open-label placebos have been small, but the results are starting to add up. A systematic review published last year in Scientific Reports evaluated 13 studies with nearly 800 participants and concluded that open-label placebos exhibit significant positive effects. The reviewers cautioned, though, that in the early stages of research in any field, positive studies are more likely to be published than those not supporting the technique. Still, the unexpected effect has many medical experts intrigued.

“It’s a paradoxical intervention,” says Ted Kaptchuk, director of the program in placebo studies and therapeutic encounter at Boston’s Beth Israel Deaconess Medical Centre and a pioneer in this research. On the surface, it doesn’t make sense, he says, but that may be because scientists don’t fully understand the way placebos work.

Placebo is more than just a pill

Inactive treatments have been utilised by doctors and other healers for centuries. As far back as the 1700s British physician William Cullen wrote that he gave a patient a treatment he was skeptical about: “I own that I did not trust much to it but I gave it because it is necessary to give a medicine and as what I call a placebo.”

The use of placebos in clinical trials really took off in the 1960s, after Congress passed an amendment that authorised the U.S. Food and Drug Administration to require pharmaceutical companies to prove that new drugs were not just safe but also effective. Clinical trials comparing a medicine to a harmless placebo became the accepted way to do that, scientists noted in the New England Journal of Medicine on the amendments’ 50th anniversary.

In traditional clinical trials, participants are never informed whether they are receiving the drug or the placebo. Scientists evaluating the trial data are not told either, so the results are supposed to be more directly comparable and less likely to introduce bias.

Over the years, study subjects who unknowingly got a placebo nevertheless improved in many cases, leading doctors to describe what’s been called the placebo effect. According to a review in the journal Neuroimmunomodulation, if a patient merely believed they might have received the active drug, the brain might release chemicals, including endorphins known to reduce pain and improve mood in ways that initiate healing. Some critics doubt the effect is real, attributing any improvements in the placebo group to fluctuating symptoms, the erratic nature of diseases, or even the desire of study subjects to be accommodating.

Either way, placebos have long been considered a necessary component of clinical trials. But keeping patients in the dark bothered Kaptchuk when he was conducting standard clinical trials early in his research career.

“There’s an element of duplicity in the placebo,” he says. In 2010 he decided to test the concept of an open-label placebo for the first time. “All my colleagues said this was nonsense. But it was a deliberate effort to move placebos out of the shadow of trickery.”

Kaptchuk and his colleagues included 80 people with irritable bowel syndrome in a randomised clinical trial; half took two placebo capsules twice each day and the others had no treatment. Researchers were careful to explain to the placebo group that the capsules contained no medication. They also told them that clinical trials have shown placebos may induce self-healing processes.

After three weeks, researchers assessed their symptom severity. Kaptchuk’s team published a report showing the placebo group fared significantly better, a finding that opened the door to the research that has followed.

Informing patients about the possible benefits of taking a placebo is crucial in open-label clinical research, says Leon Morales-Quezada, a Spaulding physician who was the principal investigator of the study Durkin participated in. “We told our patients from the beginning: We’re going to give you a placebo, but there’s a possibility it will help you to control pain and help you to decrease the opioid consumption,” he says.

Initially, people were surprised—and often skeptical, Morales-Quezada says. “They couldn’t believe what we were proposing. But at the same time, they were curious.”

Our powerful brain

Imaging studies published in 2018 showed that traditional placebos often activate neurotransmitters involved in pain and healing. That’s undoubtedly part of what’s happening with the open-label versions, says John Krystal, chair of psychiatry at Yale School of Medicine who is not involved in this research.

The fact that it’s prescribed by a physician is also key, experts say. “A placebo isn’t about the pill. It’s the ritual of the pill,” Kaptchuk says.

But clearly labeled placebos may work somewhat differently than their more traditional cousins. Experts are gaining a better understanding that, especially for pain patients, the brain may exacerbate pain and amplify bodily sensations it should otherwise ignore. In some people, the message of being instructed to swallow a pill with no physiological effects might somehow interrupt the brain’s pain signal more than if they are told the placebo could be medication, Kaptchuk wrote in the British Medical Journal in 2018.

“We never say it’s certain to work,” Kaptchuk says, noting that the uncertainty seems to play an important role in reducing the brain’s pain amplifications.

Additionally, Kaptchuk and Anthony Lembo, director of the GI motility and functional bowel disorders program at Beth Israel, point to further prospects based on their newest IBS study. As they published in Psychosomatic Medicine in April, they enrolled 262 IBS patients—up from 80 in the 2010 study. This time they also added a third group: people who were given a traditional placebo. This group was told they were either getting a placebo or peppermint oil, which some studies have shown can aid IBS. After six weeks, the open-label placebo group and the standard placebo group had similarly improved symptoms, while the no-pill controls remained the same.

When the researchers closely analysed study participants, they found differences in some responses between the two placebo groups. For example, people given open-label placebos who are prone to thinking their pain might never improve—what researchers call pain catastrophising—as well as people who most expected good results were less likely to have their symptoms resolve than those without any treatment. Neither was a factor for the standard placebo group.

“This is not about believing you’re going to get better,” Kaptchuk surmises. “In my opinion, this is about the body knowing something that’s not conscious.”

A wide range of conditions

Another way open-label placebos are being studied is by pairing them with an active treatment, as was done in Morales-Quezada’s opioid research. The idea is to condition the brain to connect the placebo to a therapeutic response. It’s supposed to work much the same way as Pavlov’s famed canine experiment, in which dogs that were fed when a bell was rung later salivated upon just hearing the bell.

In Morales-Quezada’s preliminary research, 20 in-patients with serious injuries were randomised to get honest placebos or their regular opioid treatments. After six days, opioid use remained relatively constant for those in the regular treatment arm, but it dropped by 66 percent for people taking placebos, they documented in Pain Reports. A follow-up study with similar results, which was the one Durkin participated in, is expected to be published soon.

Similarly, in a study published last year in the journal Pain, 51 patients who had spinal surgery were randomised to get honest placebos paired with their pain-relieving analgesics or the regular treatment of just analgesics, with both groups having access to opioids as needed. Over the two-week study period, the open-label placebo group used 30 percent fewer opioids but reported no higher levels of pain.

Studies that don’t involve conditioning have primarily focused on diseases that lack effective drugs. For instance, German researchers found that chronic back pain patients reduced their pain, functional disability, and depression following three weeks of taking a clearly labeled placebo. Cancer survivors with ongoing fatigue have also been helped by open-label placebos, according to a study published in Supportive Care in Cancer. And migraines improved more in people taking open-label placebos than those getting no treatment.

Prescribing placebos?

Physicians involved in this research haven’t yet incorporated the treatment into their clinical practice. After all, as GI physician Lembo notes, it’s still experimental, and there’s no easy way for a doctor to prescribe a bottle labelled “placebo” for patients to pick up at their local pharmacy. But Morales-Quezada looks forward to the day when many doctors do adopt them, especially for pain.

In truth, many doctors already quietly incorporate placebos into their practice. A National Institutes of Health survey of nearly 700 internists and rheumatologists published in the British Medical Journal in 2008 found that more than half said they regularly prescribe vitamins, over-the-counter pain relievers, or other treatments only for their placebo effect, which they rarely admitted to patients.

Ronald Williams, a 37-year-old tech entrepreneur in Los Angeles, is one patient who recently got such a prescription. After experiencing back pain last November, he went to see an orthopaedic physician who examined him and recommended an ergonomic office chair and certain neck exercises. But Williams kept pressing the doctor for medication, so he finally relented and wrote a prescription. When Williams returned a week later after adopting all three, he informed the doctor that his pain had vanished thanks to the “magic medicine.” That’s when the doctor admitted the pill was a placebo.

Williams was amused to learn of the ruse, but others might not be, says Anne Barnhill, a bioethicist at Johns Hopkins University in Maryland. Physicians have an ethical obligation not to deceive patients, even if they think concealing a placebo will be helpful to them, she says. What’s more, hidden placebos may aid a symptom in the moment, but if the patient later researches their condition and learns that what they were given is not medically effective, this could harm the doctor-patient relationship. Yale’s Krystal especially worries about this happening in minority communities, where trust in the healthcare system is already low.

Being honest about a placebo eliminates these issues, although doctors would need to inform patients very carefully, Barnhill cautions, because not everyone understands the term, and some may believe there’s active medicine involved even if they’re told otherwise.

Patients themselves seem open to the idea. In a 2016 survey of 850 people by the National Institutes of Health and Kaiser Permanente, when people were asked whether open-label placebos would be acceptable in certain scenarios, such as when the condition is not dangerous and good treatments don’t exist, some 85 percent answered yes.

“The mind plays such an important role in patient recovery,” Krystal says. Like all placebos, open-label ones take advantage of the bond between doctors and patients that is “one of the most special and unique connections among people in our society.”


Explore Nat Geo

  • Animals
  • Environment
  • History & Culture
  • Science
  • Travel
  • Photography
  • Space
  • Adventure
  • Video

About us


  • Magazines
  • Disney+

Follow us

Copyright © 1996-2015 National Geographic Society. Copyright © 2015-2023 National Geographic Partners, LLC. All rights reserved