As a surgeon, Teppo Jarvinen was trained to believe that his long hours in the operating room were key to healing his patients. Jarvinen specializes in orthopedics at Helsinki University Central Hospital, and the bread and butter of his practice had been repairing knees—specifically, torn menisci. Around 700,000 surgeries are performed each year in the United States alone to repair damage to the rubbery discs that cushion both the inner and outer knee. Surgery to repair a torn meniscus is one of the most common surgeries performed around the world, Jarvinen says, which led most surgeons to believe it was effective in relieving knee pain, stiffness, and weakness. "How could so many surgeons be wrong?" he says. But after hearing that a common rotator cuff surgery proved no better than placebo, Jarvinen began to wonder whether meniscus surgeries were as effective as he thought. So with a team of Finnish scientists, he set up an elaborate clinical trial to compare a real meniscus surgery with a sham procedure that consisted of a small knee incision but nothing more—the equivalent of giving someone a placebo pill in a clinical trial. The patient wouldn't know what procedure they received, and the surgical team wouldn't know which they were performing until they opened a sealed envelope in the operating room. Jarvinen expected to be able to tell the difference between the two groups of patients, since something as radical as surgery had to be effective. That's not what happened. Instead, the two groups looked almost identical in outcomes. Patients who received the real surgery improved, but so did those who had the sham procedure, and by roughly the same amount. The patients didn't just feel better; objective measures of flexibility and knee strength also showed that the placebo group had improved. In other words, just believing you had surgery was enough to spur your body to heal itself. "It was truly impressive," he says. His study, published in the prestigious New England Journal of Medicine, suggests that what seemed to be helping his patients was not the mighty scalpel but rather the lowly placebo. We tend to think of placebos as sugar pills, a sneaky little trick that researchers use as a benchmark to evaluate "real" medication in scientific studies. But emerging research is showing that, far from being a vehicle of deception, placebos can trigger a complex set of neurological and physiological changes that tap into the body's own arsenal of healing chemicals. Placebos, we're finding, are "real" medicine. And a growing number of doctors are putting this body-brain hack to good use.
Doctors have known about placebo for centuries, even if they didn't formally refer to it as such. Indeed, much of what medicine had to offer until recently was all placebo—see the array of snake-oil "patent medicines" and cure-alls that were healing staples until the early 1900s, when the FDA came into existence and put a stop to much of it. But it wasn't until World War II that scientists began to develop the concept of the placebo effect as a specific phenomenon. In 1946, manufacturers of a popular headache remedy used placebo-controlled trials to evaluate which compounds in their drugs had a therapeutic effect. Importantly, the researchers also noted that many patients said the dummy pills were effective—one of the first known descriptions of the placebo effect in medical literature. Thirty years would pass until scientists made any headway in understanding why.
By then, scientists knew that pain responded fairly well to placebos, but they had no idea why. In the late 1970s, the discovery of endorphins (the body's natural painkillers) was making waves in medicine. It also suggested a pathway through which placebos could reduce pain. A 1978 study in The Lancet gave a group of patients who had just had their wisdom teeth removed either a placebo or naloxone, which blocks endorphins. Not surprisingly, those who got naloxone reported an increase in pain, whereas those who got placebo reported relief, demonstrating that placebos for pain tapped into the brain's endorphin system. Showing that the placebo effect could induce real biological effects was only the first step in actually untangling what they actually were. It would take another 25 years and the development of more sophisticated neuroimaging techniques for scientists to truly begin to tackle the complex placebo response.
"The placebo effect begins long before someone takes a drug," says Tor Wager, a neuroscientist at the University of Colorado Boulder. The belief that a pill or procedure will help sets a person up to feel better.
Hearing about a treatment's benefits activates a brain circuit that is involved in forming our expectations. It's why nurses say "This won't hurt a bit," before giving a shot—it alters our expectations about how painful the needle stick will be. The brain circuit that governs expectations is in the frontal cortex, far away from the more primitive brain regions that control things like nausea and pain. Wager's work in the early 2000s, however, shows that these two brain circuits are actually well-linked. People who respond to placebo aren't simply being tricked—their belief that the treatment will work actually leads to biological changes in the body. "Positive expectations set off a neural chain reaction that we can watch in a brain scanner," Wager says.
Wager's work established placebo as an effective, side-effect-free treatment for many of humankind's most vexing day-to-day ailments, like nausea, pain and even depression (for the last, see the oft-cited study showing that placebo was more effective than either Zoloft or St. John's Wort, among others). Using them clinically, however, had a major barrier: Giving someone a placebo always involved some level of deception. In a placebo-controlled trial, participants aren't told whether they're receiving a sugar pill or the real thing. This ruse is ethically passable for a research trial, but not in an ongoing doctor-patient relationship.
For the placebo effect to be harnessed clinically, scientists would need to know whether it was contingent on this trickery. A 2010 study in PLOS ONE answered that question, by giving some patients with irritable bowel syndrome (IBS) a placebo and actually telling them what they were getting. Others got no treatment at all but still saw the doctor at regular appointments. A portion of the placebo group—who in this case knew they were getting the placebo—showed substantial improvement in their symptoms compared to those who saw the doctor but got no treatment. As it turned out, deceit didn't need to be part of helping patients benefit from placebos. This study opened the floodgates: Doctors could now consider using placebos in an above-board fashion in their everyday practices. Anecdotally, many doctors now cop to prescribing placebos—not in the form of sugar pills necessarily, but rather vitamins (many of which are unlikely to have a physical effect), or sub-clinical doses of antidepressants. And a 2013 PLOS ONE study revealed that 97 percent of general practitioners in the UK had prescribed a placebo, and three-quarters had done so in the past week.
Kathryn Hall, a geneticist at Harvard's PiPS, found out just how powerful belief itself could be when, in desperation, she sought out acupuncture for an exceedingly painful wrist. She was profoundly skeptical about acupuncture, believing its effects to be little more than a placebo. "I didn't expect it to work, but a friend convinced me," Hall said. To her surprise, one acupuncture session relieved her pain completely, and it hasn't recurred in the several years since. "It completely took care of the problem," she says.
The experience triggered Hall's curiosity about why some people respond to placebo and others don't. Neuroimaging studies had shown that after a person takes a pill they expect will work, their brain releases dopamine automatically, suggesting that the dopamine system was central to the placebo effect. So Hall focused on a gene called COMT, which helps to recycle dopamine, breaking it down and removing it from the brain. In a 2012 study on IBS treatments, Hall showed that people carrying a COMT gene that produces a less-active protein have higher levels of dopamine in their brains and a greater placebo response.
Although this study revealed that a person's response to placebo isn't random, it also doesn't entirely explain why one person may benefit from a placebo in one situation but not another, or why some people benefit from placebo pills but not injections or sham surgeries. Understanding that will require a deeper understanding of what Hall calls the "placebome," or the array of genetic variation that affects our response to placebo. Being able to predict who is more or less likely to respond to placebo is of huge interest to clinical trial designers, doctors, and of course pharmaceutical companies.
Pharma companies, in fact, have been capitalizing on what is known about the effect for decades. Everything from the color of a pill (blues make good sedatives, reds are better as stimulants, and bright yellow makes a good antidepressant), any flavorings, and even the method of administration is carefully thought out to maximize patient response, says Jensen, who worked in the pharmaceutical industry before she began her career in academia. Russians, she found in her research, generally respond best to a licorice flavor in their medication. Americans prefer pills to injections, whereas in other countries, it's the opposite. The French show best results with rectally administered treatments. Pharmaceutical companies take advantage of this information regularly, without changing a thing about their medications' active ingredients. "Medication is an active drug surrounded by so much else," Jensen says. "Only a small part of treatment response can really be explained by the molecule itself."
To the consternation of pharma execs, the placebo effect seems to be mysteriously on the rise in the US. Pain specialist Jeffrey Mogil and his colleagues at McGill University in Montreal reviewed 84 trials of pain medications from 1990 to 2003. Their results, published in December 2015, reveal that painkillers in clinical trials worked 27 percent better than placebo in 1996. By 2003, however, they only worked about 9 percent better. The study didn't assess why, but the authors believe the difference could be partly due to the frequency of pharmaceutical advertising in the US, and the general improvement in drugs over time. We expect that popping a pill will make us feel better, and it's as much this belief as the drug itself that provides the benefit. This poses a problem to pharmaceutical companies. If the placebo effect is getting stronger, it will be harder for a new drug to show a beneficial effect in clinical trials, since the effects of new drugs are compared to the effects of a placebo. A stronger placebo effect means that drugs will need to demonstrate more of an effect before they can receive FDA approval.
Placebos aren't magic—they can't cure cancer or clear severe infections. But they might be able to help many of us feel just a little bit better. The key to picking the lock on your own inner medicine cabinet is the doctor-patient relationship, Jensen says. Whether you take a sugar pill or something with an active ingredient, the warmth and confidence of the person prescribing it makes a huge difference. She recommends that people identify treatments they believe will help them, since this belief will likely bolster the benefits they receive.
When a Harvard research team gave a group of patients with IBS sham acupuncture (using retractable needles that never actually punctured the skin), people improved more when they were given warm, supportive treatment, including statements like "We're so glad to see you," and "I have had much positive experience treating IBS and look forward to demonstrating that acupuncture is a valuable treatment in this trial."
What a provider says, and how they say it, matters. It isn't just an add-on benefit of not being stuck with Dr. Asshole. It's a key part of the treatment itself.