TED KAPTCHUK BEGAN HIS CAREER AS A PRACTITIONER OF ACUPUNCTURE AND CHINESE MEDICINE. In the late 1970s, he treated a woman for bronchitis with expectorants and cough suppressants. She later became convinced that this treatment had cured her ovarian disease, and her chronic pain had vanished. Looking back, Kaptchuk thinks it was one of his first encounters with the placebo effect.

When he was studying epidemiology at Harvard in the 1990s, Kaptchuk remembers, he had to carefully design his experiments to account for the placebo effect. But the definition of that effect was never clear. “They taught that it was the effect of something that had no effect,” he says. “I realized that there was a lot of room for improvement there.” Since then, Kaptchuk has helped publish some 70 scientific papers on the topic.

Today Kaptchuk is a professor of medicine at Harvard Medical School. He also directs Harvard’s Program in Placebo Studies and the Therapeutic Encounter (PiPS), which he helped create in 2011. Headquartered at Beth Israel Deaconess Medical Center in Boston, it is the world’s only multidisciplinary institute dedicated to the study of placebos.

Q: Where did placebo research stand 10 years ago?
A: I think the field was in a real crisis. In 2001, The New England Journal of Medicine published a paper by a Danish group, a brilliant meta-analysis that said that the so-called placebo effect is really just regression to the mean and spontaneous remission—just the body’s own healing mechanisms. It’s not a psychobiological process. It raised questions about whether the field was real.

But over the past decade or so, researchers have risen to that challenge. For instance, our team has looked at the placebo effect in chronic pain, irritable bowel syndrome, acute migraine attacks and asthma. And we were able to show that, indeed, the placebo effect is stronger than no treatment. 

Q: Have we learned more about how the placebo effect works?
A: In 2005, we were just beginning to look at the neurobiology of the placebo effect. We already knew that endorphins were involved in placebo responses to pain. But many other neurotransmitters, including endocannabinoids and dopamine, have since been implicated in placebo responses. Specific and quantifiable regions of the brain are involved.

Research on the “nocebo” (Latin for “I shall harm,” as opposed to placebo, which means “I shall please”) also began about 10 years ago. That is where patients experience negative side effects when they take fake drugs. A nice set of experiments has shown that the more you describe the adverse effects, especially nonspecific garden-variety side effects—stomachache, headache, fatigue—the more clearly and dramatically the person experiences them. It has begun a discussion of how we should alert patients to possible side effects. 

The most recent advance in terms of the basic mechanism behind the placebo effect is the question of genetics.

Q: What role do genetics play?
A: We recently published a paper showing that there is sufficient evidence to begin to describe something that Kathryn Hall, a member of our team, called the placebome, the genetic likelihood that a person will respond to a placebo.

There are still many questions. How many genes are involved? What in the environment turns on those genes? But that information may be valuable. If patients are genetically more likely to respond to a placebo, they probably need a smaller dosage of a drug.

Q: Are there conditions for which the placebo response just doesn’t work?
A: There are many conditions for which there is no evidence that placebos work, like shrinking a tumor or lowering cholesterol. Right now the evidence is that the placebo is mainly about self-observation, not about changing objective pathology. Placebos don’t work on everything, and knowing the limits is going to be important. I tend to be skeptical about grandiose descriptions of placebos.

Q: Does the placebo effect always depend on deception?
A: Several experiments in the past 10 years demonstrate that it seems possible that if you give people a placebo and tell them what it is, they still get a placebo response. That’s kind of earthshaking.

But the next question is, how do we ethically harness placebos in the clinic? That’s research that needs to be done in the next 10 years. What kinds of behaviors should nurses, physicians and other health care providers use with patients to enhance the placebo effect?