clock menu more-arrow no yes mobile

Filed under:

The weird science of the placebo effect keeps getting more interesting

Take this sugar pill. Twice a day. And feel better? 

Brian Resnick is Vox’s science and health editor, and is the co-creator of Unexplainable, Vox's podcast about unanswered questions in science. Previously, Brian was a reporter at Vox and at National Journal.

The story of the placebo effect used to be simple: When people don’t know they are taking sugar pills or think they might be a real treatment, the pills can work. It’s a foundational idea in medicine and in clinical drug trials dating back to the 1950s.

Then Ted Kaptchuk came along.

Kaptchuk is a professor at Harvard Medical School, and over the past decade, he and colleagues have shown, in study after study, that giving people placebos openly — that is, telling them they are taking a placebo — helps them feel better. Specifically, they found a placebo can relieve not just pain but also anxiety and fatigue.

In February, Kaptchuk and his colleagues published the results of a clinical trial comparing these open-label placebos to double-blind placebos (the gold standard in medical research) in treating irritable bowel syndrome. Both were equally effective.

“The myth [of placebos requiring deception] has been destroyed,” Kaptchuk says. “And once you destroy the myth, you open up all new kinds of questions.”

Questions like: Can placebos become part of standard medical practice? Would patients take them? Can they be used, strategically, to reduce the consumption of addictive opioid painkillers?

The answers to these questions could change how we typically think of medicine. There’s just one problem: Researchers are not entirely sure how inert sugar pills, with no active drugs in them, work at all.

The origins of open-label placebos

When Kaptchuk first had the idea to give people sugar pills and tell them they were placebos, his team said, “Ted, this is the stupidest idea you’ve come up with yet,” he recalls. It was just taken for granted that placebos needed to be secret to work.

But his idea to be honest with patients wasn’t a lark. It came from a place of empathy and from frustration with the research he’d been doing.

For years, he tried to find ways to make the placebo effect bigger and stronger, particularly for patients with irritable bowel syndrome, a painful chronic illness that’s hard to treat. But “I was, in my heart, miserable because everything I was studying had to do with concealment or deception,” Kaptchuk says.

He knew placebos could never become part of mainstream medical practice as long as they were used surreptitiously. In the context of a clinical trial, patients can consent to the possibility of being deceived. In the real world, doctors can’t offer that option. That kept whatever healing power placebos could have outside the reach of everyday medicine.

Kaptchuk was unsettled. But then he and his team conducted an anthropological survey of patients enrolled in clinical trials for irritable bowel syndrome that involved deceptive placebos. They were simply curious: How did the patients feel about the concealment?

It turns out, as Kaptchuk and his colleagues reported in a study, many were anxious about the thought of possibly being on a placebo. They said, ‘What does that mean about my condition if I got better on placebo? ... Did I make it up in my head?’”

Research keeps confirming: Open-label placebos can help people

For Kaptchuk, that was a breaking point. “I read this and I said, ‘Hold it. Why don’t I just tell them they’re getting a placebo and address their concerns?’” He realized he could tell them, “If you get better from placebo, that’s a sign of healing. It’s not a big deal. It’s not like something’s wrong with your mind.”

He then set up a small study to find out. He took 80 patients with irritable bowel syndrome. Half were given a placebo and told it was a placebo, while the other half got nothing. At the end of the trial, the people in the placebo arm of the trial said they felt better than those who got nothing.

Since then, open-label placebos have been shown to relieve the symptoms of other conditions, treating chronic pain, hot flashes, fatigue, allergies, arthritis, anxiety, and depression. There have been long-term follow-up studies, where researchers for five years followed patients who took open-label placebos and showed sustained improvement.

This work has led Kaptchuk to write a new definition of the placebo effect. “It’s the positive health benefits people receive in the context of a clinical interaction that’s due to the rituals, symbols, and behaviors that surround the pill,” he says. “When a person’s sick and goes to a healer, the drama of medicine by itself is a potent form of healing.”

By drama, he’s being literal. Medicine is a type of theater. A pill is a prop in the story of medicine. A doctor wearing a lab coat and being, like, really attentive to you is a character in the story of medicine (providers with warmer, friendlier personalities tend to produce stronger placebo effects).

How powerful is this drama? How much relief can it bring?

In the February paper, published in the journal Pain, Kaptchuk and his collaborators did an expanded replication of their original trial with IBS patients. The difference this time was they compared three conditions: open-label placebos, double-blind placebos, and a no-treatment control group. From a methodological point of view, this study tried to do a funny thing: Compare three forms of nothing.

The results showed that, one, open-label placebos work. Seventy percent of those who took them showed at least a 50-point reduction in symptom severity (graded on a scale of 500) compared to 54 percent of those in the no-treatment control arm who experienced a 50-point drop. Around 30 percent of those on open-label placebos reported an even larger 150-point reduction, compared to 12 percent just in the no-treatment group.

“There’s merit in this line of research, no question about it,” says Beth Darnall, the director of Stanford Pain Relief Innovations Lab, who was not involved in this study. “It’s almost like a free treatment — and there’s virtually no risk involved. That could be really important.”

Of course, it’s not that everyone on open-label placebos is getting better. But if you were one of the people who experienced the 50- or 150-point drop in symptoms, she says, “that’s a big deal, right?”

The results also showed that the open-label placebos and double-blind placebos produce a comparable response. That is, whatever benefit people get from the double-blind placebo, they can also get it from open-label.

Again, what that means is “you don’t need to trick the patient,” Daniel Keszthelyi, a PhD/MD at Maastricht University Medical Center in the Netherlands who studies and treats irritable bowel syndrome and wasn’t involved in the study, says. “This is something that you can actually exploit in clinical practice.”

But scientists still have so many questions to answer.

Primarily, Kaptchuk doesn’t know how this all works. “I still don’t know what the magic is that you have to put in the soup,” Kaptchuk says.

Two psychological theories for why open-label placebos work

Researchers are exploring two ideas for why open-label placebos can be effective for certain conditions: expectations and conditioning.

“Expectations is just you believing that something is going to work,” says Darwin Guevarra, a researcher who studies the impact of placebos on emotion regulation at Michigan State University. In his studies — and in many of Kaptchuk’s — expectations are set by education. Study participants are taught about the placebo effect and are told open-label placebos could help.

But it’s a little more complicated than just believing in something. When you have an expectation for improvement, you might start to pay attention to different signals coming from your body. We have a flurry of neurological activity happening all the time. Some of the signals our body sends to our brain can be interpreted as painful, others as non-painful. So when you change your expectations, you might start ignoring the signals that say “I’m in pain” or “I’m anxious” and start looking for clues that you’re feeling better.

Neuroscientists have long appreciated the fact that our perception of our bodies (or of anything really) is just our brains making a best-guess interpretation of imperfect information from our sensory organs. So the thinking here is simple: You start paying attention to different sensory signals, and your sense of reality — i.e., how you feel — changes as well.

But expectations don’t really explain all of what’s going on with open-label placebos. In his experience, Kaptchuk says that a lot of people who sign up for clinical trials don’t really expect to get better. Signing up for a clinical trial is often a last-ditch option for relief from pain. “They’ve been through the mill already of despair and pain,” he says.

That’s where conditioning comes in. Conditioning is an automatic learned response and doesn’t require belief. The classic example is Pavlov’s dogs: The dogs learned to associate a ringing of a bell with being fed. Eventually, all it took was a bell for them to start salivating. For humans, it turns out we can associate one thing (taking a placebo pill) with a positive outcome (feeling better). “Just the act of taking the treatment leads to you feeling better even if you remove the active ingredients that are in the actual treatment,” Guevarra explains.

A recent study, also published in the journal Pain, showed how conditioning with open-label placebos could possibly work in the real world. In the study, 51 spine surgery patients were assigned to one of two conditions: undergo conditioning with open-label placebos (along with taking opioid drugs) or treatment as usual.

In the conditioning procedure, every time a patient took their usual opioid medication post-surgery, they were also instructed to take an open-label placebo pill. Then, after doing this for a day or two, the patients were then told to take the placebo pills on schedule.

The idea is that the brain learns to associate taking the placebo pill with the release of real drugs that dull pain. Those drugs release neurotransmitters in the brain eventually. “In theory, even when you just take placebo, if your brain is conditioned, it’s going to start releasing those neurotransmitters,” Kelsey Flowers, one of the lead researchers on the study, says.

All the patients in the study were tracked for 17 days after surgery. Those who experienced this conditioning with placebos ended up taking 30 percent fewer opioid drugs compared to the treatment-as-usual group. They also reported less daily pain.

The funny thing, Flowers says, is many of these people who were conditioned told the researchers they weren’t sure it worked. “That’s the beauty of what we’re trying to figure out,” Flowers says, “is if you will still have the benefits even if you don’t expect them.”

This was just a small pilot study. But it shows how this is feasible and what medicine might look like in the future if more doctors embrace the healing power of conditioning with placebos.

Beyond expectations and conditioning, it’s fair to wonder: Are patients just telling researchers what they want to hear? It’s possible. But there’s also research that finds that open-label placebos do seem to reduce neural markers of pain and stress.

Guevarra recently led a study that showed open-label placebos reduced stress as read out through electrodes attached to the head. ”We see that there’s a decrease, a gradual decrease, in the amplitude of these brain waves after people receive the placebo,” he says. “It suggests that this thing is actually reducing people’s emotional distress.”

Overall, researchers are still figuring out what plays more of a role in delivering the placebo effect. Or which one is most effective. Or how they overlap. The answer matters because it will change the way these scientists design their interventions. Conditioning can be a burdensome process. Expectations could be set by a one-time educational intervention, which might be more appealing.

Placebos seem to work best only on subjective symptoms

It’s very important to note: Placebos aren’t magic pills. “It sometimes works on some people, some of the time,” Kaptchuk says. “It’s not like a cure-all.”

Placebos (open label or concealed) appear to mostly work on subjective symptoms, such as pain. They don’t work on an objective symptom — something a doctor could see or diagnose, such as a fracture on a bone. “Placebos don’t shrink tumors, they don’t change your diabetes, and they’re not going to actually lower your blood pressure for more than 15 minutes,” Kaptchuk says.

Basically, placebos appear to work on things that pass through the brain’s perceptual systems — where they can prompt the release of opioids and other endorphins (chemicals that reduce pain) in the brain.

So does that mean their usefulness is quite limited?

Kaptchuk argues there are subjective symptoms for all objective illnesses. Cancer is caused by tumors but leaves people feeling fatigued, for instance. Then there are diseases where doctors can’t find an objective thing wrong with a person, like irritable bowel syndrome, a condition that’s estimated to afflict 10 percent of people and leaves many in chronic pain without effective treatment options.

It’s believed IBS is a problem where the brain is misinterpreting what should be normal-feeling sensations from the gut as pain. Perhaps placebos interfere in that broken pathway. “This is about a situation where there aren’t medications around,” Kaptchuk says. “That’s actually a huge part of medicine; it’s the non-glorious part.”

Is medicine ready for open-label placebos?

All the doctors and scientists I spoke to for this story said open-label placebos aren’t ready for primetime yet. There are too many unanswered questions. Keszthelyi, the gastroenterologist who studies IBS, says he could see using them in some cases when drugs aren’t being well tolerated. “I’d definitely give someone an open-label placebo if I had reason to assume there were going to be a lot of adverse effects as a result of the pharmacotherapy [i.e., real drugs] itself.”

To a degree, doctors are already prescribing placebos to patients. They’re just doing it with real drugs (doctors will sometimes prescribe drugs or vitamins knowing they won’t help patients beyond the placebo effect). Open-label placebos would allow them to drop the pretense.

Others are less confident that open-label placebos are the future. Luana Colloca, a physician, neuroscientist, and placebo researcher at the University of Maryland, worries that prescribing open-label placebos could just become a means for doctors to shoo away patients. “I truly believe that we don’t need open-label placebos,” she says.

Colloca says you don’t need placebo pills to activate the placebo effect. (Kaptchuk’s definition of the placebo effect doesn’t necessitate pills. The pill is just a convenient prop to alter a person’s expectations or provoke a conditioned response.)

Colloca’s own studies have shown that placebos don’t even need to be a physical object. Just the right words from a doctor, at the right time, can help people get more therapeutic value out of already-powerful drugs, like morphine. (Indeed, there are many creative approaches at work these days to help patients with pain management that are not placebo-based. In her work, Darnall uses elements of cognitive behavior therapy that also yield reductions in opioid consumption.)

Placebos have been overlooked as medicine. Maybe that should change.

Overall, Kaptchuk’s definition of the placebo effect — the “drama” of medicine — is a radical way of thinking about medicine that not every physician will be comfortable with. But for too long, Kaptchuk argues, mainstream medicine has waved away all this drama, saying, “It’s just the placebo effect.”

That’s a “denigration,” according to Kaptchuk. The placebo effect has been seen as a bar that needs clearing to establish what’s “real” medicine. But it is, increasingly, looking like a medicine in its own right. The placebo effect is this extra healing power that doctors can squeeze out of already-useful drugs or be used when no good drugs are available. It’s money left on the table.

Again, Kaptchuk isn’t totally sure how this all works. “But I know it has to do with not selling bullshit to patients and being honest,” he says. “What we’re selling here is honesty.”


How to catch a scientific fraud


9 ocean mysteries scientists haven’t solved yet


10 of the biggest — and smallest — scientific mysteries

View all stories in Unexplainable