Everyone knows the placebo effect: give someone a fake pill, let them believe it's medicine, and a surprising number get better. The trick, obviously, is the belief. The lie is doing the work. Which is why the strangest result in the whole field is this one: in real clinical trials, doctors have handed patients a bottle, said plainly "these are placebo pills, made of an inert substance, with no medication in them," and watched the patients improve anyway. No deception. No secret. They knew. And it still worked.
01 · The paradoxA sugar pill you're told is a sugar pill
Take a second with how odd this is. The entire folk theory of placebos is that they run on being fooled: you think you’re getting a drug, so your mind conjures the effect. Remove the fooling, and there should be nothing left. And yet the honest version, which researchers call an open-label placebo, keeps producing measurable relief. The patients aren’t confused and they aren’t pretending. They’ve been told, in plain language, that the pill does nothing chemically, and a real fraction of them feel genuinely better regardless. That single fact quietly demolishes the assumption that the placebo effect was ever about belief in a secret cure.
02 · The trialHarvard, 2010, and the honest placebo
The study that put this on the map came out of Harvard and Beth Israel Deaconess Medical Center in 2010, led by Ted Kaptchuk. His team took 80 patients with irritable bowel syndrome, a miserable, genuinely symptomatic condition, and split them in two. One group got nothing. The other got a bottle of pills labelled, unambiguously, as placebo, with a short explanation that inert pills can still trigger real mind-body effects. Three weeks later, the honest-placebo group was reporting far more relief: roughly 59% versus 35% in the no-treatment group. Told outright the pills were fake, patients still got nearly twice the improvement of taking nothing at all. It was the first randomized trial to show it, and it broke a rule everyone thought was iron.
03 · The evidenceIt kept working, in condition after condition
A single surprising trial is a curiosity. What makes open-label placebos worth taking seriously is that the result held up. A 2016 trial added honest placebos to usual care for chronic low back pain and found meaningfully less pain and disability than usual care alone. A 2018 trial gave them to cancer survivors struggling with fatigue and saw fatigue severity drop by around 29%. A 2021 meta-analysis pulling the trials together found medium-to-large average effects, concentrated, and this is the crucial pattern, in subjective outcomes like pain and well-being. The honest caveat travels with every one of these: the studies are often small, the effect varies a lot between them, and it’s a boost, not a miracle. But it is real, and it repeats.
04 · The mechanismThe pill is a cue, not a con
So if it isn’t the lie, what is it? Two things, mostly. The first is expectation: anticipating relief genuinely changes how your brain handles a symptom. The second, and the one that cracks the paradox open, is conditioning. Think of Pavlov’s dogs, salivating at a bell because they’d learned it meant food. Over a lifetime of being treated, your body has learned that the whole ritual of care, the pill in the palm, the clinic, the attentive professional, the act of swallowing something meant to help, reliably precedes feeling better. The ritual becomes a cue, and the body answers the cue. On this view the pill was never a message your conscious mind had to believe. It’s a trigger your nervous system responds to automatically, which is exactly why it can still fire when you know, intellectually, that the pill is blank.
This isn't your imagination being polite. Placebo pain relief releases your brain's own opioids, and in a classic experiment, giving people a drug that blocks opioids cancelled the relief. The sugar pill was triggering real chemistry all along.
05 · The chemistryReal relief, made in your own head
That twist deserves unpacking, because it’s the proof the effect is physical. Back in 1978, researchers gave placebo pain relief to patients and then dosed them with naloxone, the same opioid-blocker used to reverse overdoses. If the placebo relief were “all in the mind” in the dismissive sense, blocking opioids should do nothing. Instead it wiped the relief out, meaning the placebo had been prompting the brain to release its own natural painkillers, the endorphins. Later brain-imaging work confirmed it directly: a convincing placebo lights up the same pain-control pathways as an actual opioid drug. The relief is manufactured in-house, by your own neurochemistry. The pill just pulls the trigger. It’s also why the staging matters absurdly much: a placebo described as expensive outperforms a “cheap” one, two pills beat one, and an elaborate ritual like surgery is one of the most powerful placebos medicine has.
06 · The limitWhat a placebo absolutely cannot do
Here’s the line you must not cross, and where a lot of breathless coverage goes wrong. Placebos work on symptoms the brain modulates, how much pain, nausea, fatigue, breathlessness or anxiety you experience. They do not touch the underlying disease. A placebo can genuinely dial down the fatigue and nausea of cancer treatment; it cannot shrink the tumour. It won’t clear an infection, lower your cholesterol or mend a broken bone. In cancer trials, measurable tumour response to placebo runs at roughly 1 to 2%, essentially noise. The flip side of this power has a name too, the nocebo effect, where negative expectation manufactures real side effects and worse symptoms from an inert pill just as readily. The same machinery that heals a feeling can also harm one.
07 · The payoffSo why do they work when you know?
Because the placebo effect was never a trick played on a gullible mind. It’s a trained response of the body, built from a lifetime of learning that the ritual of being cared for tends to be followed by relief, and a trained response doesn’t need you to be fooled to fire. Knowing the pill is inert doesn’t switch off your conditioning any more than knowing a scary film is fake stops your heart from racing. That’s the quiet, genuinely useful truth underneath the paradox: honesty and effect can coexist. A doctor can hand you a pill, tell you exactly what it is and isn’t, and still hand you real relief, as long as you both remember it’s relief, and not a cure.
Quick questions
Do placebos work if you know it's a placebo?
Often, yes, for the right kind of symptom. These are called open-label placebos, given with full honesty: the patient is told the pill is inert. Across multiple randomized trials, honestly-prescribed placebos have still produced meaningful relief for conditions like irritable bowel syndrome, chronic low back pain and cancer-related fatigue. The effect is smaller and less reliable than people imagine, but it's real and repeatable.
What is an open-label placebo?
A placebo prescribed without any deception: the patient is told directly that the pills contain no active medication, and usually given a brief rationale for why an inert pill might still help through mind-body and conditioning effects. It's the honest cousin of the classic hidden sugar pill, and it exists partly to sidestep the ethical problem of lying to patients.
How well do open-label placebos actually work?
Modestly but measurably, and mostly for subjective symptoms. A 2021 meta-analysis of open-label placebo trials found medium-to-large average effect sizes for outcomes like pain and psychological well-being, though the individual studies varied a lot and many were small. The honest answer is that they help a real fraction of people feel better, not that they work for everyone or for everything.
What conditions do open-label placebos help with?
The evidence is strongest for conditions defined by how you feel: irritable bowel syndrome, chronic low back pain, cancer-related fatigue, migraine, allergic rhinitis and some menopausal or menstrual symptoms. What these share is a large brain-modulated component. There is no good evidence that open-label placebos treat the underlying disease process in conditions like cancer or infection.
Why does a placebo work at all, what's the mechanism?
Two main forces. Expectation: anticipating relief can itself change how the brain processes symptoms. And conditioning: like Pavlov's dogs learning to salivate at a bell, your body has learned to associate the whole ritual of being treated, the pill, the clinic, the attentive clinician, with getting better, and responds to the cue. Researchers increasingly think the ritual and bodily experience of treatment matter as much as conscious belief, which is exactly why the honest version can still work.
Does the placebo effect release real chemicals in the brain?
Yes, this is the strongest evidence that it isn't just imagination. Placebo pain relief involves the brain releasing its own natural opioids (endorphins). In a classic 1978 experiment, giving people naloxone, a drug that blocks opioids, cancelled out placebo pain relief, showing the effect ran on real brain chemistry. Later imaging confirmed placebos activate the brain's own pain-control pathways.
Can a placebo cure a disease or shrink a tumour?
No. This is the crucial limit. Placebos act on symptoms the brain modulates, pain, nausea, fatigue, anxiety, perceived breathlessness, not on the underlying biology of a disease. A placebo can genuinely reduce how much cancer-treatment fatigue or nausea someone feels, but it will not shrink a tumour, clear an infection or lower cholesterol. Objective physical outcomes essentially don't respond.
Does the colour, price or number of placebo pills matter?
Surprisingly, yes, the trappings shape the effect. Studies find that placebos framed as expensive tend to work better than 'cheap' ones, two pills can beat one, and pill colour can nudge the effect (blue tends to read as calming, for instance). It's all part of the same point: the placebo response is driven by context and expectation, not by any ingredient.
What is the nocebo effect?
The placebo's evil twin. If positive expectations can make you feel better, negative ones can make you feel worse. That's the nocebo effect. It's why people warned about a drug's side effects often report those very side effects even on a sugar pill, and why anxious anticipation can amplify pain. In healthy volunteers, these negative expectation effects can be as strong as, or stronger than, placebo ones.
Is it ethical for a doctor to give a placebo?
The traditional secret sugar pill is ethically fraught because it requires deception. Open-label placebos were developed largely to solve exactly that: if the pill can help without the doctor lying, you keep the benefit and lose the deceit. It's an active area of medical-ethics discussion, and honest placebos are seen as far more defensible than hidden ones, though not a substitute for effective treatment where one exists.
Can you buy placebo pills?
Yes. Honest placebo pills are sold commercially: the best-known brand, Zeebo, has been available since around 2015 and openly contains no active ingredients. Some people use them for symptom self-management. It's worth stressing that placebos are for symptoms, not cures, and shouldn't replace real treatment for a real medical condition.
Does sham surgery work like a placebo?
It can, strikingly. In a famous 2002 trial, patients with knee osteoarthritis who received fake ('sham') arthroscopic surgery, skin incisions, the full theatre of an operation, but no actual repair, improved just as much as those who got the real procedure. The more elaborate and invasive the ritual, the stronger the placebo response tends to be, and surgery is about as elaborate a ritual as medicine offers.
Isn't the placebo effect just people getting better on their own?
Part of it is, and honest researchers account for that. When someone improves after a placebo, some of the improvement is natural recovery, some is regression to the mean (people enrol when symptoms peak, then drift back to average), and some is a genuine placebo response. A well-known 2001 analysis argued the effect is often smaller than the hype suggests. The open-label trials are valuable precisely because they compare against no-treatment groups to isolate the real effect.
Why is the effect stronger for some people than others?
It varies a lot, and the reasons aren't fully pinned down. Expectation, the strength of prior conditioning, the warmth and confidence of the clinician, and the elaborateness of the ritual all seem to matter. Attempts to find a single reliable 'placebo-responder' personality have been largely inconclusive, so it's best understood as depending on the situation as much as the person.
Take the quiz on this
A quick 4-question check on what you just read. Get them right to earn XP: no points for just scrolling.
Our sources 13 checked
// every claim on this page was checked before it went up