Translated from the Portuguese original. Read the original (Português)
A placebo is a treatment with no specific action on the problem: a sugar pill, a saline injection, a sham procedure. The placebo effect is the improvement that comes from the context of the treatment — the expectation of getting better, the learning left by previous treatments, the ritual of being cared for, the relationship with whoever provides the care. The nocebo effect is the same logic in reverse: expecting to get worse makes you worse. Neither is imagination — both mobilize the body for real. But neither is a cure, either: what they change the most is what the person feels and reports, not the course of the disease.
A placebo is a substance or procedure with no known specific action on the condition being treated. The placebo effect is the favorable clinical change produced not by that substance, but by the psychosocial context of the treatment — expectations, associative learning, the therapeutic ritual, and the relationship with the professional. The nocebo effect is the symmetrical phenomenon: worsening of symptoms or the emergence of adverse events arising from negative expectations. The names come from Latin: placebo, "I shall please"; nocebo, "I shall harm" — the latter proposed as a technical term by W. P. Kennedy in 1961. An international consensus of 29 researchers in the field defines both as effects that occur after the administration of an inert treatment or as part of active treatments, through psychobiological mechanisms such as patient expectations (Evers and colleagues, 2018).
Three words that tend to get confused
Much of the confusion on this topic comes from using "placebo" for three different things. Tap the cards:
The distinction matters in practice. Someone who takes a placebo and gets better did not necessarily get better because of it: the condition could have been resolving on its own, the person sought help at its worst moment (and tended to return to the average anyway), and the very act of repeating measurements changes the results. To know how much effect from the placebo remains, you have to compare the group that received the placebo with a group that received nothing at all.
What the placebo changes — and what it doesn't
An experiment published in the New England Journal of Medicine in 2011 separated these two things in an unforgettable way. People with asthma went, in randomized order, through four situations: an inhaler with albuterol (a real bronchodilator), a placebo inhaler, sham acupuncture, and no intervention at all. At each visit, lung function was measured (FEV₁, how much air the person can blow out in one second), and they were asked how much they felt they had improved.
Read the two panels separately. In measured lung function, only albuterol made a difference: it increased FEV₁ by 20%, against about 7% in each of the other three situations. In self-reported improvement, the picture reverses: it was practically the same with the drug (50%), with the placebo inhaler (45%), and with sham acupuncture (46%) — and all three well above doing nothing (21%).
In other words: the placebo moved what the person felt, and moved it considerably, without opening the airways any more than doing nothing did. In a disease that can kill, relying on self-report alone would be dangerous.
What the body of evidence showsThe Cochrane review Placebo interventions for all clinical conditions (Hróbjartsson and Gøtzsche, 2010) pooled 234 trials that had, at the same time, a placebo group and a no-treatment group — 202 of them with usable outcome data, covering 60 clinical conditions. The average placebo effect was small and quite uneven: larger for outcomes reported by the patient than for outcomes assessed by an observer, with an identifiable effect on pain and nausea — and, for pain, varying a great deal from study to study, even among those at low risk of bias. In conditions with enough trials to analyze, there was no statistically significant effect on depression, anxiety, insomnia, hypertension, obesity, dementia, or smoking cessation. The authors also noted that larger effects showed up associated with "physical" placebos (such as sham acupuncture), with outcomes that depend on patient participation, with small trials, and with studies whose stated aim was precisely to study the placebo effect.
What happens in the body
The placebo effect is not "just imagining it." Part of it has already been measured in the lab, in biological signatures; another part depends on what the person learned before. Explore three lines of research:
The body releases its own painkillers
In 1978, Levine, Gordon, and Fields studied pain in people who had just had their wisdom teeth extracted. Those who responded to the placebo with relief lost that relief when given naloxone, a substance that blocks opioid receptors — and those who did not respond to the placebo did not get worse with it. The finding is consistent with the hypothesis that placebo analgesia depends on the release of opioids produced by the body itself.
In 2001, a study published in Science used positron emission tomography in people with Parkinson's disease and showed substantial release of endogenous dopamine in the striatum in response to the placebo — the same system mobilized by the disease's medication.
Treatment is also learned
Part of the effect does not depend on what the person believes, but on what they have already experienced. The smell of a doctor's office, the color of a pill, the gesture of giving an injection: all of this used to come paired with real relief and comes to function as a signal — it is respondent conditioning applied to care. Add to this what is learned through instruction ("this is strong, it will work fast") and through observing other people getting better.
In the language of behavior analysis, the therapeutic context is a set of stimuli with a history: prior pairings and verbal rules about the treatment alter what the organism does — and what the person reports feeling. That is why the same inert pill works differently in people with different histories.
The nocebo pathway runs through anxiety
Benedetti and colleagues induced worsening of pain using only words, in volunteers, and tracked what the body did. The worsening came together with hyperactivity of the hypothalamic-pituitary-adrenal axis (more ACTH and cortisol). An anxiolytic knocked down both things — the extra pain and the hormonal response — suggesting the central role of anxiety.
A cholecystokinin (CCK) antagonist, on the other hand, abolished the extra pain without affecting the hormonal response, which points to a separate biochemical pathway for the nocebo effect. In short: placebo analgesia recruits the opioid system; nocebo hyperalgesia recruits the CCK system. These are two pathways, not one with the sign flipped.
The nocebo effect: when expectation collects its due
If expecting to improve brings relief, expecting to get worse causes suffering — and that shows up clearly in the side effects reported by those who took a placebo within clinical trials. A systematic review with meta-analysis published in JAMA Network Open in 2022 pooled adverse-event reports from 12 articles on covid-19 vaccine trials, with 45,380 participants — 22,578 of them in the placebo group, which received an inert injection.
- After the first dose, 35.2% of those who received the placebo reported some systemic effect; the most common were headache (19.3%) and fatigue (16.7%).
- After the second dose, 31.8% of those who received the placebo reported systemic effects.
- Comparing the two groups, nocebo responses accounted for 76% of systemic effects after the first dose and 51.8% after the second.
- Those who received the vaccine reported significantly more effects — that difference exists and is real — but the rates in the placebo group were high regardless.
The symptom is real; what's in question is the cause. A nocebo headache hurts just the same. Saying "it's nocebo" is not saying "it's all in your head," nor that medication side effects don't exist — they do, they are frequent, and they need to be taken seriously. It is saying that part of what we feel after a treatment would have appeared even without the active ingredient, because of the scare, the attention turned toward the body, and the expectation created by the warning.
Don't stop anything on your ownUnderstanding the nocebo effect is not a reason to abandon a prescribed treatment, nor to distrust any side effect you feel. If something bothers you, report it to the professional following your case — only whoever prescribed it can assess, adjust, or discontinue it. This page is educational and does not give guidance on medication use.
Placebo without deception: what research is testing
For a long time it was assumed that the placebo effect required secrecy. In 2010, a trial with 80 people with irritable bowel syndrome tested the opposite: one group received pills openly presented as "placebo pills, made of an inert substance, like sugar pills," with the explanation that studies show improvement through mind-body processes; the other group had the same amount of time and attention from the professionals, without a pill. Those who received the placebo knowing it was a placebo reported greater overall improvement, at the midpoint and at the end of the three weeks.
A 2021 systematic review pooled 13 studies of this type — what the literature calls open-label placebo. Eleven of them entered the meta-analysis — covering back pain, cancer-related fatigue, ADHD, allergic rhinitis, major depression, irritable bowel syndrome, and menopausal hot flashes — with a moderate overall effect compared with no treatment. The authors themselves, however, describe the field as being "in its infancy": there are few studies, with moderate risk of bias and heterogeneous results. Open-label placebo is a promising line of research, not a validated treatment to recommend to anyone — and never a substitute for interventions with demonstrated efficacy.
What to do with this in practice
Two opposite conclusions tend to get drawn too hastily — "so it's all placebo" and "so placebo is useless" — and both are wrong. What the accumulated knowledge suggests is more interesting:
- Deceiving a patient is a different matter. Prescribing a hidden placebo in routine clinical practice clashes with informed consent and with trust in the relationship. In research, the Declaration of Helsinki allows a placebo group when no proven intervention exists, or when there are solid methodological reasons and no additional risk of serious harm to participants.
- Context is a legitimate tool. The 2018 expert consensus converged on two points: informing patients about placebo and nocebo effects, and training professionals in communication, so as to harness the placebo and reduce the nocebo. Explaining well, giving time, being supportive, and building an honest expectation is not a trick: it is part of the treatment already being done.
- Without comparison, anything can look like it works. The placebo response — the effect of context added to the problem's natural course and regression to the mean — is one of the main reasons why therapies with no evidence collect enthusiastic testimonials. Hence the need for the randomized controlled trial, for blinding, and for the hierarchy of evidence — and hence the ease with which pseudosciences in mental health sustain themselves.
- Getting better afterward isn't getting better because of. The same caution applies here as in the entry on correlation and causality: the order of events does not prove what caused what.
Test your understanding
Quick quiz
Three questions to reinforce the concept. You'll see the explanation right away.
Further reading
- Hróbjartsson A, Gøtzsche PC. "Placebo interventions for all clinical conditions". Cochrane Database of Systematic Reviews, 2010, CD003974.
- Wechsler ME, Kelley JM, Boyd IO et al. "Active albuterol or placebo, sham acupuncture, or no intervention in asthma". New England Journal of Medicine, 2011; 365(2):119–126.
- Haas JW, Bender FL, Ballou S et al. "Frequency of adverse events in the placebo arms of COVID-19 vaccine trials: a systematic review and meta-analysis". JAMA Network Open, 2022; 5(1):e2143955.
- Evers AWM, Colloca L, Blease C et al. "Implications of placebo and nocebo effects for clinical practice: expert consensus". Psychotherapy and Psychosomatics, 2018; 87(4):204–210.
- Kaptchuk TJ, Friedlander E, Kelley JM et al. "Placebos without deception: a randomized controlled trial in irritable bowel syndrome". PLoS ONE, 2010; 5(12):e15591.
- von Wernsdorff M, Loef M, Tuschen-Caffier B, Schmidt S. "Effects of open-label placebos in clinical trials: a systematic review and meta-analysis". Scientific Reports, 2021; 11:3855.
- Benedetti F, Amanzio M, Vighetti S, Asteggiano G. "The biochemical and neuroendocrine bases of the hyperalgesic nocebo effect". Journal of Neuroscience, 2006; 26(46):12014–12022.
- de la Fuente-Fernández R, Ruth TJ, Sossi V et al. "Expectation and dopamine release: mechanism of the placebo effect in Parkinson's disease". Science, 2001; 293(5532):1164–1166.
- Levine JD, Gordon NC, Fields HL. "The mechanism of placebo analgesia". The Lancet, 1978; 2(8091):654–657.
- Kennedy WP. "The nocebo reaction". Medical World, 1961; 95:203–205.
- World Medical Association (WMA). Declaration of Helsinki — ethical principles for medical research involving human subjects.
This entry is educational and does not replace assessment, diagnosis or treatment by a qualified health professional. Every person is unique — for a specific case, seek individual guidance.