Is the Placebo Effect Real? Why Fake Pills Change How We Feel, and the Nocebo Effect
- The placebo effect is an improvement in symptoms after a treatment with no active ingredient.
- Drug trials use placebo control groups and double-blinding to separate a medicine's effect from expectation.
- The nocebo effect is the reverse: negative expectations produce real discomfort.
Contents
You swallow a pill made of sugar and your headache fades. Is that just in your head, or did something in your body actually change? Medicine and psychology have taken that question seriously for more than seventy years. This guide covers what the placebo effect is, why it happens, how researchers keep it out of drug trials, and what it says about the way expectation shapes experience.
Why would a pill with nothing in it ease a symptom at all?
What is the placebo effect?
Placebo is Latin for "I shall please". In medicine it means a dummy pill or sham procedure with no active ingredient, and the placebo effect is the improvement people report after receiving one. In 1955 the American anaesthetist Henry Beecher pulled together a series of studies in a JAMA paper titled "The Powerful Placebo" and concluded that roughly a third of patients improved on placebo alone. Later researchers argued that his figure was inflated, because some of those patients would have got better anyway, but the paper put placebo research on the map.
Sceptics in Britain will know the debate from homeopathy. In 2017 NHS England advised GPs to stop routinely prescribing homeopathic remedies, on the grounds that there was no clear evidence they work beyond placebo. That decision rested on exactly the comparison this article is about.
Why do clinical trials use placebos?
Because expectation alone can move results, a new medicine has to beat a placebo, not just do better than nothing. One group gets the real drug, the other gets a pill that looks, tastes and weighs the same, and the outcomes are compared. When neither the patients nor the clinicians know who received what, the trial is double-blind. Blinding the doctors matters too: a researcher who knows which patient got the real drug may, without meaning to, ask questions differently or record borderline results more generously.
Placebos are not the only trap. Several other things make a treatment look better than it is:
- Regression to the mean: people tend to start treatment when symptoms are at their worst, so many would drift back towards normal anyway.
- Natural course: colds and many aches clear up on their own, whatever you take.
- The Hawthorne effect: being watched changes behaviour.
- Confirmation bias: if you want something to work, you notice the good days.
- Co-interventions: people often change sleep, diet or exercise at the same time.
Keep this list in mind next time a friend says "I took this and it cured me". The experience is real, but on its own it cannot prove the treatment caused the change. That is why the randomised controlled trial sits near the top of the evidence hierarchy: random allocation spreads age, health and hopefulness evenly across the groups, so any difference left over can be credited to the treatment.
Skimming the section headings first makes a long guide much easier to map out. Try writing down one question of your own as you read.
How can a fake pill have a real effect?
The most common explanation is expectation. Believing you will feel better changes how pain is perceived and how the body responds to stress. A classic 1978 study in The Lancet by Levine, Gordon and Fields gave dental patients a placebo painkiller; when some of them were then given naloxone, a drug that blocks the body's own opioids, the placebo relief largely disappeared. That suggested the brain was releasing natural painkillers in response to the expectation of relief.
Conditioning is the second strand. If taking a pill has repeatedly been followed by feeling better, the act of taking a pill can start to trigger the response by itself, much as Pavlov's dogs learned to salivate at a bell. In the 1970s Robert Ader and Nicholas Cohen even showed that rats could be conditioned to suppress their immune response to a flavoured drink.
The limits matter just as much. Placebos do not shrink tumours or clear infections. Their effects show up mostly in subjective symptoms such as pain, nausea, fatigue and anxiety, which is exactly where the brain has the most say.
What is the nocebo effect?
Nocebo is the mirror image: negative expectation produces real discomfort. People who read a long list of possible side effects sometimes report those very symptoms while taking a dummy pill. In trials, the placebo group regularly complains of headaches, dizziness and tiredness.
Statins are the best-known example. In a 2020 study in the New England Journal of Medicine, sixty patients in London who had stopped statins because of side effects took statins, placebo tablets and no tablets in rotating months without knowing which was which. About 90 per cent of the symptom burden they reported on statins also appeared in the placebo months. The aches were real; much of their cause was expectation.
That puts doctors in a bind. Patients have a right to accurate information, yet the way it is delivered can create symptoms. The answer is not to hide side effects but to explain them in proportion, for instance by saying how many people out of a hundred actually experience them.
Can a placebo work if you know it's a placebo?
One of the more surprising lines of research is the open-label placebo. In 2010 Ted Kaptchuk's team at Harvard Medical School gave eighty patients with irritable bowel syndrome either no treatment or pills they were told plainly were "made of an inert substance, like sugar pills". After three weeks the open-placebo group reported noticeably more relief. Similar results have since been reported for chronic back pain and fatigue.
Read these findings carefully. They involve self-reported symptoms, not blood tests or scans, and they do not mean an illness can be cured by willpower. What they do suggest is that the ritual of treatment, a caring consultation and a clear routine, carries some benefit of its own.
Why does a pill's colour or price matter?
A placebo has to match the real drug in size, colour and taste, because the slightest difference can tip patients off and change their expectations. Studies have found that these cues influence results in their own right: capsules can seem stronger than tablets, injections more powerful than pills, and in one experiment volunteers told a dummy painkiller cost more reported greater relief than those told it was discounted. The same inactive substance can produce a different experience depending on how it is presented.
Ethics come into play as well. If an effective treatment already exists, giving patients only a placebo can deny them care. The World Medical Association's Declaration of Helsinki restricts placebo use in those situations, so many trials compare a new treatment against the current standard rather than against a sugar pill.
Where else does expectation change outcomes?
The Pygmalion effect is the classroom version. In 1968 Robert Rosenthal and Lenore Jacobson told teachers that certain pupils were about to bloom intellectually; the pupils had in fact been picked at random, yet many of them made larger gains over the year. Expectations shape how people are treated, and that treatment shapes results. The placebo effect is the version that plays out in the body.
Research design terms at a glance
| Term | Meaning | Problem it prevents |
|---|---|---|
| Random allocation | Participants assigned to groups by chance | Differences between groups at the start |
| Single-blind | Participants don't know which treatment they receive | Participants' expectations |
| Double-blind | Neither participants nor researchers know | Researcher bias as well |
| Placebo control group | A group given an inactive treatment | Confusing natural recovery or expectation with drug effect |
The same habits help outside an exam. When a health claim or advert crosses your feed, ask whether there was a comparison group, how many people took part and who paid for the study. If an assignment asks you to apply these ideas, pick one experience of your own and decide whether expectation, conditioning or nocebo explains it best. CampusDown's introductory psychology notes take the same approach, starting from a personal example and working outwards to the theory.
Seen this way, the placebo effect is not just about "fake" medicine. It shows how much trust, explanation and the relationship between patient and clinician matter to how people experience treatment, and it is a good reason to ask not only whether something worked but how anyone knows.
Reviewed by the CampusDown editorial team, November 2026. Sources: Beecher, H. K. (1955). The powerful placebo. JAMA, 159(17); Kaptchuk, T. J. et al. (2010). Placebos without deception. PLoS ONE; Wood, F. A. et al. (2020). N-of-1 trial of a statin, placebo, or no treatment. NEJM, 383; Rosenthal, R. & Jacobson, L. (1968). Pygmalion in the Classroom.
FAQ
Where does the word placebo come from?
It is Latin for "I shall please".
Which paper launched modern placebo research?
Henry Beecher's "The Powerful Placebo", published in JAMA in 1955.
Can a placebo work if you know it is a placebo?
A 2010 Harvard study of irritable bowel syndrome found that patients told openly they were taking sugar pills still reported more relief than an untreated group.
What is the nocebo effect?
Symptoms or side effects that appear because a person expects them, even when the treatment is inactive.