Skip to content
Back to the question

Evidence record

Why does a sugar pill work?

Every sentence on the exploration belongs to at least one claim below. Each says what kind of statement it is, what it rests on, what the evidence does not establish, and how far it was checked. Nothing here is a score, and none of it is a verdict on a belief or a theory.

Last checked 24 September 2026.

  • Source checked: 34
  • A judgement, not a fact: 13
  • Partly checked: 6
  • Presented as open: 1

If something here is wrong, misquoted, out of date or unfair, write to support@unknownoceans.com and give the claim number, which does not change.

What the question takes for granted

A scientific question with a false premise folded into it. A sugar pill does not do one thing: across conditions its effect is mainly on what patients report, with a few measured exceptions, and part of what is credited to it in trials is the natural course of the condition. The page separates those before asking why an effect survives honesty.

  • A sugar pill works

    Contested

    Against no treatment, placebo shows no important clinical effect in general and a modest one on what patients report, clearest for pain and nausea. Some reviewers read the controlled studies more broadly.

  • What improves in a placebo arm is the placebo effect

    Contested

    A placebo arm also carries the natural course of the condition and regression to the mean. In three-armed trials, patients given nothing improved a little over half as much as patients given a placebo.

  • A placebo only works if the patient is deceived

    Contested

    Open-label trials find an effect on self-reported outcomes without deception. Whether it reaches objective measures has not been shown.

How the question is labelled

Claim: Told it is nothing

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: Placebo

Kind
Framing. Signposting. It asserts nothing a source could check.

The question

Claim: Patients given a pill described to them, truthfully, as an inert placebo still reported more improvement than patients given nothing.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

The outcome is what patients report, and they knew which group they were in. The 2025 pooled analysis finds no significant effect on objective measures.

Later work looked for

Europe PMC, 2021 on, titles with "open-label placebo" and "meta-analysis" or "systematic review": Fendel and others (2025, sixty trials) confirm an effect on self-reports and find none that is significant on objective measures; the page now reports it. A published correction to the 2021 paper changes its publication-bias test, not its pooled figures.

Claim 3 · checked 24 September 2026 · AI-assisted audit

Claim: Some placebo effects are physiological events with identifiable brain mechanisms

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

Shown for pain relief and in Parkinson’s disease. It does not mean every placebo effect has a physiological basis, and across conditions the effect is mainly on what patients report.

Later work looked for

Europe PMC, 2016 on, placebo mechanism reviews: the 2021 participant-level imaging analysis (Zunhammer and others) and the 2025 open-label meta-analysis (Fendel and others) are the later work found; neither revises this point.

Corrected

24 September 2026. Previously: “Placebo responses are real physiological events with identifiable brain mechanisms” Stated of every placebo response, which the page’s own sources do not carry: the Cochrane review finds no important clinical effect in general, and the 2025 open-label meta-analysis finds no significant effect on objective measures. The mechanisms are shown for particular effects.

Claim 4 · checked 24 September 2026 · AI-assisted audit

Claim: and effects on what patients report survive being told that the pill is inert

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

Self-reported outcomes, with patients who know their group. On objective measures the pooled effect is 0.09 and not significant.

Later work looked for

Europe PMC, 2021 on, titles with "open-label placebo" and "meta-analysis" or "systematic review": Fendel and others (2025, sixty trials) confirm an effect on self-reports and find none that is significant on objective measures; the page now reports it. A published correction to the 2021 paper changes its publication-bias test, not its pooled figures.

Corrected

24 September 2026. Previously: “and they survive being told that the pill is inert” The open-label evidence is an effect on self-reports; the pooled effect on objective measures is not significant.

Claim 5 · checked 24 September 2026 · AI-assisted audit

Claim: What they are responses to, and which conditions they touch, is narrower than the popular version and stranger than it.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s own reading. It follows from claims 40 and 41 (narrower) and claims 18 and 65 (stranger).

Claim 6 · checked 24 September 2026 · AI-assisted audit

Claim: The usual explanation is deception: the patient believes in the treatment and the belief does the work.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

The trial’s authors describe the belief as widely held; they do not measure how widely.

Claim 7 · checked 24 September 2026 · AI-assisted audit

Claim: That explanation was tested by removing the deception,

Kind
Historical claim. A statement about the past that historical evidence can test.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Placebos without Deception: A Randomized Controlled Trial in Irritable Bowel SyndromeAbstract, Background

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Trial

    The source reads: We tested whether open-label placebo (non-deceptive and non-concealed administration) is superior to a no-treatment control with matched patient-provider interactions in the treatment of irritable bowel syndrome (IBS).

Claim 8 · checked 24 September 2026 · AI-assisted audit

Claim: which is the kind of experiment that either kills an effect or makes it more interesting.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s own comment on the design.

Claim 9 · checked 24 September 2026 · AI-assisted audit

What is established

Claim: What is established

Kind
Framing. Signposting. It asserts nothing a source could check.

It survives being told

Claim: It survives being told

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

What survives is an effect on self-reported outcomes.

Later work looked for

Europe PMC, 2021 on, titles with "open-label placebo" and "meta-analysis" or "systematic review": Fendel and others (2025, sixty trials) confirm an effect on self-reports and find none that is significant on objective measures; the page now reports it. A published correction to the 2021 paper changes its publication-bias test, not its pooled figures.

Claim 11 · checked 24 September 2026 · AI-assisted audit

Claim: In a 2010 randomised trial, patients with irritable bowel syndrome were given a bottle labelled placebo pills, and told plainly that the pills contained no active medication, while also being told that the placebo effect is powerful and that the body can respond to a placebo the way Pavlov’s dogs responded to a bell.

Kind
Historical claim. A statement about the past that historical evidence can test.
Review
Source checked. The cited passage was read, and it says this.

Both groups heard the persuasive rationale before randomisation; only one then received pills.

Later work looked for

Describes how a trial was run; not a finding later work revises.

Claim 12 · checked 24 September 2026 · AI-assisted audit

Claim: Over three weeks they improved more than patients on the same schedule of visits with no pills, by 0.79 standard deviations on a self-reported global improvement scale.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Placebos without Deception: A Randomized Controlled Trial in Irritable Bowel SyndromeResults; Discussion

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Trial

    The source reads: and the 21-day endpoint (5.0±1.5 vs. 3.9±1.3, p = .002, d = 0.79) ... Our primary outcome measure was the IBS Global Improvement Scale (IBS-GIS) which asks participants ... Given that patients in both treatment arms experienced the same frequency and duration of contact time

What this does not establish

One trial, 80 patients, one centre, mostly women. The scale asks patients to rate their own change.

Later work looked for

Europe PMC, 2011 on, "open-label placebo" and "irritable bowel": a larger six-week trial by the same group (Lembo and others, Pain 2021, 262 adults), read as its abstract, found a smaller effect against no pill (d = 0.43); it repeats the direction of the 2010 result and does not change what that trial found.

Claim 13 · checked 24 September 2026 · AI-assisted audit

Claim: The authors call it a proof of principle.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

Claim 14 · checked 24 September 2026 · AI-assisted audit

Claim: Eighty patients, three weeks, and the patients could not be blinded, so they could not rule out that part of the difference sits in the reporting rather than in the symptom.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Placebos without Deception: A Randomized Controlled Trial in Irritable Bowel SyndromeAbstract, Methods; Results; Discussion, Limitations

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Trial

    The source reads: Two-group, randomized, controlled three week trial ... 80 eligible patients were randomized into the two arms ... given the impossibility of double-blind assessment of open placebo versus no-treatment control, the effects of report bias cannot be eliminated

Corrected

24 September 2026. Previously: “and nobody could be blinded” The trial had its questionnaires administered by a researcher blind to allocation; it was the patients who could not be blinded, which is the source of the report bias the authors name.

Claim 15 · checked 24 September 2026 · AI-assisted audit

Claim: A 2021 meta-analysis of eleven such trials, across back pain, cancer-related fatigue, ADHD, allergic rhinitis, depression, hot flushes and irritable bowel, found 0.72 standard deviations (0.39 to 1.05), falling to 0.49 once the four trials at high risk of bias are dropped.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Effects of open-label placebos in clinical trials: a systematic review and meta-analysisAbstract

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Meta-analysis

    The source reads: Eleven trials were eligible for meta-analysis. These trials assessed effects of OLPs on back pain, cancer-related fatigue, attention deficit hyperactivity disorder, allergic rhinitis, major depression, irritable bowel syndrome and menopausal hot flushes.

  • Supports

    Effects of open-label placebos in clinical trials: a systematic review and meta-analysisResults, overall effect and sensitivity analysis

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Meta-analysis

    The source reads: SMD = 0.72, 95% CI 0.39–1.05, p < 0.0001 ... For an explorative sensitivity analysis, we excluded four studies with a high risk of bias ... The corresponding effect size was lower, but still significant (SMD = 0.49, 95% CI 0.32–0.66, p < 0.00001)

What this does not establish

Heterogeneity was high (I² 76 per cent) until the high-risk trials were removed. The authors call the research in its infancy.

Later work looked for

Europe PMC, 2021 on, titles with "open-label placebo" and "meta-analysis" or "systematic review": Fendel and others (2025, sixty trials) confirm an effect on self-reports and find none that is significant on objective measures; the page now reports it. A published correction to the 2021 paper changes its publication-bias test, not its pooled figures.

Corrected

24 September 2026. Previously: “across back pain, cancer-related fatigue, allergic rhinitis, depression, hot flushes and irritable bowel” Two of the eleven pooled trials were in children with ADHD; the list left them out.

Claim 16 · checked 24 September 2026 · AI-assisted audit

Claim: Every primary outcome in it was a self-report.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

The authors’ own classification; the two ADHD trials used ratings by parents and a study physician.

Claim 17 · checked 24 September 2026 · AI-assisted audit

Claim: A 2025 update pooled sixty trials, in patients and in non-clinical samples. It put the effect at 0.47 in patients and, taking every trial together, found it on self-reports (0.39) but not, to any significant degree, on objective measures (0.09).

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

Most trials were small, under a hundred participants, and the objective-outcome estimate rests on seventeen comparisons with wide uncertainty.

Later work looked for

Europe PMC, 2021 on, titles with "open-label placebo" and "meta-analysis" or "systematic review": Fendel and others (2025, sixty trials) confirm an effect on self-reports and find none that is significant on objective measures; the page now reports it. A published correction to the 2021 paper changes its publication-bias test, not its pooled figures.

Corrected

24 September 2026. Previously: “” The page rested the open-label result on a 2021 meta-analysis of eleven trials. A 2025 update of sixty trials is the current evidence, and it is the first to test self-reported against objective outcomes directly.

Claim 18 · checked 24 September 2026 · AI-assisted audit

Claim: Whatever is happening is not simply a false belief about a pill, because the belief was corrected in advance.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Follows from claim 12: patients were told the pill was inert.

Other readings

Patients were also given a persuasive rationale that the pill could help, so a true-seeming expectation of benefit replaced the false belief; and the patients’ own reports may carry some of the difference (claim 15).

Claim 19 · checked 24 September 2026 · AI-assisted audit

There is a mechanism, and it is specific

Claim: There is a mechanism, and it is specific

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The neuroscience of placebo effects: connecting context, learning and healthNeurophysiology of placebo analgesia

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: Placebo effects in cortical–brainstem systems depend in part on opioid involvement ... a range of other neurochemicals, including dopamine, cholecystokinin and oxytocin, are also likely to be involved in the placebo response

What this does not establish

A mechanism for placebo pain relief, which depends in part on opioids. No single mechanism covers every placebo effect.

Later work looked for

Europe PMC, 2016 on, placebo mechanism reviews: the 2021 participant-level imaging analysis (Zunhammer and others) and the 2025 open-label meta-analysis (Fendel and others) are the later work found; neither revises this point.

Claim 20 · checked 24 September 2026 · AI-assisted audit

Claim: Placebo analgesia engages the body’s own opioid system.

Kind
What the field holds. The settled view of a field, as a review, a consensus statement or a reference work gives it. Not one study.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The neuroscience of placebo effects: connecting context, learning and healthNeurophysiology of placebo analgesia, opioid systems

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: multiple studies have shown that placebo analgesia can be blocked by the opioid antagonist naloxone ... placebo effects on both pain-related increases and decreases in fMRI activity are blocked by naloxone

  • Supports

    The neuroscience of placebo effects: connecting context, learning and healthNeurophysiology of placebo analgesia, summary

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: Placebo effects in cortical–brainstem systems depend in part on opioid involvement ... a range of other neurochemicals, including dopamine, cholecystokinin and oxytocin, are also likely to be involved in the placebo response

What this does not establish

In part: the review lists dopamine, cholecystokinin and oxytocin as also likely to be involved.

Later work looked for

Europe PMC, 2010 on, titles with "placebo" and "naloxone": Benedetti and others (2023) extend naloxone reversal to open-label placebo analgesia; nothing found that overturns the opioid role.

Claim 21 · checked 24 September 2026 · AI-assisted audit

Claim: Naloxone, which blocks opioid receptors, reduces both the relief people report and the brain responses that go with it, and abolishes the coupling between cingulate cortex and periaqueductal grey that predicts the effect.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Partly checked. Only part of the source could be read. What was not read is listed.

  • Supports

    Activation of the opioidergic descending pain control system underlies placebo analgesiaabstract

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · Research in people · Experiment

    The source reads: Naloxone reduced both behavioral and neural placebo effects ... naloxone abolished placebo-induced coupling between rACC and PAG, which predicted both neural and behavioral placebo effects

  • Supports

    The neuroscience of placebo effects: connecting context, learning and healthNeurophysiology of placebo analgesia

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: multiple studies have shown that placebo analgesia can be blocked by the opioid antagonist naloxone ... placebo effects on both pain-related increases and decreases in fMRI activity are blocked by naloxone

What this does not establish

The Eippert paper was read only as its abstract, which does not give the sample or the pain procedure. Naloxone reduced the placebo effect; the abstract does not say it removed it.

Later work looked for

Europe PMC, 2010 on, titles with "placebo" and "naloxone": Benedetti and others (2023) extend naloxone reversal to open-label placebo analgesia; nothing found that overturns the opioid role.

Claim 22 · checked 24 September 2026 · AI-assisted audit

Claim: That is as direct a demonstration of a physical mechanism as this field offers.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s judgement. The 2025 meta-analysis describes the opioid finding as what changed the field’s view of placebo as mere suggestion.

Claim 23 · checked 24 September 2026 · AI-assisted audit

Claim: Expectation is not the only ingredient.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The neuroscience of placebo effects: connecting context, learning and healthUnderstanding the role of conceptual processes

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: there is growing evidence that a combination of pre-cognitive associations and conceptual processes may be required

Claim 24 · checked 24 September 2026 · AI-assisted audit

Claim: The review cited here puts placebo effects down to several processes at once, learning and expectation and social cognition,

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

Claim 25 · checked 24 September 2026 · AI-assisted audit

Claim: which is why the same pill does different things in different hands.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s reading of the review, which says no single process or system mediates all placebo effects.

Claim 26 · checked 24 September 2026 · AI-assisted audit

Claim: What a placebo response is built from

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: Expectation

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: What the patient anticipates, which is shaped by how a treatment is described and by the setting it is given in.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The neuroscience of placebo effects: connecting context, learning and healthPsychological processes underlying placebo effects

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: Verbal suggestions and background beliefs about the treatment context can engage multiple conceptual processes: expectations of specific outcomes ... the delivery of placebo manipulations in a context that includes both rich associative cues (for example, a hospital setting)

Claim 29 · checked 24 September 2026 · AI-assisted audit

Claim: Learning

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: Previous experience of relief in a similar context.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The neuroscience of placebo effects: connecting context, learning and healthPsychological processes underlying placebo effects

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: presentation of sensory cues associated with positive outcomes (for example, pain relief) or negative outcomes (for example, shock) through classical conditioning

Claim 31 · checked 24 September 2026 · AI-assisted audit

Claim: The open-label trials tell patients that this part works automatically.

Kind
Historical claim. A statement about the past that historical evidence can test.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Placebos without Deception: A Randomized Controlled Trial in Irritable Bowel SyndromeMethods, Interventions

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Trial

    The source reads: The provider clearly explained that the placebo pill was an inactive (i.e., “inert”) substance like a sugar pill that contained no medication ... 1) the placebo effect is powerful, 2) the body can automatically respond to taking placebo pills like Pavlov's dogs who salivated when they heard a bell

  • Supports

    Effects of open-label placebos in clinical trials: a systematic review and meta-analysisIntroduction; Table 2

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Meta-analysis

    The source reads: classical conditioning is a possible mechanism for automatic self-healing ... that the body can also respond automatically after taking an inert pill

Later work looked for

Describes how a trial was run; not a finding later work revises.

Corrected

24 September 2026. Previously: “This part is said to work without any conscious belief, which is one of the accounts offered for its surviving being told the pill is inert.” Said by whom was not stated, and the review the passage cites says the opposite for most cases: conditioned responses usually depend on what the cue is believed to mean, and only a minority of placebo effects are insensitive to belief. The claim that it works without belief is the open-label trials’ own script, and is now attributed to it.

Claim 32 · checked 24 September 2026 · AI-assisted audit

Claim: The review cited here finds that in most cases it depends on belief as well,

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The neuroscience of placebo effects: connecting context, learning and healthUnderstanding the role of conceptual processes

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: decades of empirical work suggests that in many cases conditioned responses in humans and rodents alike depend on the information value of the cue ... In a minority of cases, placebo effects may be insensitive to beliefs

What this does not establish

The review says a minority of placebo effects, such as conditioned immunosuppression, may be insensitive to belief.

Claim 33 · checked 24 September 2026 · AI-assisted audit

Claim: so it is at best a partial account of why being told the pill is inert does not spoil the effect.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Follows from claims 32 and 33: the account the trials give rests on a kind of learning the review finds usually depends on belief.

Claim 34 · checked 24 September 2026 · AI-assisted audit

Claim: Attention and interpretation

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: The same sensation is rated differently depending on what it is taken to mean.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The neuroscience of placebo effects: connecting context, learning and healthPsychological processes; Box 2

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: appraisals of the significance of both symptoms and treatment ... they may independently affect the evaluation and functional consequences of pain in others

Claim 36 · checked 24 September 2026 · AI-assisted audit

Claim: Pooled brain scans of placebo pain relief show the pain pathways quieting, but by a small amount, too little to account for all of the relief people report.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

Twenty laboratory studies of evoked pain in healthy volunteers; the authors say the results may not generalise to clinical settings.

Later work looked for

Europe PMC, author Zunhammer and "placebo": the 2021 participant-level analysis pools twenty imaging studies and finds the reductions small; the page is updated to it. A 2026 paper by the same group on conditioned against verbal placebo was not read.

Corrected

24 September 2026. Previously: “and that change is measurable in the brain’s pain pathways rather than only in the report.” Stated as a plain measurement. The review the passage cites says effects on the most pain-specific regions are not consistently found, and the 2021 pooled analysis of 603 participants finds the reductions small and unlikely to be a complete explanation. Source added.

Claim 37 · checked 24 September 2026 · AI-assisted audit

Where it applies

Claim: Where it applies

Kind
Framing. Signposting. It asserts nothing a source could check.

What it can and cannot change

Claim: What it can and cannot change

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: A Cochrane review of 234 trials that carried a no-treatment arm alongside the placebo arm, 202 of them with usable outcome data, found no important clinical effect from placebo in general.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Partly checked. Only part of the source could be read. What was not read is listed.

  • Supports

    Placebo interventions for all clinical conditionsabstract, Selection criteria; Main results; Authors' conclusions

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · Research in people · Meta-analysis

    The source reads: We included randomised placebo trials with a no‐treatment control group investigating any health problem. ... Outcome data were available in 202 out of 234 included trials, investigating 60 clinical conditions. ... We did not find that placebo interventions have important clinical effects in general.

What this does not establish

Read as its abstract; the forest plots and per-condition analyses were not read. It pools sixty conditions, and only 16 of the trials were judged at low risk of bias.

Later work looked for

Europe PMC, "Placebo interventions for all clinical conditions": the 2010 version is the latest. Nothing found that revises its pooled figures.

Corrected

24 September 2026. Previously: “A Cochrane review of 202 trials that carried a no-treatment arm alongside the placebo arm found no important clinical effect from placebo in general.” The review included 234 such trials; 202 is the number with outcome data available. The page now names both.

Claim 40 · checked 24 September 2026 · AI-assisted audit

Claim: What it found was mainly an effect on what patients report: 0.26 standard deviations on patient-reported outcomes against 0.13 on observer-reported ones, clearest for pain and nausea, and very variable for pain even among the trials at low risk of bias.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Partly checked. Only part of the source could be read. What was not read is listed.

  • Supports

    Placebo interventions for all clinical conditionsabstract, Main results

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · Research in people · Meta-analysis

    The source reads: The SMD for patient‐reported outcomes was ‐0.26 (95% CI ‐0.32 to ‐0.19), and for observer‐reported outcomes, SMD ‐0.13 (95% CI ‐0.24 to ‐0.02). ... The effect on pain was very variable, also among trials with low risk of bias.

  • Supports

    Placebo interventions for all clinical conditionsabstract, Authors' conclusions

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · Research in people · Meta-analysis

    The source reads: in certain settings placebo interventions can influence patient‐reported outcomes, especially pain and nausea, though it is difficult to distinguish patient‐reported effects of placebo from biased reporting

  • Qualifies

    Placebo interventions for all clinical conditionsabstract, Main results

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · Research in people · Meta-analysis

    The source reads: The pooled relative risk for patient‐reported outcomes was 0.93 (95% CI 0.86 to 1.00) and for observer‐reported outcomes 0.93 (95% CI 0.85 to 1.02).

What this does not establish

Read as its abstract. These are the continuous-outcome trials; in trials with yes-or-no outcomes the review found no clear difference between patient-reported and observer-reported effects.

Later work looked for

Europe PMC, "Placebo interventions for all clinical conditions": the 2010 version is the latest. Nothing found that revises its pooled figures.

Corrected

24 September 2026. Previously: “What it found was an effect on what patients report:” The observer-reported effect, 0.13, was small but its interval excludes zero; the effect was not confined to patients’ reports.

Claim 41 · checked 24 September 2026 · AI-assisted audit

Claim: Parkinson’s disease is the exception hardest to explain away.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s judgement, resting on claim 43: the outcome there is a scan rather than a report.

Claim 42 · checked 24 September 2026 · AI-assisted audit

Claim: A placebo releases dopamine in the striatum of patients whose own supply is failing, which is measurable and is nobody’s report of anything.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Partly checked. Only part of the source could be read. What was not read is listed.

  • Supports

    Expectation and dopamine release: mechanism of the placebo effect in Parkinson’s diseaseabstract

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · Research in people · Experiment

    The source reads: as measured by positron emission tomography, we provide in vivo evidence for substantial release of endogenous dopamine in the striatum of PD patients in response to placebo ... mediated through activation of the damaged nigrostriatal dopamine system

  • Supports

    The neuroscience of placebo effects: connecting context, learning and healthPlacebo effects beyond pain

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: found that placebo administration increased striatal dopamine binding in patients with PD, particularly in those who perceived an improvement in clinical status with placebo treatment ... PD patients’ therapeutic expectations are correlated with placebo-induced striatal dopamine release

What this does not establish

The original paper was read only as its abstract, which does not give the sample size. Release was measured by PET; it was larger in patients who perceived an improvement.

Later work looked for

Europe PMC, 2002 on, titles with "placebo", "Parkinson" and "dopamine": Lidstone and others (2010) tie the release to expectation (not read; described by Wager and Atlas). Nothing found that reverses the finding.

Claim 43 · checked 24 September 2026 · AI-assisted audit

Claim: The distinction is between changing the course of a disease and changing what it does to a person, and the popular version of the placebo effect ignores it.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s reading. The Cochrane review opens by noting that placebo is often claimed to improve outcomes substantially in many conditions.

  • Context

    Placebo interventions for all clinical conditionsabstract, Background

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · Research in people · Meta-analysis

    The source reads: Placebo interventions are often claimed to substantially improve patient‐reported and observer‐reported outcomes in many clinical conditions

Claim 44 · checked 24 September 2026 · AI-assisted audit

Claim: The reviewers add the caution that goes with their own result: on a self-reported outcome, a placebo effect is hard to separate from biased reporting.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Partly checked. Only part of the source could be read. What was not read is listed.

  • Supports

    Placebo interventions for all clinical conditionsabstract, Authors' conclusions

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · Research in people · Meta-analysis

    The source reads: in certain settings placebo interventions can influence patient‐reported outcomes, especially pain and nausea, though it is difficult to distinguish patient‐reported effects of placebo from biased reporting

What this does not establish

Read as its abstract.

Claim 45 · checked 24 September 2026 · AI-assisted audit

Claim: Not every review reads the evidence so narrowly.

Kind
Disputed. A contested point, where the page reports a side or the disagreement.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

  • Context

    The neuroscience of placebo effects: connecting context, learning and healthIntroduction

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: have demonstrated causal effects of placebo treatment on measures that are typically used as primary disease end points ... in multiple forms of chronic pain ... Placebo effects can be as large as the effects of accepted drug treatments

  • Context

    Placebo interventions for all clinical conditionsabstract, Authors' conclusions

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · Research in people · Meta-analysis

    The source reads: We included randomised placebo trials with a no‐treatment control group investigating any health problem. ... Outcome data were available in 202 out of 234 included trials, investigating 60 clinical conditions. ... We did not find that placebo interventions have important clinical effects in general.

Other readings

Hróbjartsson and Gøtzsche find no important clinical effect in general; Wager and Atlas point to a subset of controlled studies with large effects. The page reports both.

Corrected

24 September 2026. Previously: “” The passage gave only the Cochrane reading. The mechanism review the page already cites takes a broader view of the same kind of evidence, and the brief asks for a real disagreement to be reported with who holds each side.

Claim 46 · checked 24 September 2026 · AI-assisted audit

Claim: Wager and Atlas cite controlled studies in chronic pain, depression and Parkinson’s in which placebo moved the measures trials use as primary end points, in some cases as far as accepted drugs do.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The neuroscience of placebo effects: connecting context, learning and healthIntroduction, clinical and physiological outcomes

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: have demonstrated causal effects of placebo treatment on measures that are typically used as primary disease end points ... in multiple forms of chronic pain ... Placebo effects can be as large as the effects of accepted drug treatments

    The same sentence also lists depression and Parkinson’s disease, with reference numbers attached to each word; asthma is listed with a contrary reference and is left off the page.

What this does not establish

A narrative review citing a small subset of studies; it does not pool them.

Claim 47 · checked 24 September 2026 · AI-assisted audit

How much of a drug trial is this

Claim: How much of a drug trial is this

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: Everyone in a trial receives the context, so the measured effect of a drug is the part that exceeds it.

Kind
What a model says. What a theory or model predicts or describes, which is not the same as what has been observed.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The neuroscience of placebo effects: connecting context, learning and healthIntroduction

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: Placebo effects are beneficial effects that are attributable to the brain–mind responses to the context in which a treatment is delivered rather than to the specific actions of the drug.

  • Qualifies

    The neuroscience of placebo effects: connecting context, learning and healthIntroduction

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Review of the field

    The source reads: In many cases, hidden drug administration, which eliminates patients’ treatment expectations, markedly reduces the effects of drugs

  • Qualifies

    Efficacy of antidepressants: a re-analysis and re-interpretation of the Kirsch dataabstract

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · Research in people · Meta-analysis

    The source reads: Kirsch et al.'s meta-analysis suffered from important flaws in the calculations ... Overall the results suggest that although a large percentage of the placebo response is due to expectancy this is not true for the active drug and effects are not additive.

What this does not establish

The subtraction assumes drug and context add. Hidden administration shows they can interact, and Fountoulakis and Möller argue that in antidepressant trials the effects are not additive.

Claim 49 · checked 24 September 2026 · AI-assisted audit

Claim: Where the placebo arm improves a lot, which includes pain and depression,

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

The depression figure is from four antidepressants in FDA trial data; the pain figure is Kirsch’s summary of other estimates, which were not read.

Later work looked for

Europe PMC, the published re-analysis by Fountoulakis and Möller (2011) disputes Kirsch’s calculations and his reading of them; cited as the other side.

Claim 50 · checked 24 September 2026 · AI-assisted audit

Claim: that subtraction gets difficult and the resulting effect sizes are contested.

Kind
Disputed. A contested point, where the page reports a side or the disagreement.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Other readings

Kirsch and colleagues find the drug–placebo difference relatively small; Fountoulakis and Möller re-analyse the same data and find it larger for some drugs. The page takes no side.

Claim 51 · checked 24 September 2026 · AI-assisted audit

Claim: Much of that improvement is not the placebo effect.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

Three-armed trials, mostly of psychological and physical treatments rather than drugs.

Later work looked for

Europe PMC, 2010 on, titles with "placebo response" or "placebo effect" and "no-treatment": nothing found that replaces the three-armed estimate.

Corrected

24 September 2026. Previously: “Most of that improvement is not the placebo effect.” The three-armed trials put spontaneous improvement at a little over half of the placebo arm’s change, which is "much", not clearly "most".

Claim 52 · checked 24 September 2026 · AI-assisted audit

Claim: A placebo arm also records the natural course of the condition and regression to the mean.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

Describes how a trial arm is made up; not a finding later work revises.

Claim 53 · checked 24 September 2026 · AI-assisted audit

Claim: Across 37 trials that added a no-treatment arm, patients given nothing improved a little over half as much as patients given a placebo.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

Only five of the 37 trials tested a drug; the proportions varied by condition, and the intervals are wide.

Later work looked for

Europe PMC, 2010 on, titles with "placebo response" or "placebo effect" and "no-treatment": nothing found that replaces the three-armed estimate.

Corrected

24 September 2026. Previously: “and the trials that separate those out by adding a no-treatment arm find the remainder small.” Cited to the Cochrane review, which pools placebo against no treatment but does not report how much of a placebo arm’s change is spontaneous. The three-armed analysis from the same authors does, and gives a number.

Claim 54 · checked 24 September 2026 · AI-assisted audit

Claim: Part of the long argument about how well antidepressants work is really an argument about this.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Both sides of the dispute over Kirsch’s analysis turn on the size of the placebo response and whether drug and placebo effects add.

  • Context

    Initial severity and antidepressant benefits: a meta-analysis of data submitted to the Food and Drug AdministrationDiscussion

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Meta-analysis

    The source reads: The response to placebo in these trials was exceptionally large, duplicating more than 80% of the improvement observed in the drug groups. In contrast, the effect of placebo on pain is estimated to be about 50% of the response to pain medication

  • Context

    Efficacy of antidepressants: a re-analysis and re-interpretation of the Kirsch dataabstract

    Cited · Link answered, 19 October 2026 · Abstract only, 24 September 2026 · Research in people · Meta-analysis

    The source reads: Kirsch et al.'s meta-analysis suffered from important flaws in the calculations ... Overall the results suggest that although a large percentage of the placebo response is due to expectancy this is not true for the active drug and effects are not additive.

Claim 55 · checked 24 September 2026 · AI-assisted audit

The part with no account

Claim: The part with no account

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: An effect that needs no deception is harder to explain than one that does.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s own line between sections.

Claim 57 · checked 24 September 2026 · AI-assisted audit

Why telling the truth does not spoil it

Claim: Why telling the truth does not spoil it

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: No accepted account explains the open-label result.

Kind
Open question. Something the page says is not known.
Review
Presented as open. The page says this is unresolved, and the review found that it is.

Claim 59 · checked 24 September 2026 · AI-assisted audit

Claim: Conditioning is the account the trials themselves offer patients.

Kind
Historical claim. A statement about the past that historical evidence can test.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Effects of open-label placebos in clinical trials: a systematic review and meta-analysisTable 2, rationales

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Meta-analysis

    The source reads: classical conditioning is a possible mechanism for automatic self-healing ... that the body can also respond automatically after taking an inert pill

  • Supports

    Placebos without Deception: A Randomized Controlled Trial in Irritable Bowel SyndromeMethods

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Trial

    The source reads: The provider clearly explained that the placebo pill was an inactive (i.e., “inert”) substance like a sugar pill that contained no medication ... 1) the placebo effect is powerful, 2) the body can automatically respond to taking placebo pills like Pavlov's dogs who salivated when they heard a bell

Later work looked for

Describes how a trial was run; not a finding later work revises.

Claim 60 · checked 24 September 2026 · AI-assisted audit

Claim: Reinterpretation is another: the instruction contradicts itself, this pill may help and this pill is inert, and that may change what a sensation is taken to mean.

Kind
Hypothesis. A proposed explanation, not an established one.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Effects of open-label placebos in clinical trials: a systematic review and meta-analysisDiscussion

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Meta-analysis

    The source reads: the contradictory messages embedded in the provided narrative “this placebo pill may help; it’s an inert pill without physiological effect” can produce a cognitive dissonance ... The cognitive dissonance due to the OLP instruction might alter the familiar interpretation of symptoms and can cause a less intense sensation.

What this does not establish

Proposed by Ongaro and Kaptchuk and reported by the meta-analysis; not tested there.

Claim 61 · checked 24 September 2026 · AI-assisted audit

Claim: And the people who volunteer have often run out of other options, so new hope is hard to exclude.

Kind
Hypothesis. A proposed explanation, not an established one.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Effects of open-label placebos in clinical trials: a systematic review and meta-analysisDiscussion

    Cited · Link answered, 19 October 2026 · Read in full, 24 September 2026 · Research in people · Meta-analysis

    The source reads: Many patients were frustrated by multiple unsuccessful treatments and chose this counterintuitive intervention from a state of despair ... which may have produced new hope after the previous psychological strain

What this does not establish

The meta-analysis offers this as a possible factor; it does not measure it.

Claim 62 · checked 24 September 2026 · AI-assisted audit

Claim: It is a repeatable effect on what patients report, with no agreed mechanism,

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

Repeatable in patients, where the prediction interval excludes zero; in non-clinical samples it may include no effect.

Later work looked for

Europe PMC, 2021 on, titles with "open-label placebo" and "meta-analysis" or "systematic review": Fendel and others (2025, sixty trials) confirm an effect on self-reports and find none that is significant on objective measures; the page now reports it. A published correction to the 2021 paper changes its publication-bias test, not its pooled figures.

Corrected

24 September 2026. Previously: “It is a repeatable effect with no agreed mechanism,” What repeats is an effect on self-reports; the pooled objective effect is not significant.

Claim 63 · checked 24 September 2026 · AI-assisted audit

Claim: which is an unusual and productive position for a result to be in.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s own judgement.

Claim 64 · checked 24 September 2026 · AI-assisted audit

Claim: One laboratory study found open-label relief of experimentally induced pain reversed by naloxone, as ordinary placebo relief is, in the four in ten people who responded at all.

Kind
Experimental result. What an experiment or trial found, under its own conditions. A result in animals, cells or a simulation is not a result in people.
Review
Partly checked. Only part of the source could be read. What was not read is listed.

What this does not establish

Read as its abstract. One study, of arm pain induced with a tourniquet, and only in the 40.6 per cent who responded; it says nothing about the self-reported benefit in clinical trials.

Later work looked for

Europe PMC, 2010 on, titles with "placebo" and "naloxone": Benedetti and others (2023) extend naloxone reversal to open-label placebo analgesia; nothing found that overturns the opioid role.

Corrected

24 September 2026. Previously: “” The page said nothing about mechanism work on open-label placebo. The one experiment the 2025 meta-analysis knows of is now reported, at its own scale.

Claim 65 · checked 24 September 2026 · AI-assisted audit

Claim: The size of the effect is not settled, and on objective measures it has not been shown.

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

Later work looked for

Europe PMC, 2021 on, titles with "open-label placebo" and "meta-analysis" or "systematic review": Fendel and others (2025, sixty trials) confirm an effect on self-reports and find none that is significant on objective measures; the page now reports it. A published correction to the 2021 paper changes its publication-bias test, not its pooled figures.

Corrected

24 September 2026. Previously: “The literature is also still small, and the size of the effect is not settled.” Out of date: the 2025 update pools sixty trials, so the literature is no longer small, though most trials still are. What it adds is that no significant effect on objective measures has been shown.

Claim 66 · checked 24 September 2026 · AI-assisted audit

What was read