Skip to content

Kaptchuk TJ, Kelley JM, Conboy LA, et al. Components of placebo effect: randomised controlled trial in patients with irritable bowel syndrome. BMJ. 2008;336(7651):999-1003. PMID 18390493

One-paragraph summary

Two hundred and sixty-two adults (76% women, mean age 39, SD 14) diagnosed by Rome II criteria with an IBS symptom severity score ≥150 were randomised at an academic medical centre to one of three arms for three weeks: waiting list (assessment and observation only); "limited" placebo acupuncture (the therapeutic ritual, with a business-like practitioner interaction); or "augmented" placebo acupuncture (the same ritual plus a patient–practitioner relationship deliberately augmented with warmth, attention and confidence). At three weeks half the patients were re-randomised to continue for a further three weeks. The design decomposes the placebo response into assessment, ritual and relationship, and adds them back one at a time. Every outcome rose monotonically across the three arms: global improvement scale 3.8 (SD 1.0) → 4.3 (1.4) → 5.0 (1.3); adequate relief 28% → 44% → 62%; symptom severity score change 30 (63) → 42 (67) → 82 (89); quality of life 3.6 (8.1) → 4.1 (9.4) → 9.3 (14.0) — all p<0.001 for trend. All pairwise comparisons between augmented and limited relationship were significant, and results were similar at six-week follow-up. Registered NCT00065403.

Key findings

  • The placebo response is decomposable and dose-like: components "can be progressively combined in a manner resembling a graded dose escalation of component parts."
  • The patient–practitioner relationship is the largest single component, adding 18 percentage points of adequate relief on top of the ritual (44% → 62%).
  • Assessment and observation alone produced 28% adequate relief — a baseline against which every uncontrolled IBS treatment claim should be read.
  • The absolute gain from relationship alone (18 points) exceeds the drug–placebo difference of every licensed IBS drug in this knowledge base.
  • Effects persisted at six weeks, so this is not a transient novelty response.

Limitations

  • Single centre, Rome II criteria, predominantly female, tertiary academic setting.
  • Three-week primary assessment; six-week follow-up.
  • "Augmented" and "limited" interactions were delivered by the same practitioners under instruction, so the manipulation is of behaviour rather than of a naturally occurring variable.
  • The ritual used was sham acupuncture, which is a high-salience intervention; whether the same gradient holds for a tablet is untested here — though the open-label placebo trials that followed suggest it does (Kaptchuk 2010, PMID 21203519; Lembo 2021, PMID 33605656).
  • Outcomes are entirely patient-reported, as they must be in this condition.

Why it matters

This trial converted "placebo response" from a nuisance parameter into a measurable, manipulable clinical variable, and it did so in IBS specifically because IBS is defined and treated by symptom report. Its practical implication is uncomfortable for drug-centred practice: the consultation itself has an effect size comparable to, or larger than, the drugs prescribed within it. It anchors the subsequent open-label placebo programme — in which placebos given with full disclosure beat no-pill control and did not differ significantly from double-blind placebo (d=0.10; Lembo 2021, PMID 33605656) — and it explains, mechanistically, why a randomised 2×2 factorial trial found the label "mebeverine" nearly doubling response while the drug did nothing (Rexwinkel 2025, PMID 40074185).

Cited by wiki pages

  • placebo-response-and-trial-design
  • quality-of-life-and-stigma
  • overview