This is not the most recent version of the article. View current version (20 JAN 2010)

Intervention Review

Placebo interventions for all clinical conditions

  1. Asbjørn Hróbjartsson*,
  2. Peter C Gøtzsche

Editorial Group: Cochrane Consumers and Communication Group

Published Online: 19 APR 2004

Assessed as up-to-date: 29 DEC 2002

DOI: 10.1002/14651858.CD003974.pub2

How to Cite

Hróbjartsson A, Gøtzsche PC. Placebo interventions for all clinical conditions. Cochrane Database of Systematic Reviews 2004, Issue 2. Art. No.: CD003974. DOI: 10.1002/14651858.CD003974.pub2.

Author Information

  1. Rigshospitalet, The Nordic Cochrane Centre, Copenhagen, Denmark

*Asbjørn Hróbjartsson, The Nordic Cochrane Centre, Rigshospitalet, Blegdamsvej 9, 3343, Copenhagen, 2100, Denmark. ah@cochrane.dk.

Publication History

  1. Publication Status: Unchanged
  2. Published Online: 19 APR 2004

SEARCH

This is not the most recent version of the article. View current version (20 JAN 2010)

 

Abstract

  1. Top of page
  2. Abstract
  3. Plain language summary

Background

Placebo interventions are often claimed to improve patient-reported and observer-reported outcomes, but this belief is not based on evidence from randomised trials that compare placebo with no treatment.

Objectives

To assess the effect of placebo interventions.

Search strategy

We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library Issue 4, 2002), MEDLINE (1966 to 2002), EMBASE (1980 to 2002), Biological Abstracts (1986 to 2002), and PsycLIT (1887 to 2002). We contacted experts on placebo research, and read references in the included trials.

Selection criteria

We included randomised placebo trials with a no-treatment control group investigating any health problem.

Data collection and analysis

Two review authors independently assessed trial quality and extracted data. We contacted study authors for additional information.

Main results

Outcome data were available in 156 out of 182 included trials, investigating 46 clinical conditions.

We found no statistically significant pooled effect of placebo in 38 studies with binary outcomes (4284 patients), relative risk 0.95 (95% confidence interval (CI) 0.89 to 1.01). The pooled relative risk for patient-reported outcomes was 0.95 (95% CI 0.88 to 1.03) and for observer-reported outcomes 0.91 (95% CI 0.81 to 1.03). There was heterogeneity (P=0.01) but the funnel plot was symmetrical. There was no statistically significant effect of placebo interventions in the four clinical conditions investigated in three trials or more: pain, nausea, smoking, and depression, but confidence intervals were wide.

We found an overall effect of placebo treatments in 118 trials with continuous outcomes (7453 patients), standardised mean difference (SMD) -0.24 (95% CI -0.31 to -0.17). The SMD for patient-reported outcomes was -0.30 (95% CI -0.38 to -0.21), whereas no statistically significant effect was found for observer-reported outcomes, SMD -0.10 (95% CI -0.20 to -0.01). There was heterogeneity (P<0.001) and large variability in funnel plot results even for big trials. There was an apparent effect of placebo interventions on pain (SMD -0.25 (95% CI -0.35 to-0.16)), and phobia (SMD -0.63 (95% CI -1.17 to -0.08)); but also a substantial risk of bias. There was no statistically significant effect of placebo interventions in eight other clinical conditions investigated in three trials or more: nausea, smoking, depression, overweight, asthma, hypertension, insomnia and anxiety, but confidence intervals were wide.

Authors' conclusions

There was no evidence that placebo interventions in general have clinically important effects. A possible small effect on continuous patient-reported outcomes, especially pain, could not be clearly distinguished from bias.

 

Plain language summary

  1. Top of page
  2. Abstract
  3. Plain language summary

Placebo treatments for all clinical conditions

It has been widely believed that placebo (dummy) treatments (for example sugar tablets) are associated with substantial effects on a wide range of health problems. However, this belief is not based on evidence from randomised trials that use a placebo treatment for one group of people, while another group receives no treatment. The effect of placebo treatments was studied by reviewing more than 150 such trials covering many types of healthcare problems. Placebo treatments caused no major health benefits, although they possibly had a small effect on outcomes reported by patients, for example pain.