• Asbjørn Hróbjartsson and Peter C Gøtzsche.
• Afdeling for Medicinsk Videskabsteori, Københavns Universitet, Panum Instituttet, DK-2200 København N. a.hrobjartsson@cochrane.dk
• Ugeskr. Laeg. 2002 Jan 14; 164 (3): 329-33.

IntroductionPlacebo treatments have been reported to help patients with many diseases, but the quality of the evidence supporting this belief has not been rigorously evaluated.Material And MethodsWe conducted a systematic review of clinical trials in which patients were randomised either to placebo treatment or to no treatment. A placebo intervention could be pharmacological (for instance, a tablet), physical (for instance manipulation), or psychological (for instance conversation).ResultsWe identified 130 trials that met our inclusion criteria. After excluding 16, which gave no relevant outcome data, 32 with binary outcomes and 82 with continuous outcomes were left. As compared to no treatment, placebo had no significant effect on binary outcomes, the pooled relative risk was 0.95 (95% confidence interval 0.88 to 1.02), irrespective of whether these outcomes were subjective or objective. Placebo seemed to have a beneficial effect in the trials with continuous outcomes, the overall pooled standardised mean difference of the placebo and no-treatment groups was -0.28 (-0.38 to -0.19), but the effect decreased with increasing sample size, which indicates bias. The pooled standardised mean difference was significant for the trials with subjective outcomes, -0.36 (-0.47 to -0.25), but not for those with objective outcomes, -0.12 (-0.27 to 0.03). In trials involving the treatment of pain, placebo was reported to have a moderate beneficial effect, with a pooled standardised mean difference of -0.27 (-0.40 to -0.15), but bias, for instance reporting bias, was possible.ConclusionWe found no reliable evidence that placebo treatments in general had effects of clinical importance.

5 keywords...

hide…