Placebo interventions are often claimed to substantially improve many clinical conditions. However, most reports on effects of placebos are based on unreliable studies that have not randomised patients to placebo or no treatment.
We studied the effect of placebo treatments by reviewing 202 trials comparing placebo treatment with no treatment covering 60 healthcare problems. In general, placebo treatments produced no major health benefits, although on average they had a modest effect on outcomes reported by patients, such as pain. However, the effect on pain varied from large to non-existent, even in well-conducted trials. Variations in the effect of placebo was partly explained by variations in how trials were conducted, the type of placebo used, and whether patients were informed that the trial involved placebo.
We did not find that placebo interventions have important clinical effects in general. However, 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. The effect on pain varied, even among trials with low risk of bias, from negligible to clinically important. Variations in the effect of placebo were partly explained by variations in how trials were conducted and how patients were informed.
Placebo interventions are often claimed to substantially improve patient-reported and observer-reported outcomes in many clinical conditions, but most reports on effects of placebos are based on studies that have not randomised patients to placebo or no treatment. Two previous versions of this review from 2001 and 2004 found that placebo interventions in general did not have clinically important effects, but that there were possible beneficial effects on patient-reported outcomes, especially pain. Since then several relevant trials have been published.
Our primary aims were to assess the effect of placebo interventions in general across all clinical conditions, and to investigate the effects of placebo interventions on specific clinical conditions. Our secondary aims were to assess whether the effect of placebo treatments differed for patient-reported and observer-reported outcomes, and to explore other reasons for variations in effect.
We searched the Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library Issue 4, 2007), MEDLINE (1966 to March 2008), EMBASE (1980 to March 2008), PsycINFO (1887 to March 2008) and Biological Abstracts (1986 to March 2008). We contacted experts on placebo research, and read references in the included trials.
We included randomised placebo trials with a no-treatment control group investigating any health problem.
Two authors independently assessed trial quality and extracted data. We contacted study authors for additional information. Trials with binary data were summarised using relative risk (a value of less than 1 indicates a beneficial effect of placebo), and trials with continuous outcomes were summarised using standardised mean difference (a negative value indicates a beneficial effect of placebo).
Outcome data were available in 202 out of 234 included trials, investigating 60 clinical conditions. We regarded the risk of bias as low in only 16 trials (8%), five of which had binary outcomes.
In 44 studies with binary outcomes (6041 patients), there was moderate heterogeneity (P < 0.001; I2 45%) but no clear difference in effects between small and large trials (symmetrical funnel plot). The overall pooled effect of placebo was a relative risk of 0.93 (95% confidence interval (CI) 0.88 to 0.99). 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). We found no statistically significant effect of placebo interventions in four clinical conditions that had been investigated in three trials or more: pain, nausea, smoking, and depression, but confidence intervals were wide. The effect on pain varied considerably, even among trials with low risk of bias.
In 158 trials with continuous outcomes (10,525 patients), there was moderate heterogeneity (P < 0.001; I2 42%), and considerable variation in effects between small and large trials (asymmetrical funnel plot). It is therefore a questionable procedure to pool all the trials, and we did so mainly as a basis for exploring causes for heterogeneity. We found an overall effect of placebo treatments, standardised mean difference (SMD) -0.23 (95% CI -0.28 to -0.17). 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). We found an effect on pain, SMD -0.28 (95% CI -0.36 to -0.19)); nausea, SMD -0.25 (-0.46 to -0.04)), asthma (-0.35 (-0.70 to -0.01)), and phobia (SMD -0.63 (95% CI -1.17 to -0.08)). The effect on pain was very variable, also among trials with low risk of bias. Four similarly-designed acupuncture trials conducted by an overlapping group of authors reported large effects (SMD -0.68 (-0.85 to -0.50)) whereas three other pain trials reported low or no effect (SMD -0.13 (-0.28 to 0.03)). The pooled effect on nausea was small, but consistent. The effects on phobia and asthma were very uncertain due to high risk of bias. There was no statistically significant effect of placebo interventions in the seven other clinical conditions investigated in three trials or more: smoking, dementia, depression, obesity, hypertension, insomnia and anxiety, but confidence intervals were wide.
Meta-regression analyses showed that larger effects of placebo interventions were associated with physical placebo interventions (e.g. sham acupuncture), patient-involved outcomes (patient-reported outcomes and observer-reported outcomes involving patient cooperation), small trials, and trials with the explicit purpose of studying placebo. Larger effects of placebo were also found in trials that did not inform patients about the possible placebo intervention.