Linde, Klaus; Niemann, Karin; Meissner, Karin
Are Sham Acupuncture Interventions More Effective than (Other) Placebos? A Re-Analysis of Data from the Cochrane Review on Placebo Effects.
In: Forschende Komplementärmedizin, Nr. 5: S. 259-264
Background and Objective: A recent Cochrane review on placebo interventions for all kinds of conditions found that `physical placebos' (which included sham acupuncture) were associated with larger effects over no-treatment control groups than `pharmacological placebos'. We re-analyzed the data from this review to investigate whether effects associated with sham acupuncture differed from those of other `physical placebos'. Methods: All trials included in the Cochrane review as investigating `physical placebos' were classified as investigating either (sham) acupuncture or other physical placebos. The latter group was further subclassified into groups of similar interventions. Data from the Cochrane review were re-entered into the RevMan 5 software for meta-analysis. The primary analysis was a random-effects analysis of trials reporting continuous outcomes of trials that used either sham acupuncture or other physical placebos. Results: Out of a total of 61 trials which reported a continuous outcome measure, 19 compared sham acupuncture and 42 compared other physical placebos with a no-treatment control group. The trials re-analyzed were highly heterogeneous regarding patients, interventions and outcomes measured. The pooled standardized mean difference was -0.41 (95% confidence interval -0.56, -0.24) between sham acupuncture and no treatment and -0.26 (95% CI -0.37, -0.15) between other physical placebos and no treatment (p value for subgroup differences = 0.007). Significant differences were also observed between subgroups of other physical placebos. Conclusion: Due to the heterogeneity of the trials included and the indirect comparison our results must be interpreted with caution. Still, they suggest that sham acupuncture interventions might, on average, be associated with larger effects than pharmacological and other physical placebos.