Does the choice of placebo determine the results of clinical studies on acupuncture? A meta-analysis of 100 clinical trials

Objective: To establish whether the choice of the placebo treatment used may influence the outcomes of clinical trials on acupuncture or not. Design: A meta-analysis of outcomes according to the choice of the placebo. Attention was focused on the placebo design of 117 clinical, controlled trials found after an extensive search. Studies comparing acupuncture to no treatment or a reference treatment were discarded from the analysis. A set of 90 publications could be classified into one of: two groups: i) Clinical studies with sham acupuncture as placebo, which consists of needling outside the meridian, but near to classical acupoints. This group of 45 trials was classified as energetic placebo model (EPM). ii) 45 studies using a placebo treatment consisting of needling within a segmental zone far enough away from the active points were classified as neurophysiological or metameric placebo model (MPM). In both groups of studies the proportions of significant results and the distribution of outcomes characterized by nonsignificant results with improvements greater than 35% in both groups of patients were assessed by the chi-square test. Results: The proportion of meaningful results was significantly higher in the MPM group [73.33% (33/45)], while only 33.33% (15/45) of such results were found in the EPM group (p < 0.03). In the EPM group 24/30 studies showed nonsignificant results with improvements greater than 35% in both groups of patients, while in the MPM group only 20% (6/30) of studies with this outcome could be observed (p < 0.05). Conclusion: Studies using EPM as placebo failed more frequently to show any differences between real acupuncture and placebo treatment than those using MPM as placebo. On the other hand, sham acupuncture appears almost as active as 'real' acupuncture. These results suggest that the design and the way of performing the placebo procedure determine the outcome, i.e. success or failure of a clinical trial in obtaining differences among the patients groups, in case they actually exist.