Background and objectives: For non-drug interventions such as acupuncture, it is difficult to establish placebo or sham controls that are both inert and indistinguishable. We reviewed sham-controlled clinical trials of acupuncture to investigate (a) which types of sham interventions have been used in the past; (b) in what respects true and sham interventions differed; and (c) whether trials using different types of sham yielded different results.
Methods: 47 randomized controlled trials comparing true and sham acupuncture interventions for pain and a variety of other conditions were identified from systematic reviews and through a search in PubMed. Details of patients, interventions, sham interventions and outcomes were extracted in a standardized manner.
Results: In two trials the sham intervention consisted of superficial needling of the true acupuncture points, four trials used true acupuncture points which were not indicated for the condition being treated, in 27 trials needles were inserted outside true acupuncture points, five trials used placebo needles and nine trials used pseudo-interventions such as switched off-laser acupuncture devices. True and sham interventions often differed in a variety of other variables, such as manipulation of needles, depth of insertion, achievement of an irradiating needling sensation (de-chi), etc. There was no clear association between the type of sham intervention used and the results of the trials.
Conclusion: Randomized trials investigating the specific effects of acupuncture have used a great variety of sham interventions as controls. Summarizing all the different sham interventions as "placebo" controls seems misleading and scientifically unacceptable.