Historical perspectives on using sham acupuncture in acupuncture clinical trials

This important historical review article examines how the inadequate development of sham controls in acupuncture clinical trials may have compromised the quality of scientific evidence on this therapy. The authors, led by Stephen Birch, identified a fundamental problem: acupuncture clinical trials in the West began in the 1970s, long before detailed clinical literature was available in English on the different acupuncture practices.

The historical analysis reveals that the first detailed clinical books on acupuncture in English only appeared in the mid-1980s, with knowledge of the international diversity of practices emerging only in the 1990s. This temporal gap created a significant problem: researchers developed and implemented sham control techniques based on inadequate assumptions about how acupuncture works, without fully understanding the variety of existing techniques.

The study documents the evolution of sham control methods, from early experiments with needling at 'irrelevant' points to the development of 'minimal acupuncture' (superficial needling without sensory stimulation) and, later, non-penetrating devices. A critical finding was that minimal acupuncture was adopted as a sham control without considering that superficial needling without sensation was already an established therapeutic technique in Japanese acupuncture.

The authors identify serious methodological flaws in the development of sham controls. First, the validation of these controls focused only on credibility and blinding capacity, ignoring physiological studies that could determine whether these techniques are truly inert. Second, no adequate pilot studies were conducted to test whether the sham controls had therapeutic effects of their own.

A particularly problematic finding was that the developers of non-penetrating sham devices themselves inadvertently demonstrated, through their own trials, that these techniques should not be used as controls. In some cases, the sham control was more effective than 'real' acupuncture, suggesting that gentler stimulation may be more appropriate for certain conditions.

The article argues that these methodological problems introduce bias against acupuncture, potentially underestimating its efficacy. The authors suggest that many trials concluding that acupuncture is no more effective than placebo may have used controls that actually had therapeutic effects, making the comparison invalid.

The clinical implications are significant. Systematic reviews and meta-analyses that base their conclusions on trials with inadequate sham controls may be providing misleading evidence about the efficacy of acupuncture. This affects not only future research, but also clinical and health policy decisions based on this evidence.

The study has some limitations, including its narrative nature and the possible selection of literature based on the authors' experience. Nevertheless, the analysis is comprehensive and well-documented, providing a valuable historical perspective on persistent methodological problems in acupuncture research. The authors call for a fundamental reassessment of how acupuncture trials are designed and interpreted, suggesting the need for new methodological standards that adequately consider the diversity of acupuncture practices and develop truly appropriate controls.