There are many levels at which one can, and should, evaluate a proposed medical treatment. Certainly, high level and high quality clinical trial research is likely to be the deciding factor for many interventions. But another important thing to look at, especially when interpreting mixed results or generally poor quality clinical trials, is the underlying mechanism proposed for a therapy. If it contradicts well-established science or cannot be convincingly demonstrated to be correct, this raises the bar even higher for clinical research evidence.
When it comes to acupuncture, the clinical research evidence is vast and highly variable in terms of quality and results. Overall, the preponderance of the evidence seems to support the conclusion that acupuncture is not superior to placebo treatment and exerts very broad, non-specific physiological effects indistinguishable from other kinds of mild traumatic or irritating stimulation (e.g. 1, 2, 3).
One way to help decide whether extensive further efforts to sort out potential clinical benefits of acupuncture through expensive and laborious clinical trials is to look at the potential justifications for why the practice might or might not be helpful. Obviously, anecdotes and testimonials are the primary reason people believe in the usefulness of acupuncture despite the lack of strong support from clinical research. However, these are only useful in generating hypotheses in medicine, not in proving or disproving them. Every therapy every used has been able to claim anecdotal evidence of an effect, and since it’s clear that many therapies turn out not to actually work when objectively studied, reliance on anecdotal evidence is a big mistake.
There are a couple of ways to approach the theoretical explanation of possible mechanisms for acupuncture. The traditional explanations are purely vitalistic. Mystical life energy (Ch’i) that is undetectable by scientific means, explanations about balancing non-physical, spiritual concepts such as Yin and Yang, metaphorical explanations involving Heat, Cold, Damp, Wind and so on are all part of the theoretical structure many practitioners use to explain and justify so-called Traditional Chinese Medicine, including acupuncture. These explanations are part religion, part folklore, and no scientific examination suggests they have any physical reality or any relevance to health and disease as it is understood in scientific medicine. Employing such explanations makes one’s practice essentially a religious rather than a medical one.
However, other acupuncture advocates claim a solidly scientific explanation for the proposed effects of acupuncture, involving identifiable anatomic structures, such as nerves, biochemical effects involving substances naturally produced in the body, and so on. This is a more promising theoretical approach, but despite some effort, there is not yet a consistent and convincing body of evidence to validate the clinical practice of inserting needles in specific locations (with or without twirling them, burning herbs on them, passing electrical current through them, and all the other varieties of acupuncture seen) in order to effect health. Putting needles in animals undoubtedly has some measurable physiological effects. But whether these are specific, predictable, controllable effects that have a meaningful impact on health is not established.
One question that has been investigated many times is whether the supposed acupuncture points, specific locations on the body where insertion of needles is proposed to have a particular effect, can be consistently identified by acupuncturists or associated with identifiable anatomic structures. Even among those who believe acupuncture to be clinically useful, there is controversy about whether specific locations for needle insertion are real or necessary (e.g. 3), and of course some acupuncturists argue that acupuncture points do not need to correspond to identifiable anatomic or functional locations since they follow the vitalist model in which the focus of acupuncture is on the Ch’i rather than the body per se.
There are some scientific papers that purport to identify real structures or other measurable characteristics at some traditionally used acupuncture points. Many of these reports come from China, where negative studies about acupuncture (or other alternative therapies) are almost never published, so the potential for bias in this source has to be considered. A review of the literature presented at the 2000 meeting of the American Association of Equine Practitioners by veterinarian David Ramey found no consistent body of scientific evidence to support the notion of specific identifiable functional or anatomic acupuncture points. A recent paper seems to add support to this conclusion.
A.F. Molsberger, J. Manickavasagan, H.H. Abholz, W.B. Maixner, H.G. Endres. Acupuncture points are large fields: The fuzziness of acupuncture point localization by doctors in practice. Eur J Pain. 2012 Apr 10 [Epub ahead of print]
This study took the approach of testing whether properly trained and experienced acupuncturists could consistently identify commonly used acupuncture points on a single patient. Twenty-three common points were selected and identified by the 71 test subjects. The area within which specific points were identified by 95% of the acupuncturists ranged from 2.7cm in diameter to 41.4cm in diameter. Because of the variability with which experienced acupuncturists identified common acupuncture points, the authors concluded that to stimulate fake points as a placebo control for a clinical trial, it would be best to stimulate the skin at least 6cm away from the spot identified as the real point on the face or hands and at least 12cm away from a proposed real spot anywhere else on the body.
Of course, the authors are acupuncturists and so must find a way to interpret these results that is supportive of acupuncture as a therapy. Their conclusion is that it would be more meaningful to talk about “acupuncture fields” rather than acupuncture “points.” Since the variance in the size of the area within which supposed points were identified was not associated with the experience or training of the doctor, and could not have been associated with individual differences between patients since the same patient was examined by all the doctors, it seems more reasonable to interpret the results as indicating that no specific locations associated with traditional acupuncture points can be reliably identified even by trained acupuncturists. Likely, such points don’t exist as anatomical or functional entities, especially given the lack of convincing and consistent evidence for their existence despite decades of study.
This by itself does not invalidate acupuncture as a clinical practice. However, it has bearing on the study and evaluation of acupuncture. Many studies using supposed “sham” acupuncture points may now be reinterpreted. Acupuncturists will likely interpret these results to mean that the lack of difference between effects in groups where “real” acupuncture points were used and those in which “sham” points were used is actually due to the fact that both groups benefitted from stimulation in the same “acupuncture fields.” Skeptics like myself are more likely to view this as indicating that “sham” acupuncture points are not a legitimate placebo control since specific acupuncture points can’t even be shown to exist, and so trials comparing real and fake points are simply poorly controlled. In any case, it suggests that if the central tenet of most acupuncture treatment, stimulation of the body in specific locations, is invalid because even acupuncturists don’t know where these locations are, then the concept of acupuncture as a meaningful therapy is even less convincing.