The largest and most rigorous study to date, employing a meticulous data meta-analysis, provides evidence that acupuncture is more than just an elaborate placebo for treating chronic pain. However, placebo effects do appear to play a considerable role and the contribution of the type and quality of acupuncture itself may be of relatively minor consequence. So, the debate over the value of acupuncture in chronic pain management seems unsettled.
Writing in an advance online edition of the Archives of Internal Medicine, Andrew J. Vickers, DPhil, of Memorial Sloan-Kettering Cancer Center, New York, and colleagues used data from previously published randomized controlled trials (RCTs) encompassing nearly 18,000 patients from the United States, United Kingdom, Germany, Spain, and Sweden [Vickers et al. 2012]. Going beyond the usual rigors of meta-analysis, the authors acquired individual patient data from 29 high quality randomized controlled trials (RCTs) comparing acupuncture with either sham (placebo) acupuncture or no acupunture (ie, usual care) in 3 chronic pain conditions: back and neck (musculoskeletal) pain, osteoarthritis, and chronic headache.
Patients in all RCTs had access to analgesics and other standard treatments for their pain, so either true acupuncture or sham acupuncture was an add-on therapy. Sham acupuncture, representing a placebo-control condition, included needles inserted superficially or at non-acupuncture points, devices with needles that retracted into the handle rather than penetrating the skin, or non-needle approaches such as deactivated electrical stimulation or detuned laser.
Results of this study, funded by the U.S. Natonal Center for Complementary and Alternative Medicine (NCCAM), suggest beneficial effects of accupuncture. After adjustments to eliminate biases unduly favoring acupuncture, the following standardized effect sizes were reported: [understanding effect sizes in pain research was discussed in an UPDATE here]
- Patients receiving acupuncture had less pain, with effect sizes of 0.23 (95% CI, 0.13-0.33), 0.16 (95% CI, 0.07-0.25), and 0.15 (95% CI, 0.07-0.24) Standard Deviations (SD) in comparison with sham acupuncture controls for back and neck (musculoskeletal) pain, osteoarthritis, and chronic headache, respectively.
- Standardized effect sizes for acupuncture in comparison to no acupuncture controls (ie, usual care) were 0.55 (95% CI, 0.51-0.58), 0.57 (95% CI, 0.50-0.64), and 0.42 (95% CI, 0.37-0.46) SD for back and neck (musculoskeletal) pain, osteoarthritis, and chronic headache, respectively.
These results were robust to a variety of sensitivity analyses, including those accounting for possible publication bias and weaknesses in some of the included trials. The statistically significant effects of true versus sham acupuncture in the first analysis above indicate that acupuncture is more than merely a placebo; however, the researchers concede that the effect sizes, representing differences between the two approaches, are relatively small.
The authors conclude that factors in addition to the specific effects of needling are important contributors to the therapeutic efficacy of acupuncture. Additionally, even though effect sizes for acupuncture compared with usual care in relieving pain are only moderate, the researchers advise that acupuncture is effective for the treatment of chronic pain and is therefore a “reasonable referral option.”
COMMENTARY: According to background information in the Vickers et al. article, an estimated 3 million adults in the United States alone receive acupuncture treatment each year. Yet, they add that the lack of an accepted biological mechanism to explain the physiological effects of this therapy, and its provenance in theories and practices outside of traditional medicine, still make acupuncture highly controversial.
In a commentary accompanying the Vickers et al. article, Andrew L. Avins, MD, MPH — of Kaiser-Permanente, Northern California Division of Research, Oakland — notes that the debate over acupuncture as a bona fide therapy for pain is ongoing [Avins 2012]. To date, at least 60 meta-analyses addressing unresolved questions surrounding acupuncture have been conducted without reaching definitive answers.
Research and arguments on both sides of the debate have been presented in various Pain-Topics UPDATES [here]. In the most recent UPDATE on this topic [here], Edzard Ernst, MD, PhD and others discuss broad-scale reviews of the scientific literature on acupuncture, finding that evidence in its favor for a variety of pain conditions is equivocal, inconsistent, or inconclusive in many cases.
This meta-analysis by Vickers et al. was indisputably thorough and rigorous, involving dozens of scientists from multiple countries over a period of years. Their extensive search uncovered 955 studies for consideration, attesting to the extent and diversity of research on acupuncture, but also to the inferior quality of so much of the research. The research team identified only 31 high quality studies meeting their stringent inclusion criteria and were able to obtain individual data records for 29 of those to perform their very exacting meta-analyses. That is, they did their own calculations of outcome results, not relying on those of the original study authors.
The findings of the meta-analysis are scientifically and clinically important; however, the authors acknowledge the following regarding the effects of acupuncture found in their data:
“…an important part of these total effects is not due to issues considered to be crucial by most acupuncturists, such as the correct location of points and depth of needling. Several lines of argument suggest that acupuncture (whether real or sham) is associated with more potent placebo or context effects than other interventions. Yet, many clinicians would feel uncomfortable in providing or referring patients to acupuncture if it were merely a potent placebo. Similarly, it is questionable whether national or private health insurance should reimburse therapies that do not have specific effects.
Our finding that acupuncture has effects over and above those of sham acupuncture is therefore of major importance for clinical practice. Even though on average these effects are small, the clinical decision made by physicians and patients is not between true and sham acupuncture but between a referral to an acupuncturist or avoiding such a referral. The total effects of acupuncture, as experienced by the patient in routine practice, include both the specific effects associated with correct needle insertion according to acupuncture theory, nonspecific physiologic effects of needling, and nonspecific psychological (placebo) effects related to the patient's belief that treatment will be effective.”
Also, it should be emphasized, as noted earlier, that acupuncture did not exert effects in isolation from “usual medical care,” since in all cases acupuncture (or sham acupuncture) was add-on therapy. So, effects of acupuncture, though modest, were beyond and above those typically achievable by pharmacotherapy or other traditional treatments for chronic pain.
However, it is difficult to assess just what those added benefits of acupuncture might be and their clinical impact. Effect sizes were presented in the meta-analysis as standard deviation units (essentially Cohen’s d scores). For example, the favorable effect=0.57 SD of acupuncture for osteoarthritis would denote an improvement of roughly a 1/2 standard deviation on the end-point measurement scale used to assess pain, functionality, or something else.
Vickers and colleagues note that, in deriving effect sizes, their meta-analysis combined different end points, such as pain and function, measured at different times. Despite this variation, they claim that their results were stable; for example, effect sizes generally did not change when analyses were restricted to only pain end points and measured at specific follow-up times (eg, 2 to 3 months after randomization). However, as Avins notes in his commentary, these standardized effect sizes can be problematic when trying to assign absolute measures of improvement.
Vickers et al. believe that their overall observed estimate of roughly 0.50 SD favoring acupuncture over usual care is of clear clinical importance, but Avins observes that this is difficult to substantiate. The clinical relevance of this would vary with the outcome being assessed (eg, pain, functionality, mood, etc.), how it is being measured, and the standard deviation. For example, given a hypothetical average baseline osteoarthritis pain score of 60mm on a 0-to-100mm VAS and a standard deviation of 20mm, acupuncture might exert an 18% improvement of about 11mm (0.57 effect size X 20mm) and patients would still have moderate pain measuring 49mm on the VAS. A recent UPDATE [here] discussed research finding that 12mm denotes the Minimum Clinically Significant Difference (MCSD) in a VAS score that is perceived by patients as being of any consequence. So, from a patient’s perspective, would this sort of benefit be sufficient to justify the time, inconvenience, and expense of acupuncture?
Another disquieting aspect is the admitted placebo aspect of acupuncture. Although Avins argues in his commentary that placebo-based mechanisms have a place in pain management practice, he also recognizes that in traditional medicine new therapies must clearly show superiority over placebo to gain regulatory approval. In the Vickers et al. meta-analysis, acupuncture compared with sham/placebo demonstrated statistically significant improvements but only small effect sizes (0.15 to 0.23 SD) that might not be considered of clinical significance. If acupuncture were judged according to the same stringent criteria as a pharmacotherapy would it be approved by regulatory agencies?
Unfortunately, the Vickers et al. meta-analysis does not examine sham/placebo acupuncture compared with usual care, so the differential effects attributable to acupuncture alone, devoid of the placebo component, cannot be calculated. Still, Avins argues for an appreciation of placebo effects, writing…
“Perhaps the recognition that some patients find benefit in CAM therapies (many of which may, indeed, operate primarily through placebo mechanisms) should force us to examine our perceptions of placebo effects and question why so many of us feel threatened by their existence. …. At the end of the day, our patients seek our help to feel better and lead longer and more enjoyable lives. It's ideal to understand the mechanism of action, which carries the potential for developing more and better interventions. But the ultimate question is: does this intervention work (or, more completely, do its benefits outweigh its risks and justify its cost)?”
Through their meta-analysis, Vickers and colleagues reveal evidence that benefits of acupuncture, while modest, may provide advantages going beyond usual care for patients with diverse types of chronic pain. At the same time, considerable therapeutic effects may be due to factors such as a patient’s belief that treatment will be helpful, as well as placebo and other context effects, while a much smaller acupuncture-specific component may or may not involve such issues as specific needling points and the type or depth of needling. All of this may be unsatisfactory to what Avins calls “the quack-hunter community” who continue to argue against the validity of CAM therapies that rely more on placebo or other nonspecific effects than validated biological mechanisms of action.
> Avins AL. Needling the Status Quo: Comment on “Acupuncture for Chronic Pain.” Arch Intern Med. 2012(Sep); 172(11):1-2 [abstract here].
> Vickers AJ, Cronin AM, Maschino AC, et al. Acupuncture for Chronic Pain: Individual Patient Data Meta-analysis. Arch Intern Med. 2012(Sep);172(11):1-10 [abstract here].
Don’t Miss Out. Stay Up-to-Date on Pain-Topics UPDATES!
Register [here] to receive a once-weekly e-Notification of new postings.