Medicare will cover acupuncture treatments under a proposal announced by the Trump administration on July 15, but only for Medicare beneficiaries with chronic low back pain enrolled in research to determine whether the treatment decreases their pain, improves function, or decreases the use of other medical treatments, such as opioids. According to Health and Human Services Secretary Alex Azar, the proposal is part of an effort to address the opioid addiction epidemic by providing patients with “new options while expanding our scientific understanding of alternative approaches to pain”.

This decision is part of a larger effort to decide whether acupuncture should be covered for all seniors insured by Medicare:

CMS [Centers for Medicare & Medicaid Services] is opening this national coverage analysis (NCA) to complete a thorough review of the evidence to determine if acupuncture for CLBP is reasonable and necessary under the Medicare program.

CMS says it reviewed the literature on the effectiveness of acupuncture for chronic low back pain, as well as guidelines and professional society recommendations. Unfortunately, these included some unreliable sources on acupuncture, such as the Veterans Administration and the American College of Physicians guidelines. CMS decided that

While we believe the evidence is promising, we do not believe the evidence is sufficient at this time for representative older patients with chronic nonspecific low back pain. We believe additional research would be needed in order to make a favorable coverage determination . . .

Hence, the decision to help fund that research.

Any fair review of the evidence for acupuncture answers the question the government is about to spend taxpayer money asking all over again: We already know that acupuncture is nothing more than an elaborate placebo and that it does not work, for anything. Thus, this effort is not justifiable based on any real need to do more research. Here’s what SBM’s Steve Novella, MD, had to say in a Washington Post story on the subject:

“The whole thing is a big scam,” said Steven Novella, an assistant professor of neurology at the Yale School of Medicine and editor of the “Science-Based Medicine” website. “The only honest interpretation of the data is that acupuncture is a theatrical placebo.” . . .

Novella said that the efficacy attributed to acupuncture in the 2018 review [referred to in the story] could easily be explained by various research biases and that no drug would be allowed on the market based on that level of proof.
“We never get that threshold of evidence that you need in medicine, where you get that persistent effect, and it’s replicable” across numerous studies, he said.

Exactly. The only explanation in my mind, then, is political: It is another success in the relentless campaign by CAM proponents to promote pseudoscience as the best thing since sliced bread for pain relief and therefore the cure for America’s opioid crisis. And, given the state of the evidence that prompted CMS’s proposal, I fear that similarly weak evidence produced by this research could prompt national coverage.

There are reasons other than wasting taxpayer money to oppose this proposal. CMS should not be considering Medicare coverage of acupuncture at all because of what I’ve referred to as integrative medicine’s “collateral damage”. If CMS were to decide to cover acupuncture outside of research, that necessarily means patients would be seeing licensed acupuncturists in private office settings, giving acupuncturists the opportunity to sell patients on the vast array of pseudoscience covered by the acupuncturist’s scope of practice. (In fact, as we shall see, given the nature of the clinical trials under consideration, this could well occur within the context of research covered by CMS’s current proposal.)

For example, depending on their scope of practice, licensed acupuncturists have broad authority to diagnose and treat any disease or condition with, not just acupuncture, but also bleeding, reflexology, Kirlian photography, thermograpy, homeopathy, gua sha, Chinese herbs (basically, unrefined and unproven drugs), magnets, reiki, therapeutic touch, cranial sacral therapy, Oriental medicine injection therapy (including caffeine and other substances), polarity therapy, dietary supplements, cupping, and, in New Hampshire, “insertion and retention, for days or weeks, of intradermal needles or press tacks in acupuncture points”. One study that reviewed acupuncturists’ websites showed that a majority advertised treatments for allergies and asthma that were unsupported by evidence. About 92 percent of seniors have at least one chronic disease and 77 percent have at least two, making them prime targets for cross-selling. Conditions like dementia, cognitive decline, and other mental health issues make older adults even more vulnerable.

CMS may also want to consider Florida’s experience in covering acupuncture for motor vehicle accident injuries. (Although, interestingly, cost-effectiveness analysis is apparently not a part of determining Medicare coverage.) A 2010 study by the Department of Insurance found that acupuncture was the second most expensive average provider charge per claim, clocking in at $3,674, more per claim than ER doctors, general practitioners, and orthopedic surgeons. The legislature responded by eliminating acupuncture coverage from no-fault auto insurance.

Medicare-supported studies

The 47-page “Proposed Decision Memo for Acupuncture for Chronic Low Back Pain” describes the requirements for the acupuncture research, which must be performed in clinical trials supported by the NIH or “CMS approved studies meeting AHRQ [Agency for Healthcare Research and Quality] criteria.” Of note, the NIH has already announced a “large-scale pragmatic trial to evaluate the impact of, and strategies to best implement, acupuncture treatment of older adults (65 years and older) with chronic low back pain.” Interesting language there: Why a “pragmatic trial” of a treatment with, at best, highly dubious evidence of effectiveness? Why “impact of” – what does that even mean in a research context? And why announce you are going to study “strategies to best implement acupuncture treatment” before you know the results of the “impact” studies, such as they are? Also of note: the NIH will not fund studies that propose a sham control. Why?

Here’s why, as explained by SBM’s own David Gorski, MD, PhD, himself an NIH-funded researcher:

Whenever you hear the words “pilot” and “pragmatic” in a description of a study design for an intervention designed to treat a subjective complaint like pain, you know it’s guaranteed to show a positive result, thanks to placebo effects. Acupuncturists love pragmatic studies for this very reason. It is true that pragmatic studies are designed to test interventions as used in the “real world,” but there is an assumption behind them, namely that the treatment being tested has already been shown to be efficacious in well-designed randomized controlled clinical trials. Indeed, for real interventions, frequently what is found in pragmatic trials is that the treatment doesn’t work as well in the “real world” as it did in clinical trials, usually because of the variability and lack of consistency that inevitably arise when treatments are released “into the wild.” In the case of treatments that don’t work better than placebo but do induce placebo effects, pragmatic trials almost always show more efficacy than well-designed randomized trials. Again, that’s the point.

CMS’s proposed Medicare coverage for research studies requires:

A minimum 12-week acupuncture intervention versus usual care or other intervention for chronic low back pain. Endpoints must be measured at 12 weeks, 6 months, and 12 months . . . with comparison to usual care, or other planned comparator arm.

In other words, CMS’s research will suffer from the same deficiencies as the NIH’s pragmatic trial.

In keeping with Medicare coverage requirements under the Social Security Act, acupuncture can be furnished in these studies only by physicians, or by physician assistants, nurse practitioners, and “auxiliary personnel,” all of whom (save physicians) must be licensed acupuncturists, have a graduate degree in acupuncture or Oriental Medicine from a school accredited by the Accreditation Commission on Acupuncture and Oriental Medicine, and certified by the National Certification Commission for Acupuncture and Oriental Medicine, both fox-guarding-the-henhouse organizations that, respectively, accredit acupuncture schools and test their graduates. “Auxiliary personnel” (presumably, acupuncturists who are not PAs or nurses) must be under the direct supervision of a physician, PA, NP, or clinical nurse specialist.

Thus, while physicians, depending on state licensing requirements, may have taken a “medical acupuncture” course to learn acupuncture, all others will be fully steeped in the mysticism and pseudoscience that comes with an acupuncture degree.

In its discussion of acupuncture, CMS does not acknowledge the recent political origins of TCM and contemporary forms of acupuncture, although it does (sort of) admit that it’s not really “thousands of years” old. While stating that it found insufficient evidence for covering acupuncture, CMS then turns around and declares that “Western medical acupuncture” employs the “principles of evidence based medicine”. It also dutifully marches through surrogate outcomes that are widely claimed as explaining acupuncture’s mechanism of action; e.g., endogenous opioids and release of adenosine. And, perhaps without realizing that it is doing so, CMS admits that “acupuncture” can mean whatever the practitioner wants it to mean.

[T]here is a diversity of theoretical models and techniques that are all described as acupuncture . . . [including] needling at specifically chosen points . . . [and] “non-needling” techniques focused on these points. Modern medical acupuncturists choose anatomically and physiologically important treatment points which many include both traditional acupuncture points and other non-traditional fixed points. . . . [T]here are several variations to traditional acupuncture including shallow needling, intradermal needling or intramuscular needling . . . twirling, plucking or thrusting [the] acupuncture needles. There are numerous variations of manually or electrically stimulated “needling” techniques, as well as multiple “non-needling” acupuncture techniques.

All of which raises the question: if a study uses one method of selecting acupuncture points and one method of needling (or non-needling), how generalizable are its results to other methods? And, if Medicare is to cover acupuncture nationally, how could it ever describe just which “acupuncture” it covers? Or, will it cover all acupuncture, despite the negative studies of particular methods? How would that be justifiable as “reasonable and necessary to the diagnosis and treatment of illness or injury,” the legal standard for Medicare coverage?

Finally, CMS’s criteria for studies include:

  • “The rationale for the study is well supported by available scientific and medical evidence.”
  • “The research study is in compliance with all applicable Federal regulations concerning the protection of human subjects. . .” [which requires studies to be based on sound science].
  • “All aspects of the study are conducted according to appropriate standards of scientific integrity.”

The implausibility of acupuncture (hand-waving about surrogate measures aside) should (but won’t) disqualify any study of acupuncture if these criteria are enforced.

CMS will accept comments on the proposal until August 14, 2019, and is “particularly interested in comments that address current studies of acupuncture (randomized controlled trials, observational trials, and registries) in the Medicare population.” (Thus, it would be especially helpful if those with a background in medicine or research would comment.) Many SBM readers complain about the infiltration of CAM; here’s the chance to do something about it. When CMS previously floated the idea of Medicare coverage for acupuncture, proponents came out in droves. Only “a few commenters did not support CMS coverage.” Let’s do better this time.



  • Jann J. Bellamy is a Florida attorney and lives in Tallahassee. She is one of the founders and Board members of the Society for Science-Based Medicine (SfSBM) dedicated to providing accurate information about CAM and advocating for state and federal laws that incorporate a science-based standard for all health care practitioners. She tracks state and federal bills that would allow pseudoscience in health care for the SfSBM website.  Her posts are archived here.    

Posted by Jann Bellamy

Jann J. Bellamy is a Florida attorney and lives in Tallahassee. She is one of the founders and Board members of the Society for Science-Based Medicine (SfSBM) dedicated to providing accurate information about CAM and advocating for state and federal laws that incorporate a science-based standard for all health care practitioners. She tracks state and federal bills that would allow pseudoscience in health care for the SfSBM website.  Her posts are archived here.