Jann Bellamy recently wrote about the American Academy of Family Physicians (AAFP) and their policy on approved content for continuing medical education (CME) activities. The AAFP said, “given the current lack of evidence, the COCPD does not deem it appropriate to learn how to practice functional medicine or implement related techniques within an AAFP-certified CME activity.”
Moreover, they said:
The AAFP Credit System strives to certify CME that is evidence-based, objective and balanced, which means benefits have been proven and all possible risks, side effects and negative outcomes should be acknowledged. Ideally, CME includes fair, balanced consideration of all usual and customary approaches and not only the approach preferred by the CME provider.
Not practicing what it preaches
The AAFP speaks with forked tongue. Their own CME program is far from fair and balanced. It promotes acupuncture, dry needling, cupping, and referral to alternative practitioners.
For many years, the AAFP has offered a CME program called FP Essentials, for which they award 60 CME credits a year. Every month, they publish a monograph on a topic important to family physicians. I have subscribed for as long as I can remember, and I thought I could rely on it to keep me up-to-date with evidence-based information. I can no longer trust it. In their recent monograph on Musculoskeletal Therapies, there is a whole section (constituting a whopping 1/4 of the material presented) devoted to acupuncture, dry needling, and cupping! One-fourth of what they think family physicians should know about treating musculoskeletal conditions is non-evidence-based alternative medicine. This is not acceptable.
Their concept of evidence-based medicine is exactly the kind of thing that led us to establish the Science-Based Medicine blog. They are willing to accept any published studies at face value without considering prior plausibility and without recognizing the factors that so often lead to false conclusions in alternative medicine research.
They recommend acupuncture as first-line therapy for chronic back pain. They say there is evidence that it may benefit neck pain, lateral elbow pain, peripheral joint osteoarthritis, shoulder pain, and fibromyalgia. They say, “acupuncture consistently has been shown to be more effective than no treatment.” They seem not to understand that doing anything will appear more effective than no treatment, and that studies comparing acupuncture to no treatment are useless.
A truly fair, balanced consideration of all the evidence on acupuncture led David Colquhoun and Steven Novella to conclude that acupuncture is a theatrical placebo. What’s more, its effect on pain is questionable. Edzard Ernst et al. did a systematic review of systematic reviews of acupuncture for pain. They found 57 systematic reviews; the results were a mix of negative, positive, and inconclusive results. There were only four conditions for which more than one systematic review reached the same conclusions, and in only one of the four did they agree on a positive conclusion (neck pain). (Think about this: If acupuncture worked well for pain, wouldn’t the results be more consistent, and wouldn’t it work for pain in all parts of the body rather than just in the neck?) They explained how inconsistencies, biases, conflicting conclusions, and recent high quality studies throw doubt on even the most positive reviews.
They also demolished the “acupuncture is harmless” myth by reporting 95 published cases of serious adverse effects including infection, pneumothorax, and five deaths. Admittedly some but not all of these might have been avoided by better training in anatomy and infection control.
The AAFP monograph provides many references to acupuncture studies, but it is apparent that they are cherry-picking. Colquhoun and Novella’s articles and Edzard Ernst’s review of reviews are conspicuously absent.
Assuming acupuncture works and trying to explain how it works
The AAFP monograph describes several hypothetical mechanisms of action. It says it was originally believed to work by manipulating the life force “qi” at acupuncture points along meridians. It doesn’t question that explanation. It doesn’t point out that qi, acupoints and meridians are mythical concepts from a prescientific era. It says there is some evidence that acupoints have a discrete histologic architecture with a corresponding high electrical conductivity capacity. It also mentions the gate theory of pain, endogenous endorphin release (Note: this also occurs with placebos, and they acknowledge that it is an unsatisfactory explanation because the pain relief lasts longer than the endorphin half-life), and the connective tissue theory. It says, “The connective tissue theory uses the ancient concept of meridians” – with no comment about the validity of that concept. It claims that the neurotransmitter theory may explain the effects of acupuncture on conditions such as depression, anxiety, and substance abuse. (What effects? Systematic reviews are negative.) It mentions that “placebo effects also may be involved”, but It doesn’t acknowledge that placebo could be the sole mechanism and that all the published evidence is compatible with the hypothesis that acupuncture is a theatrical placebo.
Describing the published evidence
It describes the published evidence for acupuncture for low back pain, neck pain, lateral elbow pain, peripheral joint osteoarthritis, shoulder pain, and fibromyalgia. It admits that most of the trials it cites are low quality, insufficient evidence, inconclusive, sometimes negative, or not clinically significant. One wonders why it bothered to cite them.
It doesn’t differentiate between studies of traditional acupuncture and variants like electroacupuncture and ear acupuncture. It does mention that the fibromyalgia study was for electroacupuncture, and it concluded, “Thus, electroacupuncture should be considered for patients with fibromyalgia.” In fact, that recommendation is one of the four Key Practice Recommendations listed at the beginning of the monograph.
After covering acupuncture, the monograph goes on to discuss dry needling. Mark Crislip reviewed dry needling for us a couple of years ago. He stressed that it had nothing to do with acupuncture. He was not impressed by the published evidence; in fact, one of the systematic reviews he found showed that the placebo was more effective than the dry needling! The monograph chose to cite another, more positive review showing that dry needling was more effective than no treatment but not more effective than traditional management, with risk of bias and little evidence of long-term benefit. They call this “low- to moderate-quality evidence”. Whaaat?
The monograph says cupping produces hematomas (in other words, it deliberately injures the skin) and carries a risk of infection. Cupping is theorized to “increase circulation to resolve stagnant blood and lymph, which is thought to be a source of pathology”. (A source my pathology course in medical school failed to mention.) It describes the literature as “conflicting in terms of evidence of efficacy”. It acknowledges that good research is difficult due to the variety of practices and the impossibility of blinding. Since there is a lack of good evidence, why does it devote six paragraphs to cupping in a monograph about all musculoskeletal therapies? It could have just explained what cupping is and said there is no good evidence; two sentences would have been sufficient.
Issues to consider
Under this heading, it comments that patients who expect these “integrative” practices to work will report benefit even with sham procedures. It wonders whether skeptics will respond. It compares rates of acceptance by various groups (women, people of Asian descent, etc.) and reports the reasons patients give for choosing acupuncture: “The holistic and collaborative nature of treatment”. It mentions that acupuncture sessions can cost up to $200 and that insurance may not pay.
It provides guidelines for referring patients for integrative and alternative medicine therapies. It starts by saying there doesn’t appear to be increased liability for doctors to refer patients for these services. Think about that. Can you imagine commenting on liability in a discussion about referral to an orthopedic surgeon or other reputable practitioner? It says the referring doctor should ensure that the integrative practitioner is trained, licensed, and has malpractice insurance (!). Referral documents should include diagnosis, previous and ongoing treatments, and goals of therapy. The alternative treatment should have a favorable risk to benefit ratio compared to other treatments for the same conditions. It should be based on a reasonable expectation of favorable outcome and be based on the expectation that it will offer greater benefits than no treatment (but as I pointed out, any treatment will offer greater placebo benefits than no treatment). The thought process leading to the referral should be documented in the medical record, and the referring physician is required to monitor the results of treatment.
It describes how the physician can get training in these modalities. “Using the learned skill set will depend on the clinician’s practice environment.”
Conclusion: A betrayal
In response to Jann Bellamy’s article, one of our commenters said, “There are plenty of CAM methods that are safe, while also having enough supporters to have tentative credibility with the public. The public mostly doesn’t care too much about the science.” I’m sure this is true, but the AAFP should care about the science. And the AAFP’s members should care whether the AAFP maintains its reputation as a reliable source of the best science-based medical information.
Another commenter said, “Sugar pills and saline injections have also been consistently shown to be more effective than no treatment. Should we bring those back too?”
There is an obvious disconnect between what the AAFP says are its standards for CME and the CME it actually offers. I feel angry and betrayed by my own professional organization, an organization that I was once proud to belong to.