I thought I could trust the American Academy of Family Physicians (AAFP) to recommend the best medicine available. My family practice residency followed their philosophy. By learning what the AAFP taught I was able to pass the board certification exam with flying colors. Their flagship journal American Family Physician rates its recommendations by strength of evidence.
As I learned about alternative medicine, the many pitfalls of evidence-based medicine, and the need for science-based medicine, I began to suspect the AAFP of a less-than-rigorous approach. On several occasions, I wrote letters to the editor of American Family Physician, but they were not receptive to my arguments.
Imagine my shock when the AAFP’s continuing medical education program FP Essentials featured a monograph on “Integrative Medicine”! According to Mark Crislip, “If you integrate fantasy with reality, you do not instantiate reality. If you mix cow pie with apple pie, it does not make the cow pie taste better; it makes the apple pie worse.” Readers of SBM will be aware that integrative medicine is a marketing term, not a scientific term. It is designed to allow practitioners to provide treatments that are not supported by science, to facilitate the infiltration of so-called “quackademic” medicine into our hospitals and medical schools.
The monograph begins with a summary of “Key Practice Recommendations” rated by “Strength of Recommendation Taxonomy (SORT).” Not a single recommendation earned evidence rating A – “based on consistent and good quality patient-oriented evidence.” They were all rated SORT B and C, based on inconsistent or limited-quality evidence, consensus, usual practice, opinion, or case series. I don’t want to know what someone’s opinion is; I want to know what the evidence shows.
They claim that manipulative therapy “may improve pain and function in the short and moderate term in patients with acute and chronic back pain, neck pain, rotator cuff conditions, and temporomandibular joint disorders.” The research they cite for most of these indications is far from definitive. They say, “manual therapies have been used to manage infantile colic.” Yes, they have been used for colic, but they are not effective. They offer evidence of effectiveness from a Cochrane review, but mention that in studies where parents were blinded, the difference was not significant, suggesting that “benefits may not have been attributable to the therapies.” Simon Singh was famously sued by British chiropractors for saying chiropractors supported bogus treatments, one of which was manipulation for colic. The General Chiropractic Council subsequently issued a report saying “the evidence does not support claims that chiropractic treatment is effective for childhood colic, bed-wetting, ear infections or asthma.” The suit was withdrawn.
The monograph at least acknowledges that neck manipulation carries risks, saying “an estimated 1 in 20,000 neck manipulations result in vertebral artery dissection or aneurysm with potential stroke.” I question whether that is an acceptable risk, considering that study results varied, and considering that neck manipulation is only effective for neck pain when combined with exercise.
Two of the four authors of the monograph are credentialled medical acupuncturists. Isn’t that a source of bias, if not a conflict of interest? Of course, they are not going to agree with Colquhoun and Novella that acupuncture is a theatrical placebo. They say there is strong evidence supporting the effectiveness of acupuncture for allergic rhinitis, chemotherapy-induced nausea and vomiting, chronic low back pain, headache, knee osteoarthritis, migraine prophylaxis, postoperative nausea and vomiting, and postoperative pain. They say there is moderate evidence supporting the effectiveness of acupuncture for 37 other conditions, including obesity, constipation, dry eye, allergies, female infertility, and asthma. I think most non-acupuncturist providers would disagree that the evidence is strong or moderate. They don’t mention Edzard Ernst’s systematic review of systematic reviews of acupuncture for pain.
They found “evidence” for dry needling, laser acupuncture, acupressure, and auricular acupuncture. They offer mechanisms of action, but these are mostly speculative or based on animal studies. Before trying to explain how something works, it is necessary to prove that it does work.
Pitfalls in evidence
In addition to the sections on manual therapy and acupuncture, the monograph includes sections on herbal supplements and cannabis/cannabis-related drugs. I won’t go into detail on those sections. There is a pattern throughout the monograph of accepting evidence that would not be acceptable in rigorously science-based medicine. One recurring problem is the acceptance of studies that compare an alternative treatment to usual care or no care. That’s a recipe for finding false positive results.
They are very adept at finding studies that support their beliefs, but I wish they were equally adept at searching out disconfirming evidence. I wish they would adopt my “SkepDoc’s rule” – before accepting a claim, try to find out who disagrees with it and try to understand why. I wish they would read the Science-Based-Medicine blog and absorb its lessons.
Not all bad
I’m trying to see the good side of this. Alternative medicine is very popular at present. Patients have heard good things about acupuncture, supplements, etc. They want to try them and want more information. If they are consulting an integrative medicine doctor, that means they are not going to someone who only does alternative medicine. It could be worse. The monograph is very clear about some treatments that have no evidence for efficacy, such as evening primrose for hyperlipidemia. For neck manipulation, they point out the risk of stroke. For herbal remedies, they cover side effects, drug interactions, contaminants, poor regulation, and poor quality control. They do not recommend or even discuss homeopathy, energy medicine, therapeutic touch, drinking urine, reflexology, iridology, or many other totally quacky treatments. An integrative medicine doctor may be able to exercise some control over the number of useless treatments their patients are willing to try. Many of their patients will perceive improvement from useless treatments due to placebo effects. That may be what they want, but shouldn’t it be the duty of the doctor to educate them and give them what they need?
Conclusion: The AAFP’s embrace of integrative medicine is unfortunate, but not unexpected
I can no longer trust them or respect their judgment. It is distressing to learn that my former idols have feet of clay. I respect the right of patients to try anything they want to try, but I can hope they have access to accurate scientific information and can exercise true informed consent. I’m not going to waste any more time complaining to the AAFP, because I’m convinced they are not prepared to listen. But I can hope that by writing about the true state of the evidence on SBM I will be able to reach at least a few people who will be less likely to start down that slippery slope.
This article was originally published in the Science-Based Medicine Blog.