“Complementary and alternative medicine” (CAM) goes for the bandwagon fallacy again

ResearchBlogging.orgIt figures.

Right around the time of my blogiversary yesterday, when I had intended nothing more than a brief by characteristically self-indulgent bit of navel-gazing twaddle (at which, I succeeded brilliantly, I might add; no one–and I mean no one–does self-indulgent navel-gazing twaddle better than I do), what should be there tempting me from my intended day off? Lots of news stories about a report published by the Centers for Disease Control about a yearly survey regarding the usage of so-called complementary and alternative medicine (CAM) therapies in the U.S. Predictably, headlines in major news outlets proclaimed:

Wow, from theose headlines and stories, you might think that CAM usage is skyrocketing right here in the ol’ U. S. of A. You might think that the the 2007 National Health Statistics report on CAM use by Americans had found some massive sea change in American’s usage of unscientific medical modalities.

You’d think wrong.

I won’t go into detail about this, mainly because Steve Novella deconstructed this very report three months ago and then posted a followup yesterday. He looked at the numbers then; he looked at the numbers now. He found little change in CAM usage over the last 20 years. Yet, it’s very easy to spin the report as indicating that CAM is hugely popular.

But to what end?

If you’ve learned on thing from me reading this blog, I hope it’s that CAM, or “integrative medicine” (a.k.a. unscientific woo) is not a distinct category of medicine based on any rational or logical division, at least not as defined by its advocates. However, there is a logical division between CAM and “conventional medicine”; it’s just not the one that CAM advocates like. That’s because the logical, rational, and scientific difference between “conventional,” or science- and evidence-based medicine (otherwise known by its detractors as that evil “Western” medicine) and so-called CAM is that the former is based on science and evidence, and the latter is not. The former has considerable evidence to support its efficacy; the latter often doesn’t even have scientific plausibility going for it. (Homeopathy, anyone?) In other words, to boil it all down, the former is (mostly) proven to work; the latter is not, consisting of either therapies that are unproven or have actually been proven not to work any better than a placebo. Given that most, if not nearly all, CAM therapies are nothing more than elaborate placebo, one exception being herbal remedies, this is not surprising. Of course, this is not what CAM advocates want to hear, much less admit to be true. So, given the utter lack of scientific support for the vast majority of CAM modalities and the weak support for the remainder, what’s left if you’re a CAM advocate trying to prosletyize about how great CAM is?

Argumentum ad populum, of course. Otherwise known as the bandwagon fallacy or appeal to popularity, it is one of the most favorite logical fallacies of all. If CAM really were that popular, it would say nothing about its efficacy anyway. After all, what percentage of the population believes in astrology? That John Edward or Sylvia Browne can communicate with the dead? In alien abductions? In ghosts? Does the large number of people who believe in all these things mean that they must have some validity to them? Of course not. But that’s exactly what CAM believers would like you to conclude from surveys like this when in reality the apparent popularity of CAM says very little about whether CAM modalities work or not. Science does, and in general it hasn’t been kind in its verdict–and rightfully so, given the implausibility of modalities such as reiki, therapeutic touch, or homeopathy. The only CAM modalities that may have a chance of working, at least based on an assessment of current research, are herbal remedies, and they are in essence crude, contaminated drugs with variable strength from lot to lot.

It’s worse than that, though, as Steve points out. In order to inflate the numbers of people using CAM, the designers of this and other CAM usage studies have broadened the definition of CAM to be in essence meaningless and to incorporate modalities that, when used appropriately, are firmly within the science-based medicine camp. (When not used properly, apparently they’re CAM.) For instance, if you’ve ever had massage therapy, apparently you’re a CAM user, even though massage therapy is a perfectly legitimate modality for muscle soreness and relaxation. It only becomes woo when practitioners start to make bizarre claims for it. The same principle is in operation for modalities like yoga or Tai Chi, both of which are primarily exercise. Another example is chiropractic, which is nothing more than spinal manipulation and, assuming it’s done only for back pain. A good physical therapist can do the same thing at least as well. However, as I’ve said before, chiropracters become physical therapists with delusions of grandeur when they start claiming that they can treat allergies or conditions not directly related to the spine. The seemingly reasonable claims are used to try to legitimize the claims that are flat-out woo. The same is true for various nutritional therapies and supplements. When used according to the evidence, they are firmly in the realm of science-based medicine; but CAM practitioners make claims far beyond what can be justified by the evidence. Some surveys of this nature even include prayer; given that 90% or more of people in the U.S. believe in God or a higher power and that large proportions of people pray, especially when they are sick, including prayer and “spiritual” approaches under the CAM rubric allows even more inflation of the apparent numbers.

What disturbed me more than the CDC survey that Steve so masterfully took on not once but twice, however, is a part of the study that dealt with children, the results of which were published in the December issue of Pediatrics. This is argumentum ad populum taken to a whole new level. If adults use this stuff, it’s one thing. They are, after all, adults and can do choose any therapy they like. But to inflict unscientific therapies on children? Having seen what “alternative” (or, as the quacks like to call it, “biomedical”) therapies autistic children are subjected to, I have grave concerns about subjecting children to woo that go far beyond my concerns about adults using unscientific treatments. Some of the findings of this study are disturbing indeed. The article appears to be freely available; so you can read it yourself. However, I can’t resist mentioning a a couple of “highlights”:

  • “41% used multivitamins.” Of course, what’s “alternative” about a multivitamin?
  • “14% of all chiropractic visits were for pediatric patients.” Why children would need spinal manipulation for anything is hard to imagine.
  • “A range of 2% to 10% of children use homeopathic remedies, most often for respiratory problems, teething, otitis media, and other conditions related to the ears, neck, and throat.” Homeopathy is quackery, period.
  • “In Seattle, Washington, 70% of homeless adolescents reported using some form of CAM and among 9th- and 12th-grade students in Massachusetts, herbal remedies were used by up to 20% of respondents.” Of course, homeless teenagers would probably not have the means to see or pay for real medicine; so it’s not surprising at all that they would turn to CAM.
  • “Children with special health care needs are frequent users of CAM. The rate of CAM use for this population is estimated to be 30% to 70%. In a recent survey of families of children with developmental disabilities, families wanted their clinicians to be able to counsel them about CAM options.” This is not at all surprising. Look at all the unscientific and sometimes even dangerous “biomedical” therapies for autism there are out there. And that’s just autism.

Then the authors make a howler of a statement that utterly floored me (and demonstrated that they are in fact believers):

Although many CAM therapies have not yet been evaluated formally in children, a 2002 review identified more than 1400 RCTs and 47 systematic reviews of pediatric CAM. Formal evaluation has suggested that the quality of RCTs of CAM is as good as that of RCTs of conventional medicine, and the quality of systematic reviews of CAM exceeds that of systematic reviews of conventional medicine. It should be noted that publication bias in CAM research is opposite that of conventional medicine; that is, negative studies are more likely to be published in well-known journals, and positive studies are more likely to be published in foreign-language journals. Those interested in promoting an evidence-based approach to the use of CAM therapies must be cognizant of the bias created by applying language restrictions in their search strategy. Other approaches to evidence-based CAM include n-of-1 evaluation, whereby methodologic rigor (eg, blinding, randomization) is combined with an individualized approach fundamental to many CAM therapies.

Reverse publication bias? Systematic bias? What are these guys smoking? The reason foreign language reports are more likely to be positive is because they are more likely not to be methodologically sound, at least if they come from China and are about any form of traditional Chinese medicine, where negative trials are rare. Such trials are less likely to be properly blinded and less likely to use appropriate placebo or “sham” acupuncture controls. The interpretation made in this report is almost Orwellian in its language. It especially worries me that this report was apparently assembled as an official project of the AAP. It leads me to think that the AAP has fallen down the rabbit hole of woo, along with Yale and Beth Israel. And, of course, it’s not because this woo doesn’t work that clinical trials keep failing to identify a treatment effect. Oh, no:

There are some unique considerations when examining the efficacy of CAM, including heterogeneity of both products and practices. Lack of regulation of many commonly used practices exacerbates heterogeneity, making treatment effect difficult to measure. The relative lack of CAM expertise in conventional institutions results in inadequate peer review and undue difficulties when attempting to obtain institutional review board approval to study CAM in children.

Or maybe IRBs are smarter than the woo-meisters give them credit for. At least, I’d like to thinks so. Their charge is to protect human subjects, but the Common Rule gives them a special charge to be especially vigilant when it comes to vulnerable populations such as children. Perhaps IRBs are pretty good at recognizing unscientific bullshit when they see it. Of course, my concern is that they are still not good enough at it, given how many dubious and arguably unethical CAM studies on children manage to get IRB approval.

When you see all these reports of how CAM usage is supposedly increasing rapidly and that so many people use CAM, all I can say is: Be skeptical. In reality, CAM usage hasn’t changed very much throughout the years, at least among adults. Moreover, the number of modalities that advocates claim under the CAM rubric includes a number of what should be solidly “conventional” therapies, along with religious and “spiritual” interventions that huge swaths of the population have always done without calling it “CAM,” such as prayer for illness. Steve Novella and others are right to call it a “bait and switch.”

And I’m tired of seeing arguments of popularity presented to browbeat physicians to accept unscientific CAM modalities. Indeed, it’s exactly the same sort of game that antivaccine propagandist and uber-crank David Kirby plays when he points out how many people have made statements that would lead one to conclude that they believe vaccines cause autism or that vaccines might cause autism, regardless of their qualifications as experts (or, more frequently, their lack of qualifications) and regardless of context (bureaucrats trying to placate parents; scientists using the usual caveats of uncertainty that they do about research; or quotes made years ago before the question was as settled as it is now). It’s almost as though CAM advocates are trying to bludgeon science-based physicians into an attitude of “if you can’t beat ’em, join ’em.”

Actually, there’s no “almost” about it. That’s exactly what they’re doing.


K. J. Kemper, S. Vohra, R. Walls (2008). The Use of Complementary and Alternative Medicine in Pediatrics PEDIATRICS, 122 (6), 1374-1386 DOI: 10.1542/peds.2008-2173