Most complementary and alternative medicine (CAM) modalities are complete bunk. Among the most mind-numbingly stupid modalities are therapeutic touch (a misleadingly named “therapy” in which a practitioner does not actually touch the patient at all and instead manipulates the “energy field” that hovers about six inches from the patient’s body), homeopathy (a “medicine” that consists of pure water—the “active” ingredient is removed by repeated dilutions), and ear-candling (a therapy that actually works just like it sounds—you stick a freakin’ candle in your ear). Such silly, insane, and utterly astonishing alternative treatments are so little removed from claims about magic and fairies with pixie dust that they warrant the same amount of dismissive derision and should be brushed aside as worthless without even attempting to investigate them. Unfortunately, because we live in a society in which CAM is increasingly becoming incorporated into mainstream medicine, researchers are often forced to investigate such tripe to finally reveal the surprising, scientifically-supported conclusion that drinking plain water is no more beneficial than, say, drinking Aquafina, and that sticking candles in your ear doesn’t do much of anything except present a risk of setting your hair on fire or getting hot wax in your face. It is relieving to see that science is investigating such pressing, eminently worthy research questions based solely on the fact that this crap has saturated our culture through ingenious marketing and political clout. After all, why should science let plausible mechanistic explanations and scientifically-feasible hypotheses generate research when they can simply waste resources on foolish, faddish nonsense promoted by charlatans and funded by the gullible and the desperate?

Homeopathy, therapeutic touch, and ear-candling are easy targets, of course. Other CAM treatments are not nearly as implausible. Herbal medicine, for instance, is highly plausible on the face of things, although it can be implemented in a dangerous fashion. St. John’s wort, for instance, is an herbal remedy commonly used to treat depression, but if it is mixed with other common antidepressant medications (medications a depressed person is likely to be taking), it can lead to adverse side effects like serotonin syndrome, which can be deadly. Of course, herbs and other medicines based on plant products are not outside the realm of western medicine. It is well-known that plants can generate various chemicals that can be used to treat ailments; aspirin, for example, is derived from the bark of the willow tree. What makes herbal medicine dangerous is that, unlike modern pharmaceuticals, the herb’s active ingredient is not isolated for more effectiveness, many of the herbs are not efficacious at all or in certain contexts in which they are marketed, and as I mentioned above they can have serious interactions with other drugs. And it goes without saying that many people forego more effective treatment in favor of these nearly worthless alternatives, setting themselves up for worsening medical conditions. So while it is true that not all CAM treatments are complete bunk, the large majority of them are, and those that aren’t complete bunk can still be characterized as semi-bunk or half-bunk for the unreasonable dangers they can present.

Besides herbal remedies, of course, acupuncture is probably one of the most highly touted alternative remedies. When initially described—acupuncture involves inserting needles into certain specific “meridian” points on the body through which life energy supposedly travels—acupuncture seems almost as crazy and absurd as a practice like ear-candling. However, a few studies have shown that acupuncture may be efficacious for treating various pain problems, and even more have tried to dubiously argue that acupuncture is effective even when its pain-relief results fared no better than placebo.

To get an idea how prevalent ideas about acupuncture’s efficacy have become, one need only peruse the popular press or simply Google the term “acupuncture.” As an example, here is a quote from Natalie Angier’s pop-science book The Canon:

When patients with lower back pain reported pain relief from bona fide needling but not from sham acupuncture, even the most skeptical Western doctors had to concede that the 5,000 year-old practice might have its limited uses. (p. 33)

Et tu, Brute? To be credulously stabbed in the back with acupuncture needles by a fine science journalist like Natalie Angier shows just how well such alternative modalities have saturated the healthcare industry. But though Angier is normally a fine journalist, she is certainly in err here. In the book, she refers to a study that tested traditional acupuncture (where needles were inserted into the “official” acupuncture nodes that harness the energy or qi or whatever) against sham acupuncture (in which the needles were inserted into “sham” spots not traditionally used in traditional acupuncture). The results, she said, showed that traditional acupuncture outperformed the sham acupuncture. Now, I am not certain what study she is referring to, but it has been widely demonstrated that traditional acupuncture does not outperform random needling. Even before the publication of her book in 2007, there were many studies that showed such results. As an example, a study by Brinkhaus et al published in February of 2006 showed that, among 298 patients assigned to acupuncture, minimal acupuncture, and a waiting-list control group that received no treatment, there was no significant difference between the acupuncture group and the minimal acupuncture group in relieving chronic low back pain. (“Minimal acupuncture” in this study consisted of superficial needling in non-acupuncture points.) Thus, for Angier to claim that “even the most skeptical” had to concede that acupuncture worked was an amazing overstatement. One medical study, much less a study of pain that could easily be marred by placebo effects, is not nearly enough to convince a skeptic, as the quality of the study could be poor, the interpretation could be misleading, and at any rate, studies like these must be replicated to be accepted without question.

As I said, though, Angier made her remarks concerning acupuncture in 2007. Has time and further research vindicated acupuncture? Not a chance. Acupuncture studies have actually been quite numerous since then, and most of them show that acupuncture performs no better than placebo. On top of that, the placebo form of acupuncture in many of these studies has improved significantly. In the studies mentioned previously, the placebo form of acupuncture consisted of needling non-acupuncture areas of the body. Later acupuncture placebos went even further, of course, and these forms of “sham” acupuncture did not consist of needling at all. Instead, a shaft would be placed on the patient’s body and produce the sensation of a needle without any sort of skin penetration at all. Thus, these placebos controlled not only for the specific placement of acupuncture needles, but also for the act of needling itself.

One such placebo-controlled trial using this form of sham acupuncture (it involved a blunt needle touching the skin and then being immediately retracted into the shaft) ironically showed that the sham performed better than the true acupuncture (Goldman et al). This study followed 123 patients who experienced persistent arm pain from overuse injuries, and the sham acupuncture group experienced more pain relief than the true acupuncture group during treatment (without the attendant pain and soreness of having a needle jammed into one’s self). Thus, this study showed that the placement of the needles and the mechanism of needling itself have no effect on pain.

As I mentioned earlier, though, scientists must be wary of studies, especially those involving pain, and the results must be replicated. Of course, many more studies have shown that acupuncture performs only as well as sham acupuncture and other placebo forms. After one analysis of 13 studies about acupuncture’s effects on pain (involving nearly 3,000 patients), the authors concluded:

Our findings correspond with several Cochrane reviews on acupuncture for various types of pain, which all concluded that there was no clear evidence of an analgesic effect of acupuncture.

Clearly, the evidence for acupuncture grows shakier with more research, and conveniently the higher quality research tends to show more negative results for its effectiveness. This makes sense given the initial implausibility and the seeming insanity of the treatment modality. But if acupuncture doesn’t work, how is it that some studies manage to show a positive result?

While I lack the resources to do an exhaustive examination of the literature, it seems that those studies that purport to show a positive effect stemming from acupuncture either do so through misleading interpretations of much less compelling data or else through poor study methodology.

As an example of poor study methodology, consider the “positive” results attained in a study by Carlsson and Sjölund: They showed that two acupuncture groups had greater pain relief from low back pain than a placebo group receiving treatment with a fake transcutaneous electrical nerve stimulation (TENS) device. The study is even featured on the Acupuncture Today website, which exclaims that acupuncture shows “significant” pain relief for low back pain after 1, 3, and 6 months. (The abstract for the actual study, interestingly, only claims “significant” pain relief for months 1 and 3.) Why are there scare quotes around “significant”? Because it is being used in a statistical sense. When a person off the street says they feel a significant reduction in pain, that usually means the pain has decreased by quite a bit. In statistical medical journal jargon, of course, “significant” pain relief can mean as little as having minimal change. Looking at a graph of the data for this study, it appears the the difference in pain amounts to about 10 mm on a 100 mm pain scoring scale—which translates into “minimal” pain relief. (I don’t have access to the actual numbers, but it appears from the graph that the pain relief at 6 months may have just missed the “significant” cutoff at just under 10 mm.)

So yes, the study showed that acupuncture caused “significant” (that is, minimal) pain reduction in the lower back. Where’s the flaw in the methodology? Well, the flaw is that they included a form of acupuncture in which the needles are electrified (called electroacupuncture) and then combined that with the “true” acupuncture group. Obviously, the traditional Chinese healers from 5,000 years ago were not using electrified needles. But perhaps this shows that electroacupuncture works. Once again, this line of reasoning is flawed. As you may have noticed, the “placebo” group of this study involved the use of a fake TENS device. TENS devices are frequently used in the treatment of low back pain. Their mechanism of action is to stimulate nerves with electricity. That is, the standard, Western treatment of low back pain involves electrical stimulation, which is exactly what the electroacupuncture does. It is reasonable to assume, then, that the actual acupuncture needling has no effect whatsoever, and the limited pain relief is the result of the electrical stimulation. And given that other studies have isolated needling as a variable and shown it no better at pain relief than sham acupuncture with no needle penetration, it would appear that this study demonstrates nothing whatsoever about acupuncture, and at best shows that standard treatment with an electrical stimulation device is beneficial. Remember, kids: If you’re going to do medical research, isolate your goddamn variables. If the only way to legitimize your alternative, 5,000 year-old treatment is to haphazardly vivisect it and sew its corpse onto standard medical treatments like transcutaneous electrical nerve stimulation, then it would seem patients would be better off with the standard care without the elaborate, ritualized nonsense attached to it.

Naturally, when the data itself can’t be manipulated into showing a positive result through poor methodology, many study authors and journalists instead prefer to interpret the results in a stunningly incompetent manner.

Now, if I were researching the efficacy of a certain drug and compared it with a placebo sugar pill in a study, subsequently finding that the placebo sugar pill and the drug both improved outcomes, it would be highly unreasonable for me to exclaim that the drug—and placebo sugar pills—work. Obviously, the purpose of a placebo control is to account for the placebo effect. If a treatment does no better than a placebo, that means the treatment doesn’t work—the pain relief may be caused primarily by psychological effects like self-reporting biases and so on. In the acupuncture world, of course, a study that shows true acupuncture performing only as well as the placebo acupuncture is reason to go about proclaiming a success, that both the placebo acupuncture and the regular acupuncture work! When this sort of nonsense is tried by drug researchers it seems stupid and crazy, but for whatever reason acupuncture researchers and credulous journalists routinely get away with making such claims about the placebos in acupuncture trials.

In a study by Cherkin et al, for example, the researchers studied pain-reducing effects of acupuncture for low back pain in three acupuncture groups versus a group receiving standard care (the acupuncture groups also received standard care in addition to being needled, though). The acupuncture groups consisted of a group receiving “individualized” acupuncture care, a group receiving standard acupuncture, and a group receiving superficial acupuncture that consisted of toothpicks that didn’t penetrate the skin. Not surprisingly, all three acupuncture groups performed better than the standard care group in terms of pain improvement. This was expected, naturally, because the acupuncture groups also received standard care, and the slight improvement could easily be attributed to the placebo effect, a conclusion tacitly acknowledged in the last sentence of the abstract. The important comparison is between the placebo “toothpick” acupuncture and the “real” forms of acupuncture, though. All of these performed the same, which just goes to show, once again, that the placement of the needles in certain spots does not matter and that the actual needling through skin also does not matter in terms of pain relief. If the placebo acupuncture performs as well as the real versions, this is no success to crow about. And yet headlines read, “Acupuncture, Real or Fake, Helps Aching Back,” while the researchers themselves often make similar claims. The interpretation of these results in the media is simply ignorant and stupid, enforcing the misconception that acupuncture actually works.

Thus, the initial gut-reaction to poking oneself with multitudes of needles turns out to be correct: Acupuncture is a nonsensical waste of time, producing health benefits only on par with a placebo and certainly not warranting the time, money, and effort that go into such a treatment. More and more, studies show that acupuncture does not work compared with sham treatments and other placebos, and those studies that do show benefits are severely flawed, either by using crazy hybrid acupuncture needles incorporating traditional Western treatments (like the electroacupuncture needles that mimic TENS devices) or just using bad methodology in general. In the end, the positive results of acupuncture studies are largely illusory, created either by the media, as they whore for attention with a catchy headline, or created by the patients themselves as a placebo effect.

References:

Angier, N. The Canon: A Whirligig Tour of the Beautiful Basics of Science. Houghton Mifflin Harcourt. 2007.

BMJ-British Medical Journal (2009, January 29). Pain relieving effects of acupuncture are limited. ScienceDaily. Retrieved June 4, 2009, from http://www.sciencedaily.com­ /releases/2009/01/090127202048.htm.

Brinkhaus B, Witt CM, Jena S, et al. Acupuncture in patients with chronic low back pain: A randomized controlled trial. Arch Intern Med 166(4):450-457, 2006.

Carlsson CP, Sjölund BH. Acupuncture for chronic low back pain: A randomized placebo-controlled study with long-term follow-up. Clin J Pain 17(4):296-305, 2001.

Cherkin DC, Sherman KJ, Avins AL, et al. A randomized trial comparing acupuncture, simulated acupuncture, and usual care for chronic low back pain. Arch Intern Med 11;169(9):858-866, 2009.

Devitt M.  Acupuncture provides long-term relief of low back pain. Acupuncture Today 3(3), 2002. Retrieved June 4, 2009, from http://www.acupuncturetoday.com/archives2002/mar/03lowbackpain.html.

Goldman RH, Stason WB, Park SK, et al. Acupuncture for treatment of persistent arm pain due to repetitive use: A randomized controlled clinical trial. Clin J Pain 24(3):211-218, 2008.

Steenhuysen J. Acupuncture, real or fake, helps aching back: Study. ABC News Health, May 11, 2009. Retrieved June 5, 2009, from http://abcnews.go.com/Health/PainNews/wireStory?id=7560289.

(This post is dedicated to the recent commenter who claimed alternative medicines work, specifically citing acupuncture. Thanks for making me waste hours of my time writing this lengthy reply.)

This entry was posted on Friday, June 5th, 2009 at 12:12 AM and is filed under Alternative Medicine, Skepticism. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

Tweet