Acupuncture needling

A recent editorial entitled “CAM in the Real World: You May Practice Evidence-Based Medicine, But Your Patients Don’t” published in Headache: The Journal of Head and Face Pain by Robert Cowan, a headache specialist, addresses the use of complementary and alternative medicine (CAM) in the treatment of headaches. Unfortunately he propagates many common misconceptions about CAM in the article.

I do agree with one point – physicians need to be more aware of CAM treatments and their patients’ use of them. We should be directly asking our patients about such use, in a non-judgmental way, and we should be familiar enough with common CAM treatments so that we can provide knowledgeable guidance to our patients.

Cowan begins by, in my opinion, grossly exaggerating the current popularity of CAM. He writes:

As much as 82% of headache sufferers use complementary and alternative approaches.

The reference he cites, however, states:

Adults with migraines/severe headaches used CAM more frequently than those without (49.5% vs 33.9%, P < .0001); differences persisted after adjustment (adjusted odds ratio = 1.29, 95% confidence interval [1.15, 1.45]). Mind–body therapies (eg, deep breathing exercises, meditation, yoga) were used most commonly.

Only 4.5% of adults with migraines/severe headaches reported using CAM to specifically treat their migraines/severe headaches.

In order to get the inflated figure you have to include treatments that are, at best, in the gray zone. This is a game proponents of CAM have been playing for years – the definition is vague enough that they can include common practices an inflate the apparent popularity of CAM. Even the lower figure of 49.5% includes people who are simply exercising.

Cowan also did not mention that the same study found that only 4.5% of migraine patients use CAM to treat their migraine. That would have undercut his premise that use is common so that headache doctors need to pay attention.

He continues with perhaps the second most common CAM trope:

There is limited evidence suggesting the vast majority of these treatments are harmful (regardless of the evidence they are helpful), and most have withstood “the test of time,” having been handed down over hundreds, even thousands of years.

“The test of time” has proven to be utterly worthless. Galenic medicine, based on balancing the four humors, survived for a couple thousand years even though its interventions almost certainly caused harm and no benefit. It only faded away in the wake of scientific medicine.

The plural of anecdote is not data, and this includes anecdotes over a long period of time. Long use of treatments can be explained by cultural inertia, in the absence of any real benefit.

Further, lack of evidence of harm is not very compelling. Anecdotal evidence is useful in eliminating immediate obvious harm, such as deadly poisons, but not useful in exposing long term harm or low probability harm.

There are now plenty of example of herbal remedies, used for centuries, that only recently were discovered to have significant toxicity, such as aristolochia and renal failure or kava kava and liver damage. Many herbal supplements also contain heavy metal contaminants, or don’t even have the ingredients on their labels, opening the door for unknown allergic reactions or drug-drug interactions.

Risks are not limited to herbal drugs. Chiropractic neck manipulation has been linked to a rare but serious complication of stroke and even death. A review of acupuncture side effects found they occurred in 11.4% of treatments.

Much of the harm of CAM is also indirect. Patients expending time and limited resources on treatments that are ineffective may delay or even forgo science-based treatments. Even for self-limiting or purely symptomatic syndromes, there is potential harm in patients being convinced that an unscientific treatment is effective due to practitioner recommendation and placebo effects. If a patient feels their migraines are improved due to acupuncture, perhaps they will be more open to using acupuncture to treat their cancer and delay definitive treatment. (Yes – some “medical acupuncturists” advocate using acupuncture to treat cancer.)

It is unreasonable to expect practitioners in an unscientific paradigm to be science-based in the application on their interventions.

I also have to wonder which treatments Dr. Cowan believes have been around for thousands of years. Perhaps some herbs have been in use for that long, and those that work are mostly already incorporated into modern pharmacology. Acupuncture is often claimed to be thousands of years old, but prior to the 20th century acupuncture had more in common with bloodletting than what passes today for acupuncture. Chiropractic and naturopathy are less than two centuries old, and homeopathy is a little over two centuries.

Cowan also does not discuss evidence for lack of efficacy, and like many CAM proponents pretends that this does not exist. There is sufficient evidence, for example, that acupuncture does not work for any specific indication, and specifically does not work for migraine (despite the claims of acupuncturists).

Cowan makes the usual excuse, however:

The most widely used of these are Ayurveda, classical Chinese medicine, homeopathy, chiropractic, and naturopathy. Because these are medical systems rather than discrete interventions, studies are much harder to come by and in general, each has its own internal logic. It is much more difficult to evaluate a system which is based on centuries of trial and error or of an oral tradition.

Why? If any of these interventions work, regardless of their internal logic or history, careful clinical trials should show an effect. I would also point out that astrology has an internal logic. That is not sufficient to make a philosophy legitimate or recommend it as a medical system.

He mentions homeopathy in a few places in the article, without ever pointing out that homeopathy is based purely on magical thinking, that its premises run contrary to modern science, and that a large body of clinical evidence shows that homeopathy does not work for anything.

Here, however, is the worst statement in the entire article:

To summarize, what is presently accepted as conventional medicine came to be so by caveat. Other medical systems have neither been subject to the rigorous vetting that Western medicine approaches have undergone in order to further develop nor have they been demonstrated to be ineffective or dangerous in any systematic way. In other words, there are no scientifically accepted criteria for inclusion or exclusion of specific modalities as CAM or as conventional. Conventional medicine is exactly that – medicine by convention.

He is arguing that scientific medicine won out over competing medical systems because Flexner (a German educator and not a physician) arbitrarily decided that science would be the basis of modern medicine. This is an astounding argument, considering that the application of scientific principles transformed the practice of medicine and has arguably doubled the human life-expectancy since its inception.

Cowan again also ignores the fact that competing systems at the time not only lacked scientific evidence, many of them were demonstrably absurd. Homeopathy was ridiculed long before the Flexner report.

Further, regardless of history, scientific evidence is clearly the best tool we have to determining if a treatment is safe and effective. There is simply no viable alternative. He is also flatly wrong when he argues that CAM systems have not been demonstrated to be ineffective – they largely have, as I pointed out for homeopathy and acupuncture.


Cowan tries very hard to make it seem as if he is being objective and is simply looking out for the best interests of patients. In the end, however, in this article he has simply repeated the misconceptions and distortions typical of CAM proponents.

He makes a false dichotomy between “Western” medicine and alternative medicines. He exaggerates CAM’s popularity, and blurs the line between CAM and scientific medicine by including treatments that are solidly science-based, such as nutrition and exercise.

He pretends that the scientific basis of medicine is just an arbitrary historical choice, rather than the transformational and defining aspect of modern medicine that finally dragged the medical profession out of the era of witch-doctors.

He pretends that we do not already have sufficient evidence for lack of efficacy for most CAM modalities. I would add that a science-based approach would also consider plausibility, which most CAM modalities lack.

In the end Cowan used an elaborate and indirect justification for the same list of magical and disproved therapies that CAM proponents have been promoting for years.

I agree that physicians need to be more familiar with CAM treatments and discuss such issues with patients when appropriate. They should do so, however, from a solid understanding of the actual science and history of such treatments.


Posted by Steven Novella

Founder and currently Executive Editor of Science-Based Medicine Steven Novella, MD is an academic clinical neurologist at the Yale University School of Medicine. He is also the host and producer of the popular weekly science podcast, The Skeptics’ Guide to the Universe, and the author of the NeuroLogicaBlog, a daily blog that covers news and issues in neuroscience, but also general science, scientific skepticism, philosophy of science, critical thinking, and the intersection of science with the media and society. Dr. Novella also has produced two courses with The Great Courses, and published a book on critical thinking - also called The Skeptics Guide to the Universe.