Does NIH recommend acupuncture?

"NIH recommends acupuncture" - that's the headline on a recent web article on an acupuncturists' website. Now, I know that NIH has granted millions of dollars to acupuncture researchers who have tried their best to prove it works, but I didn't think NIH officially endorsed it. For one thing, despite these wasted millions, the best evidence shows that acupuncture doesn't work, not even a little bit.

What the article was referring to illustrates the dangers of NIH's far-too-gentle treatment of acupuncture and other "alternative" practices. It also shows how promoters of questionable or downright bogus practices can distort the facts to suit their own ends.

What the acupuncturists found was an NIH news site that ran an article called "Understanding acupuncture: time to try it?" (Note the question mark.) The article was written by Harrison Wein, a science writer at NIH who interviewed a handful of researchers, most of them promoters of acupuncture themselves who are funded by NCCAM.

(I've written about NCCAM before, so here I'll just remind readers that it was created by Congress as a way to earmark funds for bad science that can't pass muster in normal peer review. NCCAM grants over $200 million per year to its stable of mediocre scientists.)

Wein did a very poor job of describing the complete implausibility of acupuncture, but he did at least point out that sham acupuncture works just as well as "real", and it doesn't even matter if you use toothpicks that don't pierce the skin. So does his article endorse acupuncture? In a sidebar titled "If you want to try acupuncture," it says:

• "Don’t use acupuncture as a replacement for conventional care"
• "Don’t rely on a diagnosis of disease by an acupuncturist who doesn’t have conventional medical training"

and a few similar things, none of them recommendations for acupuncture. The article concludes by asking the question again:

"Should you try acupuncture? Studies have found it to be very safe, with few side effects. If you’re thinking about it, talk to your doctor."

So no, not an endorsement. Just a much-too-gentle recommendation to talk to your doctor first. The writer (Wein) regurgitates the claims of acupuncture promoters hook, line, and sinker - and he doesn't cite any evidence for his statement that it is "very safe." In fact, acupuncture often causes infections, sometimes serious ones. And as I wrote last month in The Atlantic: in 1995, a 40-year-old Norwegian woman visited an acupuncturist for relief from fibromyalgia. As described in The Lancet, she died two hours later, and an autopsy revealed that the needle had punctured her left ventricle.

Think that's just one unfortunate anecdote? Well, in the journal BMJ last year, researchers reported that acupuncture infections are a significant problem worldwide, and that they have been under-reported for years. Infections caused by acupuncturists have led to "joint destruction, multi-organ failure, flesh-eating disease and paralysis" as well as hepatitis B and C and mycobacteriosis.

So I don't agree that acupuncture is "very safe." After I gave an interview on Minnesota Public Radio last month, I was attacked by acupuncturists claiming that they don't use "dirty needles" (which I never claimed) - but in saying this, the attackers reveal their own ignorance. Needles aren't the problem: it's that every site on the skin as well as the acupuncturists hands that need to be sterilized. You see a photo on the website of the American College of Acupuncture and Oriental Medicine of a bare-handed person inserting needles into someone's skin. In case they remove it after reading this article, here it is:

All this should be beside the point, though, because acupuncture simply doesn't work. One of the acupuncture researchers (funded by NCCAM) interviewed in the NIH news story, Karen Sherman, had this response to the data showing that toothpicks work as well as needles:

"For example, when researchers have compared inserting needles with just pressing a toothpick onto acupuncture points, they’ve often found both treatments to be successful. But Sherman questions whether these are really controls. Many traditional acupuncturists would consider them true treatments, too. The important thing, in their view, is to hit the right spot, not necessarily how deep you go."

So when the evidence doesn't show what she likes, Dr. Sherman changes the definition of placebo. This is called "moving the goalpost," and it's a classic sign of bad science (and a bad scientist). Actually, she goes even further, saying "I don’t really think you can come up with a great placebo needling." In other words, in Dr. Sherman's world, acupuncture can't really be tested. I guess it's just magic.

Even though the NIH article doesn't recommend acupuncture, it uncritically repeats some ridiculous claims, such as:

"… the body contains a delicate balance of 2 opposing and inseparable forces: yin and yang. Yin represents the cold, slow or passive principle. Yang represents the hot, excited or active principle. Health is achieved through balancing the 2. Disease comes from an imbalance that leads to a blockage in the flow of qi—the vital energy or life force thought to regulate your spiritual, emotional, mental and physical health. Acupuncture is intended to remove blockages in the flow of qi and restore and maintain health."

I want to laugh at this, but it appears on an honest-to-goodness NIH website. As my colleagues at Science-Based Medicine put it, "acupuncture is a pre-scientific superstition." It does not deserve our respect, nor should we take it seriously.

There's no scientific evidence whatsoever that "the flow of qi" has any physiological basis. The passage above should have been followed with a bit of real information, such as: "Scientifically, there is no more evidence for qi than there is for the tooth fairy or Santa Claus. However, some researchers argue that Santa Claus is quite a bit more plausible." But that would perhaps be hoping for too much backbone from NIH. Note to NIH: don't give pseudoscientists the imprimatur of legitimacy by repeating their claims. And pay better attention to what you allow on your website.