NCCAM director Stephen Straus believes that there is promising evidence that acupuncture is effective in treating certain conditions. Harvard University's Marcia Angell is skeptical.

Director, National Center for Complementary and Alternative Medicine, National Institutes of Health

How has the practice of acupuncture in this country evolved?

Acupuncture [is] a very good illustration of the potential of alternative medicine. When I was in medical school Nixon went to China and James Reston wrote about his experience with acupuncture. What had seemed bizarre and arcane suddenly became fashionable in a country that once again was interested in things Asian...Today, forty years later, acupuncturists practice in every major medical center in every state and in every strip mall throughout the country.

Americans are fascinated by acupuncture, millions have had acupuncture, and many individuals think it works. Now, what happened in those thirty years? Part of it is a cultural and social phenomenon, part of it has to do with the fact that the community of acupuncturists were aggressive and seeded themselves throughout the country. Part of it had to do with huge immigration waves from Asia that brought those techniques as well as the acceptance of those techniques.

But two other very important things happened...There was decent evidence emerging that it was useful for some indications, and there were data suggesting how it works. Starting in the 1970s, scientists in China and elsewhere began to show that acupuncture activates certain parts of the nervous system and engages processes that can relieve pain the same way that morphine relieves pain. So if you understand the mechanism and you have some evidence that it works, it becomes a lot easier to accept it as a rational alternative to other approaches.

What do the data say about the benefits of acupuncture?

There are thousands of publications; most of them are not worth their salt. Many are decent studies, but they're small. There are few large studies and we today at the National Center for Complementary and Alternative Medicine are funding the largest ever studies of acupuncture.

We do not yet know whether it is better than conventional non-steroidal anti-inflammatory drugs for the pain of degenerative arthritis. That's a common use, and we don't know. There was enough data to suggest that it was a worthwhile investment. The study at the University of Maryland is enrolling hundreds of aging Americans who have degenerative painful arthritis in the knee and asking whether acupuncture is successful in relieving them of that pain.

Why is there reasonable data that acupuncture is good for dental extractions, but not for degenerative arthritis of the knee? It's because dental extractions are common, it's a very good model to study acute, temporary pain. It's sometimes harder to intervene effectively in chronic pain, so in research you'll often study that from which you can get the best answers. It's not that dental pain is as important a public health problem as arthritis.

When it comes to acupuncture, physicians today struggle to provide comfort for their patients...Now, acupuncture makes sense for pain, because there's a great deal of data saying that it's beneficial for pain. We just don't have data saying it's as good as our other treatments or that it's better than our other treatments or it's more cost-effective than our other treatments.

But for a given individual who is struggling with pain, and you want comfort for that person, acupuncture is practiced widely, there are standardized needles, there's reasonable agreement on approaches for certain uses of acupuncture, and there's reasonable data for certain uses and we have reasonable explanations of how it works. That's enough for physicians to recommend it. It's not enough for physicians to say, "I'm going to use it to treat your lung cancer, I'm going to use it to treat your brain tumor."

	Senior Lecturer, Harvard Medical School

I have never seen a good study that shows that acupuncture works. Now the NIH a few years ago had a panel that ostensibly reviewed the world literature on acupuncture and suggested that there was evidence to the effect that it might work for two conditions. One I think was toothache, the pain from toothache, and the other was nausea of chemotherapy I believe. And that was it. A lot was made of that. It was suggested that the NIH had found and it was implied through its own research that acupuncture works, period. But in fact, looking at the whole body of research, it might work for these very, very narrow conditions out of the whole body of conditions for which it's been tried. Both of these conditions are subjective. It's not like reversing heart failure or curing your cancer. You're reporting that your toothache feels better. You're reporting that you're less nauseated. So you have to ask how well were these studies controlled? And a control means making sure it's not a placebo effect, that the subjects didn't imagine that they felt better, or feel better because they believed in it. So to control then for the placebo effect in acupuncture requires that the patients see needles going in in the same way that they might with acupuncture, that they be fooled. And I don't know whether these studies that were purportedly positive involved control groups where you fooled the control group or not. I just don't know. Other than that, I don't know of any good studies that have shown that it works. Certainly nothing like the claims that are made for it.

It's generally accepted as, it works. And yet it's based on a philosophy so primitive that it's amazing to me that people could imagine that it does. I mean, these meridians of electrical fields and so forth. It's based on a philosophy that goes back to the period of time in which it developed. Which is quite primitive. …

What do you say to people who believe that acupuncture works?

Well it speaks to the power of the placebo effect, which is extremely powerful when it comes to subjective complaints like pain or nausea, particularly if they're not terribly severe. If you were in agony because your leg was crushed, probably the placebo would not work. But I suspect in most cases, acupuncture works through a placebo effect. People believe in it, the discomfort is not that severe and they're not objective complaints, but just subjective complaints. And so this placebo effect kicks in, they feel better. We can all see this. If you have a cold and you're feeling crummy and then something really nice happens to you, you forget the cold for a couple of hours. The placebo effect operates on all of us every day and I would suspect that that's what's going on with acupuncture. …