Europe Travelblog  -

18 May '09 - + 25 - 16 Disappointed in NPR

I very rarely find myself yelling at the radio when I'm listening to NPR, but this story about acupuncture this morning did the trick.

I'm very disappointed by this report.  The study discussed did not show that acupuncture works.  What it showed is that acupuncture is no better than a fake version of acupuncture that the recipient thinks is real. This means that acupuncture itself provides no benefit other than the recipient thinking that there is a benefit.

That is called the placebo affect. Acupuncture has never been shown to be anything more than a placebo.

Yonkers (a patient interviewed in the story) is seeing some benefit from her monthly treatment, but there are two strong possibilities that are much more likely than acupuncture efficacy:

1. The placebo effect
2. Spending an hour a month in a soothing environment, a spa where "I know it's just going to be time to relax and be by myself," gives her what she needs. Have her go to a spa without the needles, and she'll probably see a benefit. Have her be stuck with needles while children are banging pots around her, and there's probably no benefit.

This story should have been about how the ritual and the relaxation create a placebo effect.  Instead, it just promoted woo and fake medicine.

Steven Novella of the Skeptic's Guide to the Universe has a lot more information about this inaccurate spin of acupuncture's failure.

Update: Below is an adaptation of a comment I left on the NPR story

For all those who feel discouraged by the belittling treatment acupuncture receives, there's one simple way to stop that treatment: prove that acupuncture works.

The opposition to acupuncture does not come from a bias or bigotry against Eastern or alternative treatments; it comes from a bias _for_ treatments that work. Unfortunately, acupunture has never been shown to work.

Sure, there are plenty of anecdotal reports about how individuals think it helped them, but the plural of anecdotes is neither data nor evidence. It's anecdotes.

Simply create a well-designed, blinded, reproducible study that correctly isolates the variables to show that acupuncture itself has a specific effect. It would also be nice if there were a plausible mechanism for how it works, but that can come later. First, show unambiguously that it works, and we'll be on board.

After all, as Tim Minchin notes, what do we call alternative medicine that has been shown to work? Medicine.

However, every well designed study so far has found the same result; there's no demonstrated evidence that acupuncture works better than a placebo. Until that changes, we have no reason to respect acupuncture more than snake oil.

two comments, already:

Man, you and I really are reading all the same blogs these days.

Rich Unger - 18 May '09 - 17:35

Do you listen to the SGU? I think it was one of the first podcasts I got subscribed to.

Lance Finney (link) - 18 May '09 - 21:17