Plus, the usual media gullibility
Starting with a provocative statement and finishing with nonsense worthy of Lewis Carroll, the headline in the September 11 Calgary Herald read; Acupuncture helps relieve headache, back pain, whether placebo effect or not: study.
According to the article, “The new analysis was published online Monday in Archives of Internal Medicine. The federal government’s National Center for Complementary and Alternative Medicine paid for most of the study, along with a small grant from the Samueli Institute, a non-profit group that supports research on alternative healing.” Researchers with Memorial Sloan-Kettering Cancer Center in New York and several universities in England and Germany wrote that the results ‘provide the most robust evidence to date that acupuncture is a reasonable referral option,’ .
Archives of Internal Medicine, researchers led by Andrew Vickers, an epidemiologist and biostatistician at Memorial Sloan Kettering Cancer Center, report that stated that; “The effects of acupuncture are statistically significant and different from those of sham or placebo treatments,” says Vickers. “So we conclude that the effects aren’t due merely to the placebo effect.”
Dr. Vickers may be an epidemiologist and biostatistician, but clearly, needs to go back and take an introductory course on statistics. This is a first year statistics students error, misinterpreting statistical significance as meaning practical importance. Statistical significance means the difference was detectable by the measurements system (including things like the sample size, sensitivity of the measuring instrument). Differences in telephone numbers are statistically significant. The question is whether these differences are important.
Confusing statistical significance as a measure of practical importance is a sign of incompetence or worse, the desperate act of researchers to get something, anything, published. Or perhaps it’s just desperate act of NCCAM (the black hole of pseudo-scientific research at $2.0 billion and counting) to actually produce a finding. In essence, Dr. Vickers and his team, are making a pun, switching the statistical meaning of significance (detectable) with the everyday meaning (important) in the hope that; (i) you won’t notice, and (ii) you will believe this statistical con-game for real science. Either way, the joke is on anyone believing it and American taxpayers paying for it.
The actual results showed a decline from a baseline pain level of 60 (on a scale of 0 to 100), to 30 for those receiving acupuncture, to 35 for those receiving fake acupuncture (the placebo), and 43 for people receiving the usual care and no acupuncture. From 35 to 30–are you kidding me? Like loaded guns, some people shouldn’t be left alone with a statistical software program.
Equally bad is the level of journalism involved. Close to 500 words on the subject of which about 40 are dedicated to the voice of reason from Dr. Stephen Barrett. A few more words from the good doctor, who actually knows what he is talking about, wouldn’t hurt. Where’s the journalistic balance?
The author, Lindsey Tanner, is supposed to be a “Medical Writer” for The Associated Press, although, here again, clearly not a very good one. A health writer, should actually understand what a placebo is. Ms. Tanner clearly doesn’t. She is too busy selling pseudo-scientific nonsense as science. Perhaps she should be the “Alternative-Medical Promoter” for Associated Press. Now that would fit.