Wednesday, September 26, 2007

The public has been exposed to the media over hyping (and misrepresenting) a study that finds acupuncture to be more beneficial than "western medicine". I was first exposed to this by an article on MSNBC entitled "Acupuncture - real or fake - best for back pain." Emphasis added.

Let's take a closer look. Here is a link to the full text. Luckily, this study, and the reporting of it, seems to be recieving great analysis and coverage. This includes Orac over at Respectful Insolence; Dr. Steven Novella's NeuroLogica Blog; and Dr. Ben Goldacre at Bad Science. These two are more intellingent than me, and, most importantly, are able to express themselves better. Please take the time to read their posts. Especially Orac's.

I would just like to highlight a few problems with this study. Understand, though, most studies have flaws and that does not mean they are worthless. However, these flaws must be known because they greatly effect how a study can be applied to you/me/us.

1. This study, as designed, is wholly unreproducible. Why? The "conventional treatment" group was not controlled (well). Some received all, some, one, etc of the "allowable" conventional treatments. Maybe acupuncture is a cure all, maybe not. However, another study using the same methods, cannot be done because the conventional treatment group can not be reproduced.

2. The inclusion/exclusion criteria eliminated the vast majority of patients we see (and for those reading, what you have). Therefore, cannot be applied to much of the LBP population. Orac argues this point well:
One point that leaps right off the page is that the patient population studied had had chronic low back pain for at least 6 months and, based on that duration and the patients' willingness to try acupuncture, we can reasonably infer that their pain probably wasn't responding particularly well to conventional therapy. This makes it unsurprising that the reported response rate in the standard therapy group was so low, given that it was just getting more of the same treatment.
3. There was no blinding. All the participants knew what they were receiving, and, presumably, the same practicioner was providing the "real" and "sham" acupuncture. Although is both the real and sham acupuncture did just as well...does that mean that all acupuncture is sham or that any needling is real acupuncture? Hmm..... In my opinion this opens the door to rater bias and a very large placebo effect (hmm, the 'I'm receiving a "new" treatment' group does better than the 'I'm receiving the same crap that hasn't seemed to work in the past' group).

What to make of this study then? Those with chronic LBP not linked to any known causes (sciatica, DDD, OA, Surgery, etc) that "conventional" treatment has not worked, improved with "shamish" acupuncture compared to the same old crap. Not quite the same as "Best for back pain".


Post a Comment