As many of you may know, my mate Simon Singh is currently being sued for libel by the British Chiropractic Association. Chris French has recently put a great summary of the issues in the Guardian, and I would urge you to support Simon’s campaign for freedom of speech in science.
Aspects of the case revolve around the evidence (or lack of it) for the chiropractic treatment of children with colic. Although there is some anecdotal support for this notion, in my opinion such stories would not cut the mustard in the scientific arena (as the gag goes, the plural of anecdote is not evidence). After all, mothers may be give an inaccurate report of their children’s behaviour, or may have sought out treatment when the condition was especially bad and thus any effect could simply due to the symptoms becoming less severe over time.
That’s why controlled studies are so important. In my opinion, the only decent study to date was carried out by Olafsdottir and colleagues in 2001. Unlike the handful of other experiments into the topic, this study used proper blind procedures and controls, with, for example, the mothers involved not knowing whether their child was receiving chiropractic treatment or simply being held. The results showed that chiropractic spinal manipulation was no more effective than the placebo condition. I am not the only one to think that this is a vital study. For example, a 2002 review of chiropractic treatment for infant colic in the Archives of Disease in Childhood (here) only cited 4 studies – one of which was the Olafsdottir experiment, which was singled out for its sound methodology.
Yesterday I put in the search terms ‘chiropractic research infant colic’ into Google, and looked at the chiropractic sites that offer evidence about the efficacy of the chiropractic treatment for infant colic. I couldn’t find one that described the Olafsdottir study.
Of course, it is up to them which evidence they list, but I would have thought that a major study like that deserved a mention somewhere. Am I missing something? Can you find one?