Tag Archives: effect size

Acupuncture as an example for rational thinking

11 Sep

I’ve written previously about using homeopathy as an example for discussing rational thinking, but in the last couple of days I’ve come across a couple of articles about another ‘alternative medicine’, acupuncture , that might be useful in getting to deploy higher-order rational thinking skills.

On Tuesday 11th September 2012, the Daily Mail reported a meta-analysis of acupuncture studies under the headline ‘acupuncture ‘does help to relieve pain’ say US researchers’. There is a certain irony, in that six days earlier the Mail had reported a story headlined ‘The hidden perils of acupuncture’, that detailed a range of cases in which NHS acupuncture patients had come to some harm including some cases of punctured lungs ! Irony aside, my interested was peeked by a quotation from original story, that has the studioes lead researchers saying:

”Although the data indicate that acupuncture is more than a placebo, the differences between true and sham acupuncture are relatively modest, suggesting that factors in addition to the specific effects of needling are important contributors to therapeutic effects’

As you might imagine, the Mail doesn’t enlarge on what ‘relatively modest’ means, but a brief bout of Googling led me to the Guardian’s version of the sample story that does list the size of the effects found:

Differences better actual acupuncture and sham controls

   Back and neck pain  .23  standard deviations

   Osteoarthritis            .16  standard deviations

   Chronic Headache   .15  standard deviations

So, in reality whilst this study did find that acupuncture did have some benefit beyond placebo in fact the benefits found were tiny. It’s interesting that whilst reporting of this study is widespread, from India (‘   Acupuncture proved effective for chronic pain  ‘) to the USA (‘Acupuncture may actual work after all’) very few of the articles bother to mention that the effect found were so small.

This little saga seems to provide a number of useful teaching opportunities. ‘How would you go about testing acupuncture’ is a great question as it lead to interesting discussion of the placebo effect and especially how you might create placbo acupuncture. Simon Singh and Edzard Ernst’s book ‘Trick or Treatment’ a great chapter on this. Beyond that, this story is a nice illustration of the importance of ‘looking at the numbers’ . Saying that this study showed that acupuncture ‘worked’ for back and neck pain, osteoarthritis and chronic headaches might be technically ‘true’, but without looking at the numbers you get no idea of how well they ‘worked’. For the statisdtcally mined this is a nice example of why students should always report significance AND effect size.

Finally, this story is another example of the importance of reading beyond newspaper headlines and where possible looking at the original source material.

%d bloggers like this: