One of the great things about being on holidays is when you can spend as much time as you want reading: a rare luxury which I have been indulging this week by reading five books at once, one of which was Bad Science, by Ben Goldacre, based on his column in the Guardian (see http://www.badscience.net/)
It is fascinating if depressing reading: Goldacre's world is one in which the public are constantly being hoodwinked by a world of charletans, quacks and journalists with humanities degrees, who all trick us into thinking that science is a complicated and boring speciality. Journalists fail to explain or understand the methodology behind experiments, and worse, they and most of us, their readers, do not even realise that we do not understand, and think we are experts. If we all understand some basic concepts, Goldacre maintains, we would not be vulnerable.
One of the things that it is most important to understand is statistics and probability, so that we can correctly interpret the results of experiments.
Goldacre explains how systematic review can judge all the available evidence on a subject as well as evaluating the validity of the methodology used in experiments to produce evidence based authoritative information. Such evidence based information provided by the Cochrane Institute and others has helped to improve practices and has saved thousands of lives.
Medicine is a life and death issue and myths such as that linking MMR vaccinations with autism have led to a rise in death rates as parents fail to innoculate their children properly.
I found it difficult to put the book down (so I didn't) and I found myslelf thinking that statisticians must get very frustrated that other people don't share their insight into what is really happening in the world. The book also reminded me of my favourite book, Stuart Sutherland's Irrationality in which he argues for medical informatics. ( see http://www.guardian.co.uk/books/2007/feb/24/featuresreviews.guardianreview27) When I first read Sutherland's that book I believed that Sutherland's pleas for medical practice to be based on statistical evidence was extreme: now twenty years on it has become best practice. Goldacre says this is the medical leap forward that the twenty first century will be remembered for.
Bad Science by Ben Goldacre Fourth Estate,2008 price £12.99
Irrationality by Stuart Sutherland Pinter and Martin,)1992 reprinted 2007, price £8.99
Comments