Rapid Responses to:

LETTERS: James Penston The urge to sprinkle statistics is irresistible BMJ 2007; 334: 440-a [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Count me out

Stephen J Senn   (3 March 2007)

Count me out 3 March 2007
Stephen J Senn, Professor of Statistics University of Glasgow, G12 8QQ Send response to journal: Re: Count me out	If, in all the centuries since Galen in which physicians were slaughtering their patients, they had been content to only use treatments that were effective by the robust standards that Penston proposes, they would have had little to do but millions of patients would have been saved. Alas, the problem is not that physicians have had a natural therapeutic conservatism that modern statistical methods have overthrown but that they have been only too glad to rush in and bleed patients as a cure for everything. Indeed, your rival publication, the Lancet, is named after the favoured weapon of mass destruction. In the words of Artemus Ward, "It's not the things we don't know that harm us, it's the things we know that ain't so". It is only since doctors learned to count that they have begun to discover how many things ain't so. To those who doubt it, I recommend a strong dose of David Wootton's excellent book, Bad Medicine1. Reference 1Wootton, D (2006) Bad Medicine: Medicine since Hippocrates, Oxford University Press, Oxford Competing interests: I am a statistician and therefore biased in favour of evidence and logic.