Rapid Responses to:

EDUCATION AND DEBATE: David J Spiegelhalter, Jonathan P Myles, David R Jones, and Keith R Abrams Methods in health service research: An introduction to bayesian methods in health technology assessment BMJ 1999; 319: 508-512 [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Bayes Theorem---12 years ago

James E Parker   (23 August 1999)

Bayes Theorem---12 years ago 23 August 1999
James E Parker, Paediatrician -retired 289 McCallum Rd Abbotsford BC V2S 8A1 Send response to journal: Re: Bayes Theorem---12 years ago	Bayes' theorem was invoked in a 'classic' paper by Meyer and Pauker in the New England Medical Journal in 1987 titled ' Screening for HIV : Can we afford the false positive rate ?(NEJM.1987; 317: 238-241 ). The philosophy imparted had much to do with attitudes to HIV screening and it's legacy on the present world pandemic. They assumed a false positive rate of 0.005% for combined Western blot and ELISA testing and a seroprevalence rate in the general population of 0.01% (this represented the incidence of HIV in US female blood donors in 1986 ). Thus testing a population of 100,000 people would reveal 10 bona fide cases and 5 false positives. These odds were considered an argument against screening. That female blood donors were not truly representative of the general population and that random false positivity in testing could be appreciably reduced by repeat testing went unnoticed. This would seem to underscore a statement in the final paragraph of the article by Spiegelhalter, Miles,Jones and Abrams. 'The perceived problems with the bayesian approach largely concern the source of the prior and the interpretations of the conclusions'. James E Parker MB FRCPC