Ethics of clinical trials from bayesian perspective: Medical decision making should use posteriors, not priors

BMJ 2003; 326 doi: https://doi.org/10.1136/bmj.326.7404.1456-a (Published 26 June 2003) Cite this as: BMJ 2003;326:1456
  1. Stephen Frankel (stephen.frankel@bristol.ac.uk), professor of epidemiology and public health,
  2. Jonathan Sterne, reader in medical statistics,
  3. Shah Ebrahim, professor in epidemiology of ageing
  1. Department of Social Medicine, University of Bristol, Bristol BS8 2PR

    EDITOR—The article by Lilford is a curious contribution to the debate about recruitment to randomised trials.1 While each sentiment seems, individually, to be reasonable, in composite they lead to the dangerous conclusion that patients should be denied the evidence that protects them against prejudice based medicine.

    It is remarkable that a piece apparently arguing from the viewpoint of the Reverend Bayes omits the key feature of bayesian inference—that …

    View Full Text

    Log in

    Log in through your institution


    * For online subscription