Letters

Non-randomised studies cannot be ignored

BMJ 1994; 309 doi: https://doi.org/10.1136/bmj.309.6960.1016b (Published 15 October 1994) Cite this as: BMJ 1994;309:1016
  1. J G Thornton,
  2. R J Lilford
  1. Institute of Epidemiology and Health Services Research, University of Leeds, Leeds LS2 9LN

    EDITOR, - T H Bloomfield's and John F Stratton's analogy to cars, while colourful, is poor. Try ours. Imagine a successful motor racing team that claims to improve its cars' performance with a package of special fuel, oil, and tyres but also has expert drivers. Let us pretend that trials of all three components combined are impossible. Nevertheless, when other teams perform trials of the special fuel and oil separately and of both fuel and oil combined they show no clear benefits, …

    View Full Text

    Sign in

    Log in through your institution

    Free trial

    Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
    Sign up for a free trial

    Subscribe