Intended for healthcare professionals

Endgames Statistical Question

Relative risks and confidence intervals

BMJ 2010; 341 doi: (Published 04 August 2010) Cite this as: BMJ 2010;341:c4166
  1. Philip Sedgwick, senior lecturer in medical statistics1,
  2. Louise Marston, research statistician2
  1. 1Centre for Medical and Healthcare Education, St George’s, University of London, Tooting, London
  2. 2Department of Primary Care and Population Health, University College London, London
  1. Correspondence to: P Sedgwick p.sedgwick{at}

    Researchers investigated if a standardised programme of combined manual therapy and exercise improved shoulder pain and function in people with chronic rotator cuff disease.1 A randomised, placebo controlled trial of 22 weeks duration was used. The placebo comprised inactive ultrasound therapy and application of an inert gel. A total of 120 participants were recruited.

    One of the primary outcomes was participants’ global rating of overall change. At 11 weeks, the relative risk of a self reported successful outcome (defined as “much better”) for the standardised programme compared with placebo was 1.43 (95% confidence interval 0.87 to 2.34).

    Which of the following, if any, are true for the self reported change at 11 weeks?

    • a) The sample relative risk is an estimate of the population parameter

    • b) The …

    View Full Text

    Log in

    Log in through your institution


    * For online subscription