Intended for healthcare professionals

Endgames Statistical Question

Why randomise in clinical trials?

BMJ 2012; 345 doi: (Published 22 August 2012) Cite this as: BMJ 2012;345:e5584
  1. Philip Sedgwick, reader in medical statistics and medical education
  1. 1Centre for Medical and Healthcare Education, St George’s, University of London, Tooting, London, UK
  1. p.sedgwick{at}

Researchers evaluated the efficacy of a 23-valent pneumococcal polysaccharide vaccine in preventing pneumonia in people at high risk. A randomised, placebo controlled, double blind trial was used. A total of 1006 nursing home residents in Japan were recruited. Participants were randomly allocated to 23-valent pneumococcal polysaccharide vaccine (n=502) or placebo (n=504). All participants were followed for at least 26 months.1

The primary endpoints were the incidence of all cause pneumonia and pneumococcal pneumonia. The researchers reported that all cause pneumonia and pneumococcal pneumonia were both significantly more frequent in the placebo group than in the vaccine group: the incidence per 1000 person years was 91 versus 55 (P<0.0006) and 32 versus 12 (P<0.001), respectively.

Which of the following, if any, did random allocation of participants facilitate?

  • a) Minimisation of allocation bias

  • b) Minimisation of confounding

  • c) Minimisation of selection bias

  • d) Double blinding

  • e) Minimisation of ascertainment bias


Answers a, b, d, and e are true, whereas c is false.

The nursing home residents were allocated to vaccine or placebo …

View Full Text

Log in

Log in through your institution


* For online subscription