Intended for healthcare professionals

Endgames Statistical Question

Sources of bias in randomised controlled trials

BMJ 2010; 341 doi: (Published 08 December 2010) Cite this as: BMJ 2010;341:c6995
  1. Philip Sedgwick, senior lecturer in medical statistics
  1. 1Section of Medical and Healthcare Education, St George’s, University of London, Tooting, London, UK
  1. p.sedgwick{at}

Last week’s question described a cluster randomised controlled trial that assessed the long term effects of an obesity prevention programme in schools.1 The intervention was delivered over one school year, and it included education focused on promoting a healthy diet and discouraging consumption of carbonated drinks. The control group received no intervention.2

In total, 644 children aged 7-11 years from six schools were recruited and followed for three years after randomisation. The main outcomes were anthropometric measures, such as height, weight, and waist circumference. All of these measurements were taken by a single investigator not blinded to treatment allocation.

Which of the following statements, if any, are true?

  • a) Individual children were randomised to the intervention or control

  • b) Children were likely to display the Hawthorne effect

  • c) Lack of blinding to the allocated treatment might have resulted in resentful demoralisation

  • d) Lack of blinding to the allocated treatment might have resulted in dilution bias


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

In a …

View Full Text

Log in

Log in through your institution


* For online subscription