Endgames Statistical Question

Clinical trials: units of randomisation

BMJ 2014; 348 doi: https://doi.org/10.1136/bmj.g3297 (Published 16 May 2014) Cite this as: BMJ 2014;348:g3297
  1. Philip Sedgwick, reader in medical statistics and medical education
  1. 1Centre for Medical and Healthcare Education, St George’s, University of London, London, UK
  1. p.sedgwick{at}sgul.ac.uk

Researchers evaluated the efficacy of a short course of oral prednisolone for acute asthma in children of school age. A randomised double blind placebo controlled crossover trial study design was used. The intervention was parent initiated treatment with the corticosteroid prednisolone (1 mg/kg/day) or placebo when an episode of acute asthma occurred. For each episode, treatment with prednisolone or placebo was chosen at random. Participants were 230 children aged 5-12 years with a history of recurrent episodes of acute asthma. Each child could contribute a maximum of eight episodes of acute asthma to the data collection over a three year study period.1

The primary outcome measure was the mean daytime symptom score over seven days according to a paediatric asthma diary. The daytime symptom score measured the children’s perspectives about their asthma symptoms and how they were limited by their symptoms. The participants contributed a total of 308 episodes of asthma that needed parent initiated treatment: 155 episodes were treated with prednisolone and 153 with placebo. The mean daytime symptom score was 15% lower in episodes treated with prednisolone than in those treated with placebo (P=0.023).

The …

View Full Text

Log in

Log in through your institution


* For online subscription