Endgames Statistical Question

Randomised controlled trials: “within subject” versus “between subject” designs

BMJ 2014; 349 doi: https://doi.org/10.1136/bmj.g6435 (Published 24 October 2014) Cite this as: BMJ 2014;349:g6435
  1. Philip Sedgwick, reader in medical statistics and medical education1
  1. 1Institute for Medical and Biomedical Education, St George’s, University of London, London, UK
  1. p.sedgwick{at}sgul.ac.uk

A study investigated the effectiveness of morphine in relieving the sensation of breathlessness in patients in whom the underlying cause was maximally treated. A randomised double blind placebo controlled crossover (“within subjects”) design was used. The intervention was four days of 20 mg sustained release oral morphine or corresponding placebo. No washout period was included. Laxatives were provided as needed. Participants were 48 patients who had not previously been treated with opioids. Most of the patients had chronic obstructive pulmonary disease (42; 88%).1

The main outcome measures were dyspnoea, as measured in the morning and evening on a 100 mm visual analogue scale, plus quality of sleep and side effects. Outcome measures were recorded at the end of each four day treatment period. Participants reported significantly different dyspnoea scores when treated with morphine: an improvement of 6.6 mm (95% confidence interval 1.6 to 11.6; P=0.011) in the morning and 9.5 mm (3.0 to 16.1; P=0.006) in the evening. Participants also reported significantly better sleep during the period when taking morphine (P=0.039). Significantly more participants reported distressing constipation while taking morphine (P=0.021) despite the use of laxatives. It was concluded that sustained release oral morphine at low dosage provides significant symptomatic improvement in refractory dyspnoea.

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

  • a) All trial participants acted as their own control

  • b) The crossover trial study …

View Full Text

Sign in

Log in through your institution

Subscribe