Endgames Statistical Question

Treatment effects and placebo effects

BMJ 2015; 350 doi: https://doi.org/10.1136/bmj.h267 (Published 17 January 2015) Cite this as: BMJ 2015;350:h267
  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

Researchers investigated the effectiveness of iron supplementation on reducing fatigue in non-anaemic women with unexplained fatigue. A double blind randomised placebo controlled trial was performed. In total, 144 women aged 18-55 years were recruited from an academic primary care centre and eight general practices in western Switzerland. Women were randomised to oral ferrous sulphate (80 mg/day of elemental iron daily; n=75) or placebo (n=69) for four weeks.1

The primary outcome was self reported fatigue as measured by a 10 point visual analogue scale, ranging from 1 (no fatigue at all) to 10 (very severe fatigue). At baseline, mean fatigue values were 6.37 points in the intervention group and 6.46 points in the placebo group. The mean decrease in fatigue after four weeks was significantly greater in the intervention group than in the placebo group (1.82 v 0.85 points; difference 0.97, 95% confidence interval 0.32 to 1.62; P=0.004). It was concluded that non-anaemic women with unexplained fatigue may benefit from iron supplementation.

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

  • a) The mean change in the primary outcome for the intervention group represented the treatment response

  • b) The difference between the intervention group and placebo group in the mean change of the primary outcome represented the treatment …

View Full Text

Log in

Log in through your institution


* For online subscription