Intended for healthcare professionals

Endgames Statistical Question

Meta-analyses: standardised mean differences

BMJ 2013; 347 doi: (Published 06 December 2013) Cite this as: BMJ 2013;347:f7257
  1. Philip Sedgwick, reader in medical statistics and medical education1,
  2. Louise Marston, senior research statistician2
  1. 1Centre for Medical and Healthcare Education, St George’s, University of London, London, UK
  2. 2University College London, London, UK
  1. p.sedgwick{at}

Researchers undertook a meta-analysis of the effects of local anaesthesia for pain control during hysteroscopy. Randomised controlled trials were included if they compared local anaesthesia with no intervention, placebo, oral analgesics, or conscious sedation. Participants were women undergoing diagnostic or operative hysteroscopy as outpatients without general anaesthesia. The primary outcome was pain associated with the procedure.1

In total, 15 trials were included. Four methods of administration of local anaesthesia were identified—intracervical, paracervical, and transcervical injections plus topical application. The trials used different scales to assess the pain associated with the procedure, including continuous visual analogue scales and numerical scales. The standardised mean difference in pain between treatment groups (local anaesthesia minus control) was derived for each trial. The results of the meta-analysis were presented in a forest plot, with those for the subgroup of paracervical injections shown (fig 1).

Fig 1 Effects of local anaesthesia compared with control treatment on pain during outpatient hysteroscopy for the subgroup of administration by paracervical injection

Which of the following statements, if any, are true for the subgroup of paracervical injection of local anaesthesia?

  • a) For each trial, the standard error of the mean difference was used to calculate the standardised mean difference

  • b) For each trial, the standardised mean difference was on the same scale as the original measurement of pain

  • c) The standardised mean differences allowed a direct comparison of treatment effects across …

View Full Text

Log in

Log in through your institution


* For online subscription