Endgames Statistical Question

What is number needed to harm (NNH)?

BMJ 2013; 347 doi: http://dx.doi.org/10.1136/bmj.f4869 (Published 02 August 2013) Cite this as: BMJ 2013;347:f4869
  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 investigated whether 10 mg, 25 mg, or 50 mg metoclopramide combined with 8 mg dexamethasone, given intraoperatively, was more effective in preventing postoperative nausea and vomiting than 8 mg dexamethasone alone. A four armed, parallel group, double blind, randomised controlled trial was performed. The intervention consisted of 8 mg dexamethasone alone (no metoclopramide) or 8 mg dexamethasone plus 10 mg, 25 mg, or 50 mg metoclopramide. Participants were 3140 patients who received balanced or regional anaesthesia during surgery.1

The main outcome measures were postoperative nausea and vomiting within 24 hours of surgery, plus occurrence of adverse drug reactions. The table shows the proportion of patients experiencing postoperative nausea and vomiting plus number needed to treat for each treatment group. The most common adverse drug reactions were hypotension and tachycardia; the table also shows the proportion of patients experiencing these adverse events plus number needed to harm for each treatment group.

View this table:

Association between dose of metoclopramide and nausea and vomiting, plus adverse events (hypotension and tachycardia)

The researchers concluded that the addition of 50 mg metoclopramide to 8 mg dexamethasone (given intraoperatively) is an effective, safe, and cheap way to prevent postoperative nausea and vomiting. It was suggested that a reduced dose of 25 mg metoclopramide intraoperatively, with additional postoperative prophylaxis in high risk patients, might be equally effective and cause fewer adverse drug reactions.

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

  • a) Number needed to harm is a measure of the harm (adverse drug reactions) caused by the intervention when compared with 8 mg dexamethasone alone

  • b) It is estimated that on average, for every 11 patients given 50 mg metoclopramide one would experience an adverse drug reaction

  • c) The larger the value of the number needed to harm, the lower the risk of adverse …

Sign in

Free trial

Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
Sign up for a free trial

Subscribe