Misleading meta-analysisBMJ 1995; 311 doi: https://doi.org/10.1136/bmj.311.7015.1303b (Published 11 November 1995) Cite this as: BMJ 1995;311:1303
- M R Law,
- N J Wald
- Reader Professor Department of Environmental and Preventive Medicine, Wolfson Institute of Preventive Medicine, St Bartholomew's Hospital Medical College, London EC1M 6BQ
One incorrect meta-analysis does not invalidate them all
EDITOR,--Guilt by association is dangerous practice in medicine as in law, yet James Le Fanu seeks to discredit the results of all meta-analyses1 because of the incorrect result of one: small trials of intravenous magnesium in acute myocardial infarction suggested an effect not confirmed by a large trial.2 The misleading suggestion was probably due to publication bias.2 Meta-analyses are not immune to error and must be evaluated as critically as individual studies.
Le Fanu considers that …