Search tips
Search criteria 


Logo of bmjThis ArticleThe BMJ
BMJ. 2007 July 28; 335(7612): 178.
PMCID: PMC1934467
Shortcuts from Other Journals

Systematic reviews don't last for ever

Systematic reviews and meta-analyses are powerful tools for informing clinical decisions. But they soon begin to show their age. Researchers recently estimated that about a quarter of reviews (23%, 95% CI 15% to 33%) need updating within two years of publication, and 15% (9% to 24%) need updating within a yearyear.

figure scut2807.f1

In a sample of 100 systematic reviews indexed in the American College of Physicians Journal Club between 1995 and 2005, 57 needed to incorporate new evidence by September 2006. It took a median of three years for these reviews to become out of date, and 7% (3% to 14%) were out of date by the time they were published. Cardiovascular reviews had a shorter survival than the rest.

The researchers deliberately chose reviews likely to be relevant to practising doctors, and all the reviews in their sample mathematically synthesised the evidence on at least one outcome. So these estimates may not apply to all published systematic reviews. It seems clear, however, that this kind of evidence has a relatively short shelf life, they say. Authors and journal editors should aim for speedy publication and update reviews first if there's a delay. Doctors should probably check for more recent evidence once reviews reach their first birthday.


  • Ann Intern Med 2007;147:224-33

Articles from The BMJ are provided here courtesy of BMJ Publishing Group