High performance of Cochrane’s search filters to identify RCTs in MEDLINE

Re-posted on 13 October 2020, by Froeks Kamminga (Methods Liaison Officer) from the Cochrane Community web pages)

The efficient retrieval of evidence from randomized controlled trials (RCTs) is core to Cochrane's work. Cochrane developed its own search filters to identify reports of RCTs in 1994 (updated most recently in 2008). These search filters, published in the Cochrane Handbook for Systematic Reviews of Interventions, are widely used across Cochrane and by other systematic reviewers, and also to identify trial records for CENTRAL, Cochrane’s database of RCTs.

As Cochrane’s RCT filters have not been validated for many years and newer filters are now available, Cochrane commissioned York Health Economics Consortium (YHEC) information specialists, to test available RCT filters for MEDLINE. YHEC’s study found that, even though there are slightly more sensitive filters available, the Cochrane sensitive filter offers better relative precision (0.14) with very high sensitivity (0.96) – meaning it still performs very well for finding RCTs in systematic reviews and can continue to be used to capture RCTs for CENTRAL.

For more information, please see Julie Glanville’s publication in the Journal of the Medical Library Association:  

Glanville J, Kotas E, Featherstone R, Dooley G. Which are the most sensitive search filters to identify randomized controlled trials in MEDLINE? Journal of the Medical Library Association: JMLA. 2020 Oct 1;108(4):556. http://jmla.mlanet.org/ojs/jmla/article/view/912