A comparison of Embase randomised controlled trial filters

Tags: Poster
Martin S1, McHugh J1
1BMJ Publishing Group, UK

Background: The BMJ Evidence Centre has a transparent and rigorous methodology. Our products are based on systematic searches of Medline, Embase and the Cochrane Library for systematic reviews and randomised controlled trials. Our challenge was to develop new filters for Embase searches to increase their specificity, without losing a great deal of sensitivity, in order to increase our productivity for a variety of BMJ Evidence Centre products.

Objectives: To compare the new Embase randomised controlled trial filter used for BMJ Evidence Centre searches against other published Embase randomised controlled trial filters.

Methods: We have done an extensive literature search for all other available filters to use as comparisons and have based our methods on the relative recall method as reported by Sampson M. et al. (2006).

Results: We will report specificity and sensitivity; highlight the strengths and weaknesses of our research methods; discuss considerations and outcomes and present our conclusions.


1. Sampson M et al. An alternative to the hand searching gold standard: validating methodological search filters using relative recall. BMC Research Methodology 2006 6:33 [http://www.biomedcentral. com/1471-2288/6/33]