Background: The National Institute for Health and Care Excellence (NICE) ‘Guidelines manual’ outlines a core number of databases that should be routinely searched for identifying reports of randomised controlled trials (RCTs), including MEDLINE, Embase and CENTRAL. Current practice is to search each bibliographic database from its inception. However, the Cochrane Handbook recommends that searches of MEDLINE and Embase for trials should be undertaken in knowledge of the fact that both MEDLINE and EMBASE have been searched systematically for reports of trials and that these reports of trials have been included in CENTRAL (section 6.3.3.2).

Objectives: To estimate the accuracy and reliability of the processes used by the Cochrane Collaboration to identify reports of RCTs from MEDLINE and Embase, for CENTRAL. To validate the Cochrane Handbook approach of searching CENTRAL supplemented by searches of MEDLINE and Embase for the more recent years. A further objective is to develop a common filter(s) for identifying reports of RCTs from MEDLINE and Embase for the more recent years that optimises the balance between sensitivity and precision.

Methods: A sample of ‘included’ RCTs from NICE guidelines from each of the four National Collaborating Centres (NCCs) and Internal Clinical Guideline (ICG) team will be used to validate the Cochrane Handbook approach for identifying reports of RCTs. The sample will also be used to objectively develop RCTs filters for searching MEDLINE and Embase for the more recent years.

Results: This is an ongoing project, interim data will be presented on the number and proportion of RCTs indexed in MEDLINE and Embase which are also indexed in CENTRAL.

Conclusions: We plan to draw conclusions about the validity of the Cochrane Handbook searching approach compared to the current ‘all years’ searching approach which will inform the methods for searching for RCTs for NICE guidelines.