CENTRAL contains all records from MEDLINE indexed with the Publication Type term ‘Randomized Controlled Trial’ or ‘Controlled Clinical Trial’ that are indexed as human studies. These records are downloaded quarterly from MEDLINE by Wiley-Blackwell as part of the build of CENTRAL for publication in The Cochrane Library. For further details see:
o www3.interscience.wiley.com/cgi-bin/mrwhome/106568753/CENTRALHelpFile.html
A substantial proportion of the MEDLINE records coded ‘Randomized Controlled Trial’ or ‘Controlled Clinical Trial’ in the Publication Type field have been coded as a result of the work of The Cochrane Collaboration (Dickersin 2002). Handsearch results from Cochrane entities, for journals indexed in MEDLINE, have been sent to the US National Library of Medicine (NLM), where the MEDLINE records have been re-tagged with the publication types ‘Randomized Controlled Trial’ or ‘Controlled Clinical Trial’ as appropriate. In addition, the US Cochrane Center (formerly the New England Cochrane Center, Providence Office and the Baltimore Cochrane Center) and the UK Cochrane Centre have conducted an electronic search of MEDLINE from 1966–2004 to identify reports of randomized controlled trials, identifiable from the MEDLINE titles and/or abstracts, not already indexed as such, using the first two phases of the Cochrane Highly Sensitive Search Strategy first published in 1994 (Dickersin 1994) and subsequently updated and included in the Handbook. The free text terms used were: clinical trial; (singl$ OR doubl$ OR trebl$ OR tripl$) AND (mask$ OR blind$); placebo$; random$. The $ sign indicates the use of a truncation symbol. The following subject index terms (MeSH) used were exploded: randomized controlled trials; random allocation; double-blind method; single-blind method; clinical trials; placebos. The following subject heading (MeSH) was used unexploded: research design. The Publication Type terms used were: randomized controlled trial; controlled clinical trial; clinical trial.
A test was carried out using the terms in phase three of the 1994 Cochrane Highly Sensitive Search Strategy but the precision of those terms, having already searched on all the terms in phases one and two as listed above, was considered to be too low to warrant using these terms for the above project (Lefebvre 2001). It was, however, recognized that some of these terms might be useful when combined with subject terms to identify studies for some specific reviews (Eisinga 2007).
The above search was limited to humans. The following years were completed by the US Cochrane Center (1966–1984; 1998–2004) and by the UK Cochrane Centre (1985–1997). The results have been forwarded to the NLM and re-tagged in MEDLINE and are thus included in CENTRAL. This project is currently on hold. If the US Cochrane Center can attract funding for this project they will continue the electronic search of records entered into MEDLINE in 2005 and beyond. Any updates to this situation will be described in the CENTRAL Creation Details file in The Cochrane Library:
o www3.interscience.wiley.com/cgi-bin/mrwhome/106568753/CENTRALHelpFile.html
CENTRAL includes from MEDLINE not only reports of trials that meet the more restrictive Cochrane definition for a controlled clinical trial (Box 6.3.a) but also trial reports that meet the less restrictive original NLM definition (Box 6.3.b), which used to include historical comparisons. There is currently no method of distinguishing, either in CENTRAL or in MEDLINE, which of these records meet the more restrictive Cochrane definition, as they are all indexed with the Publication Type term ‘Controlled Clinical Trial’.