Article type
Year
Abstract
Introduction: Systematic reviews need to be based on as high a proportion as possible of the available evidence, in order to provide reliable information on which to base health care decisions. Previous work within the Collaboration has focussed on the systematic electronic searching of MEDLINE and systematic handsearching of general and specialized health care journals. EMBASE is stronger than MEDLINE in its coverage of health care journals published in Europe and pharmaceutical journals. Identifying reports of trials in EMBASE, however, is not straightforward, as the indexing term RANDOMIZED-CONTROLLED-TRIAL was not introduced to EMBASE until 1994, and since then has not always been applied consistently.
Objective: To identify reports of controlled trials in EMBASE from inception (1974) to 1998, and make them accessible through The Cochrane Controlled Trials Register, with the permission of the database producer, Elsevier.
Methods: A highly sensitive search strategy was used to identify potential reports of controlled trials from EMBASE. Online hosts offering access to both MEDLINE and EMBASE simultaneously were used to identify and download records from EMBASE, having de-duplicated those records which were already coded as reports of trials on MEDLINE (and thus were already included in The Cochrane Controlled Trials Register). Handsearchers scanned the printed EMBASE abstracts and identified reports of controlled trials, which were checked by a second experienced scanner prior to inclusion.
Results: Thus far, approximately 70,000 EMBASE records have been downloaded. Of these, 20,000 have been fully processed. Approximately 9,000 newly identified controlled trials from EMBASE are included in The Cochrane Controlled Trials Register in the Issue 2, 1998 version of The Cochrane Library.
Discussion: Our preliminary results indicate that EMBASE is a rich source of reports of controlled trials, and that such databases need to be searched systematically for all reports of trials to ensure that The Cochrane Controlled Trials Register can be reliably considered to be the most important single source of published reports of trials.
Objective: To identify reports of controlled trials in EMBASE from inception (1974) to 1998, and make them accessible through The Cochrane Controlled Trials Register, with the permission of the database producer, Elsevier.
Methods: A highly sensitive search strategy was used to identify potential reports of controlled trials from EMBASE. Online hosts offering access to both MEDLINE and EMBASE simultaneously were used to identify and download records from EMBASE, having de-duplicated those records which were already coded as reports of trials on MEDLINE (and thus were already included in The Cochrane Controlled Trials Register). Handsearchers scanned the printed EMBASE abstracts and identified reports of controlled trials, which were checked by a second experienced scanner prior to inclusion.
Results: Thus far, approximately 70,000 EMBASE records have been downloaded. Of these, 20,000 have been fully processed. Approximately 9,000 newly identified controlled trials from EMBASE are included in The Cochrane Controlled Trials Register in the Issue 2, 1998 version of The Cochrane Library.
Discussion: Our preliminary results indicate that EMBASE is a rich source of reports of controlled trials, and that such databases need to be searched systematically for all reports of trials to ensure that The Cochrane Controlled Trials Register can be reliably considered to be the most important single source of published reports of trials.