STUDY OBJECTIVES: The Cochrane Database of Systematic Reviews (CDSR), one of the databases in the Cochrane Library, is a compilation of regularly updated systematic reviews with meta-analytic summary statistics. We conducted a study of the CDSR to evaluate the relevance of this database to emergency medical practice. METHODS: Using criteria that were determined a priori, 2 reviewers independently screened the titles of 795 completed reviews from the April 2000 CDSR for possible relevance to emergency medicine practice. Five reviewers independently ranked summaries of these reviews for degree of relevance. Agreement was measured using kappa statistics. Disagreements were resolved by consensus or adjudication. RESULTS: Screening of 795 completed reviews identified 136 (17%) as "possibly" and 151 (19%) as "likely" relevant (simple agreement, 74%; kappa=0.82). Further independent assessment indicated 95 (12%) of the 151 "likely" systematic reviews were directly relevant to emergency medicine practice, whereas 44 (6%) were indirectly relevant (simple agreement, 77%; kappa=0.45). Cochrane Review Groups producing the most emergency medicine-relevant systematic reviews included acute respiratory infections (14 [10%]), injuries (14 [10%]), pregnancy and childbirth (13 [10%]), stroke (12 [10%]), and airways (11 [8%]). In contrast, only 1 (0.1%) of the reviews produced by the Heart Review Group was considered directly relevant to emergency medicine practice. CONCLUSION: More than one third of CDSR reviews have some relevance, and 12% are directly relevant, to emergency medical practice or discharge planning. Although certain conditions are well covered, other key emergency medicine areas are not. Emergency physicians should consider the Cochrane Library an important evidence-based resource for emergency medicine therapeutic interventions and should examine ways of increasing the number of reviews relevant to emergency medicine.
STUDY OBJECTIVES: The Cochrane Database of Systematic Reviews (CDSR), one of the databases in the Cochrane Library, is a compilation of regularly updated systematic reviews with meta-analytic summary statistics. We conducted a study of the CDSR to evaluate the relevance of this database to emergency medical practice. METHODS: Using criteria that were determined a priori, 2 reviewers independently screened the titles of 795 completed reviews from the April 2000 CDSR for possible relevance to emergency medicine practice. Five reviewers independently ranked summaries of these reviews for degree of relevance. Agreement was measured using kappa statistics. Disagreements were resolved by consensus or adjudication. RESULTS: Screening of 795 completed reviews identified 136 (17%) as "possibly" and 151 (19%) as "likely" relevant (simple agreement, 74%; kappa=0.82). Further independent assessment indicated 95 (12%) of the 151 "likely" systematic reviews were directly relevant to emergency medicine practice, whereas 44 (6%) were indirectly relevant (simple agreement, 77%; kappa=0.45). Cochrane Review Groups producing the most emergency medicine-relevant systematic reviews included acute respiratory infections (14 [10%]), injuries (14 [10%]), pregnancy and childbirth (13 [10%]), stroke (12 [10%]), and airways (11 [8%]). In contrast, only 1 (0.1%) of the reviews produced by the Heart Review Group was considered directly relevant to emergency medicine practice. CONCLUSION: More than one third of CDSR reviews have some relevance, and 12% are directly relevant, to emergency medical practice or discharge planning. Although certain conditions are well covered, other key emergency medicine areas are not. Emergency physicians should consider the Cochrane Library an important evidence-based resource for emergency medicine therapeutic interventions and should examine ways of increasing the number of reviews relevant to emergency medicine.