作者: Pamela Royle , Ruairidh Milne
DOI: 10.1017/S0266462303000552
关键词:
摘要: OBJECTIVES To analyze sources searched in Cochrane reviews, to determine the proportion of trials included reviews that are indexed major databases, and compare quality these with those from other sources. METHODS All new systematic Library, Issue1 2001, were restricted randomized controlled (RCTs) or quasi-RCTs selected. The recorded, checked see whether they four databases. Trials not how could be identified. found databases was compared RESULTS range number per review ranged between one twenty-seven. Controlled Register = 78.5%, MEDLINE 68.8%, Embase 65.0%, Science/Social Sciences Citation Index 60.7%. Searching another twenty-six after (CCTR), MEDLINE, only 2.4% additional trials. There no significant difference CCTR, trials, respect adequate allocation concealment sample size. CONCLUSIONS a large variation exhaustiveness literature searches. CCTR single best source RCTs. Additional database searching retrieved small percentage extra Contacting authors manufacturers find unpublished appeared more effective method obtaining better