发表于《循证医学》的干预类系统评价/Meta分析的方法学质量分析  

Methodology assessment of papers on systematic reviews and meta-analysis of interventions published in Journal of Evidence-Based Medicine

作　　者：石新彤[1,2] 葛龙[1,2] 安妮[1,2] 周为文[1,3] 徐俊峰[1,2] 马继春[1] 田金徽[1] 

机构地区：[1]兰州大学循证医学中心,甘肃兰州730000 [2]兰州大学第一临床医学院,甘肃兰州730000 [3]兰州大学第二临床医学院,甘肃兰州730000

出　　处：《中华医学图书情报杂志》2013年第12期50-54,共5页Chinese Journal of Medical Library and Information Science

摘　　要：目的:评价发表在《循证医学》期刊上的干预类系统评价/Meta分析(SR/MA)的方法学质量。方法:手工检索《循证医学》杂志2001-2011年发表的所有干预类系统评价/Meta分析文献,共纳入70篇,提取文献的基本信息,采用AMSTAR量表评价文献的方法学质量,将数据输入Excel,用SPSS17.0和Meta Analyst软件进行统计分析。结果:质量评价标准以Cochrane偏倚风险评估工具为主(34篇,48.6%),AMSTAR的发布、基金资助、作者数量、作者单位性质和数量,对纳入系统评价/Meta分析的方法学质量总评分影响不大,差异不具统计学意义。结论:《循证医学》杂志发表的干预类系统评价/Meta分析方法学质量中等,存在前期研究设计方案缺失、检索策略不够完善、未提供排除文献列表和未交待利益冲突等问题。Objective To assess the methodology of papers on systematic reviews and meta-analysis of interventions published in Journal of Evidence-Based Mecdicine . Methods Basic data were extracted from 70 papers on systematic reviews and raeta-analysis of interventions published in Journal of Evidence-Based Medicine from 2001 to 2011. Methodology used in these papers was assessed according to the AMSTAR Scale. The data were input into the Excel and analyzed using the SPSS7.0 and Meta-Analyst software. Results The methodology used in 34 papers （48.6%） was assessed using the Cochrane bias risk assessment tools. Fund support, number of authors and their affiliated institutions did not significantly affect the total score of methodology used in the 70 papers. Conclusion The methodology used in papers on systematic reviews and recta-analysis of interventions published in Journal of Evidence-Based Medicine is not quite valid as its early design program and retrieval strategies are imperfect, and it does not provide the excluded literature list and the interest conflict.

关 键 词：干预类 系统评价 META分析 AMSTAR 循证医学 

分 类 号：G250.252[文化科学—图书馆学] R4[医药卫生—临床医学]