我国“系统评价/Meta分析”冠名学位论文文献检索情况分析

Retrieval Status of Master and Doctoral Dissertations Entitled Systematic Review/Metaanalysis in China

目的分析我国以"系统评价/Meta分析"冠名的学位论文文献检索情况。方法计算机检索中国学位论文全文数据库,检索时限从1980年至2015年3月,查找以"系统评价/Meta分析"冠名的学位论文,提取其基本资料和检索数据库名称、检索策略和文献筛选方法等资料后,采用Stata 12.0进行统计分析。结果最终纳入1 639篇学位论文,其中1 295篇(79.0%)学位论文同时检索了中、英文数据库,107篇(6.5%)只检索了中文数据库,237篇(14.5%)只检索了英文数据库;1 137篇(69.4%)检索了2个及以上中文数据库,1 316篇(80.3%)检索了2个及以上英文数据库。检索频率较高的中文数据库依次是CNKI、VIP、CBM、WanFang Data,英文数据库依次是PubMed、The Cochrane Library、EMbase和Ovid。89.9%的学位论文报告了检索起止时间,73.4%进行了手工检索,68.6%对参考文献进行了追溯,23.3%对会议文献进行了追溯,23.3%检索了搜索引擎,26.8%联系了原文作者/机构。45.8%的学位论文详细报告了文献的筛选步骤,37.5%报告了文献筛选流程图,仅有14.9%报告了每个数据库的检出文献数量。结论目前我国以"系统评价/Meta分析"冠名学位论文对数据库检索仍不够全面。建议今后系统评价/Meta分析研究者应至少检索3种常用数据库,同时加强追溯参考文献、会议文献和搜索引擎使用,完善检索策略和文献筛选等方面的信息。

Objective To analyze the retrieval status of master and doctoral dissertations entitled systematic review/ meta-analysis （SR/MA） in China. Methods The China Dissertation Database was searched for master and doctoral dissertations entitled SR/MA from 1980 to March 2015. Two reviewers independently extracted the basic characteristics of included dissertations, the name of retrieval databases, retrieval strategies and the method of screening literature. Then related data were analysed by Stata 12.0 software. Results A total of i 639 master and doctoral dissertations entitled SR/ MA were included for data analysis. Among them, 1 259 （79.5%） retrieved both Chinese databases and English databases, 107 （6.5%） retrieved Chinese databases alone, and the other 237 （14.5%） retrieved English databases alone. 1 137 （69.4%） retrieved two or more Chinese databases, while 1 316 （80.3%） retrieved two or more English databases. The most commonly retrieved Chinese databases were CNKI, VIP, CBM, and WanFang Data; while the most frequently retrieved English databases were PubMed, The Cochrane Library, EMbase and Ovid. 89.9% reported the duration of search time, 73.4% reported manual retrieval, 68.6% reported reference retrieval, 23.3% reported conference retrieval, 23.3% reported using search engine and 26.8% reported contact with experts and authors. 45.8% reported their steps of screening papers, 37.5% provided a flow screening chart, and only 14.9% reported the number of papers from each database. Conclusion The current study suggests that the retrieved databases of master and doctoral dissertations entitled SR/MA in China are still not enough. We suggest the authors of dissertations on SR/MA retrieve at least 3 common databases,reference, engines, conference paper at the same time, and improve information on search strategy and paper screening methods.