《中华流行病学杂志》55篇Meta分析文献的质量评价

A systematic evaluation on the quality of Meta-analysis in articles published in the Chinese Journal of Epidemiology

目的评价《中华流行病学杂志》公开发表Meta分析文献的方法学质量和报告质量。方法计算机检索万方医学网，检索时间为1998年1月至2012年12月，同时辅以手工检索，查找《中华流行病学杂志》发表的Meta分析文献。由2名调查员分别按照纳入和排除标准筛查文献，并采用AMSTAR量表和PRISMA量表评估纳入文献的方法学质量和报告质量。采用SPSS13．0软件进行数据统计和分析。结果共纳入文献55篇，方法学质量评价结果显示高质量者仅2篇（3．6％），中等质量35篇（63．7％），低质量18篇（32．7％）；2008年以来发表的文献方法学质量有所改善，但仍然存在纳人和排除的研究文献清单提供不全、检索不够系统全面、纳入研究质量评价不受重视等问题。报告质量的评价结果显示纳入文献在报告题目、文献信息来源的描述、方法效应指标的概括以及Meta分析结果的综合方面质量较高，但是在摘要、资料检索和收集方法、单个研究内部偏倚评估、讨论中对证据总结、局限性及资金支持等方面报道不全面。结论《中华流行病学杂志》发表的Meta分析文献为提供更为可靠的病因／危险因素的估计提供了证据，但其方法学质量和报告质量仍需提高。

Objective To assess the methodology and quality on Meta-analysis used in papers being published in the Chinese Journal of Epidemiology. Methods Computerized literature searching was carried out in Wanfang Medical Online to collect articles that Meta-analysis was used in the Chinese Journal of Epidemiology since it was founded till December, 2012. Manual retrieval was also conducted. Two researchers independently screened for literature and extracted data. Disagreements were resolved through discussion or by resort to a third reviewer if consensus was not reached. Qualities on methodologies or on the processes of reporting and reviewing, were evaluated by both AMSTAR and PRISMA scales. Statistical calculations and analyses were performed using SPSS 13.0. Results Fifty-five papers on meta-analyses were included in this study. Results on the qualities of methodology or evaluation showed that only 2 articles （3.6%） were rated as high, 35 articles （63.7%） as moderate and 18（32.7%） as low. The quality on methodology being used in literature had improved since 2008. However, there were still some problems seen in the following areas as the list of studies （included and excluded）, comprehensive search on literature, quality of the included studies having been assessed and documented, etc. Results on evaluation of quality showed that the included reviews had high quality on the titles of the report, sources of information, summary measures and synthesis of results. However, areas as： structured summary, methods on searching, data collection, risk of bias in individual studies, summary of evidence, limitations, funding etc., were still lack of comprehensive reports. Conclusion Articles on Meta-analysis published in the Chinese Journal of Epidemiology provided substantial evidence for more reliable information on the etiology and risk factors of the studies. However, both of the qualities on methodology and reports in the included literature presented problems at different levels that called for careful improvement.