应用AMSTAR声明评价通心络疗效及安全性系统评价的文献方法学质量的报告

摘要

目的：使用AMSTAR声明评价有关通心络疗效及安全性的系统评价文献的方法学质量。方法：计算机检索包括中国期刊全文数据库（CJFD）、重庆维普中文科技期刊数据库（VIP）、中国知网数据库（CNKI）、中国生物医学文献数据库（CBM）、万方数据库（WanFang Data）和中国中医药期刊文献数据库（TCM）、Medline数据库、Pubmed数据库、ISI Web of Knowledge数据库、Web of Science数据库在内的国内外各大数据库，检索时限从建库至2014年12月20日。由5名评价者根据纳入与排除标准独立进行文献筛选，并应用AMSTAR声明清单对纳入的所有系统综述文献逐一进行评价。结果：最终纳入22篇文献。其中17篇系统评价，5篇Meta分析，所有文献在方法学及报告质量上都存在不同程度的缺陷，主要集中在方案与注册、文献的检索收集、纳入排除标准、真实性评价、数据处理及基金等几个方面；纳入的所有文献对AMSTAR声明的平均报告率为57.02%。结论：从目前平均报告率的高低可以看出，应用AMSTAR声明评价有关通心络疗效及安全性的系统评价报告的方法学质量相对偏低，影响了文章的质量，说明目前系统评价在论文报告的各个环节都存在不同程度的缺陷，这势必会影响临床证据的可靠性及临床决策的判定，提示撰写此类文章需遵照AMSTAR声明条目以提高文章报告的方法学质量。%This study was aimed to assess the methodological quality of systematic literature reviews on the efficacy and safety ofTong-Xin-Luo(TXL) using the AMSTAR statement. The literatures were searched from major domestic and international databases include the CJFD Database, VIP Database, CNKI database, CBM Database Online, WanFang Data and TCM Database, Medline, Pubmed, ISI Web of knowledge, Web of science, from the database was built to 20th December, 2014. And 5 evaluators screened the literatures independently according to the inclusion and exclusion criteria. Then, all systematic reviews were assessed using the list of AMSTAR statement. The results showed that 22 literatures were finally included, which contained 17 systematic reviews and 5 meta-analysis. In the included literatures, there were different degrees of impairments on the methodological and reporting quality, which were mainly focused on program and registration, retrieving and collecting literatures, inclusion and exclusion criteria, authentic assessment, data processing, funds, and etc. The average reporting rate of AMSTAR statement was 57.02% in all included literatures. It was concluded that the methodological quality of systematic literature reviews on the efficacy and safety of TXL was relatively low from the current level of average reported rate, which affected the quality of the article. It indicated that there were different degrees of impairments in all links of test reports in current systematic reviews, which will be bound to affect the reliability of clinical evidence and the determination of clinical decision. It showed that the methodological quality of writing of such articles should be in compliance with entries of AMSTAR statement to improve its quality.