论文部分内容阅读
目的对国内医院药学人员发表的系统评价/Meta分析进行方法学和报告质量评价,了解该领域人员循证药物评价存在的质量问题,以期规范和提高研究质量。方法计算机检索CNKI、Wan Fang Data、VIP、CBM、CMCI、Pub Med、EMbase、The Cochrane Library数据库,时限均为建库到2016年3月17日。根据纳入排除标准筛选文献和提取数据,采用PRISMA和AMSTAR分别进行报告质量和方法学质量评估,采用Excel 2013和SPSS 20.0软件进行数据统计与分析。结果共纳入1 018篇系统评价/Meta分析,包括中文871篇和英文147篇。PRISMA平均分为18.41±2.84分,AMSTAR平均分为7.38±1.28分。英文文献量表评分均高于中文文献,大部分文献的报告存在一定缺陷,且为中等质量。报告质量存在的问题主要表现在结构式摘要、目的、方案与注册、补充分析与资助来源;方法学质量存在的问题主要表现在前期设计方案、考虑发表情况和纳入与排除清单。单因素分析表明,发表在量表发布后(P<0.000 1)、作者单位级别高(P<0.000 1)、发表语种为英文(P<0.000 1)和有资金支持或阐明利益冲突(P<0.000 1)的研究报告质量和方法学质量较高。Pearson分析表明,PRISMA和AMSTAR(P<0.000 1)、引用量和AMSTAR(P=0.045)之间分别具有相关性特征。结论国内医院药学领域的循证药物评价发展迅速,方法学与报告质量逐年提升,但仍存在不同程度的问题有待改进;未来应积极开展后效评价的相关研究。
OBJECTIVE To evaluate the methodology and quality of reporting of systematic reviews and meta-analyzes published by domestic pharmacies in pharmacies and to understand the quality problems in the field of evidence-based drug evaluation in order to standardize and improve the quality of research. Methods The databases of CNKI, Wan Fang Data, VIP, CBM, CMCI, Pub Med, EMbase and The Cochrane Library were searched by computer. The deadline for the database was March 17, 2016. According to the screening criteria included in the exclusion criteria and the data extraction, PRISMA and AMSTAR were used to assess the quality of reporting and methodology respectively. The data were analyzed and statistically analyzed by Excel 2013 and SPSS 20.0 software. Results A total of 1 018 systematic reviews / meta-analyzes were included, including 871 Chinese and 147 English. PRISMA averaged 18.41 ± 2.84 and AMSTAR averaged 7.38 ± 1.28. The scale of English literature scale is higher than that of Chinese literature, most of the reports of the literature have some defects and are of medium quality. The main problems in the quality of reporting are structured abstracts, objectives, programs and registrations, supplementary analyzes and sources of funding. The problems in methodological quality are mainly reflected in the previous design plans, taking into account the publication status and the inclusion and exclusion list. Univariate analysis showed that there was a high level of unit rankings (P <0.0001), published in English (P <.0001) and financially supported or clarified a conflict of interest (P < 0.000 1) The quality of research reports and methodologies are high. Pearson analysis showed that there was a correlation between PRISMA and AMSTAR (P <0.000 1), reference volume and AMSTAR (P = 0.045), respectively. Conclusions The evaluation of evidence-based medicine in the field of pharmacy in domestic hospitals has been rapidly developed. The quality of methodologies and reports has been increasing year by year. However, there are still some problems that need to be improved to some extent. In the future, relevant researches on post-evaluation should be actively carried out.