2006～2011年我国中医药护理临床试验的方法学质量评价  

Methodological Evaluation on Domestic Clinical Trials on Traditional Chinese Medicine Nursing from 2006 to 2011

作　　者：梅丽娟[1] 郑国华[2] 陈庆月[1] 林润[1] 颜钰[3] 杨志红[3] 

机构地区：[1]福建中医药大学护理学院,福州350122 [2]福建中医药大学中西医结合研究院,福州350122 [3]福建中医药大学药学院,福州350122

出　　处：《中国循证医学杂志》2012年第6期735-739,共5页Chinese Journal of Evidence-based Medicine

摘　　要：目的评价近6年来我国中医药护理临床试验的方法学质量。方法计算机检索CNKI、VIP、WanFang Data和CBM disc数据库中发表的中医药护理的临床试验文献,检索时限为2006年1月～2011年9月。同时手检2010年1月至2011年9月国内主要护理期刊,采用Cochrane偏倚风险评估表进行临床试验文献质量评估。结果共纳入临床对照试验854篇,其中随机对照试验706篇(82.7%),半随机对照试验108篇(12.6%),非随机对照试验40篇(4.7%)。方法学质量分析显示:91.8%的纳入研究(784篇)均描述了组间基线资料可比性,并有明确的诊断标准(498篇,58.3%)和排除标准(178篇,20.8%);97.3%的纳入研究(831篇)统计方法运用正确。但只有55篇(6.4%)报告了随机序列的产生方法;10篇(1.2%)阐述随机序列隐匿方法;22篇(2.6%)采用了盲法;98篇(11.5%)报告进行了随访,93篇(10.9%)报告了安全性,20篇(2.3%)报告了失访及退出情况,2篇进行了意向性分析;所有研究均未报告研究方案,无法判断是否存在选择性报告偏倚,但21篇(2.5%)试验结果中的结局指标明显少于方法中的结局指标数,可判断为存在偏倚。符合Cochrane低风险标准2条及以上的研究共81篇,其中2009年10篇(12.3%)、2010年26篇(32.1%)、2011年前9个月27篇(33.3%),呈上升趋势。结论按Cochrane偏倚风险评估标准,近6年来,中医药护理临床试验研究总体质量不高,均存在不同程度的缺陷,但研究质量有逐年增高的趋势。Objective To evaluate the methodological quality of clinical trials on traditional Chinese medicine （TCM） nursing in recent six years. Methods Such databases as CNKI, VIP, WanFang Data and CBM were searched for collecting clinical trials on TCM nursing published from January 2006 to September 2011, and domestic primary nursing journals were also searched from January 2010 and September 2011. Methodological quality of included studies was as- sessed using quality assessment criteria of the Cochrane systematic review guideline. Results A total of 854 clinical trials were retrieved, including 706 （82.7%） randomized controlled trials （RCTs）, 108 （12.6%） quasi-randomized controlled tri- als and 40 （4.7%） non-randomized controlled trials. In the methodological quality analysis, the comparability of baseline was mentioned in 784 trials （91.8%）, a total of 498 （58.3%） reported definite diagnosis criteria. 178 （20.8%） reported exclu- sive criteria. 831 studies （97.3%） applied relevant statistical methods properly. However, only 55 trials （6.4%） mentioned the method of randomization sequence. 10 studies （1.2%） described the method of randomiztion assignment. Blinding was mentioned in 22 studies （2.6%）. 98 trials （11.5%） did prospective follow-up. 93 trials （10.9%） had safety description. 20 trials （2.3%） reported lost and with drawl cases, but only 2 conducted intention-to-treat analysis. It was hard to deter- mine whether there was selective reporting bias or not because all the studies did not have protocols. Only 21 studies （2.5%） mentioned the lack of outcome indicators which could be the evidence for existing of bias. By annual analysis, there were 81 trials which conformed to at least 2 low risk criteria. 10 trials （12.3%） was published in 2009, 26 trials （32.1%） published in 2010, and 27 trials published by September 2011, indicated an uptrend. Conclusions According to the Cochrane Col- laboration＇s tool for assessing risk of bias, the overall

关 键 词：中医药 护理 临床试验 质量评价 

分 类 号：R248[医药卫生—中医临床基础]