运用CONSORT 2010标准评价2012年国内主要期刊关于糖尿病随机对照试验文献报告的质量  

Quality assessment of the reports of randomized controlled trials concerning diabetes published in China in 2012 by Consolidated Standards of Reporting Trials Statement 2010

作　　者：王博[1] 刘珺[2] 宣淼[3] 周尊海[1] 宋晓敏[1] 顾蕾[1] 张玄娥[1] 唐文佳[1] 朱文[1] 蔡晓莺[1] 张秀珍[3] 

机构地区：[1]上海市杨浦区中心医院内分泌科,200090 [2]上海市第十人民医院中医科,200072 [3]上海市同济医院内分泌科,200065

出　　处：《中华内分泌代谢杂志》2014年第6期529-533,共5页Chinese Journal of Endocrinology and Metabolism

基　　金：国家自然科学基金（81001573）

摘　　要：运用临床试验报告的统一标准（CONSORT 2010）声明评价2012年国内期刊发表的随机对照试验（RCT）文献的方法学质量.以“随机,糖尿病”为自由词,搜索2012年万方数据库中RCT期刊文献,依据CONSORT 2010声明,对文献质量进行综合评价,并选择中华系列杂志及中国糖尿病杂志,与其他杂志进行对比.共检索5 269篇文献,经筛选后共纳入RCT文献2 591篇,其中纳入中华系列和中国糖尿病杂志文献70篇,其他文献2 521篇.2 591篇文献中,19篇（0.7％）文题能识别随机临床试验,2 574篇（99.3％）有结构式摘要,2 510篇（96.9％）阐述背景,2 053篇（79.2％）描述试验设计,1 936篇（74.7％）有受试者合格标准,2 588篇（99.9％）描述干预措施,10篇（0.39％）预先设定主要和次要结局指标,120篇（4.6％）阐述产生随机序列方法,0篇（0.0％）描述分配隐藏,28篇（1.1％）实施盲法,308篇（1 1.9％）描述基线资料,941篇（36.3％）描述危害和意外效应,385篇（14.9％）描述偏倚和不确定原因,2 505篇（96.7％）解释结果.与其他杂志比较,中华系列和中国糖尿病杂志在描述试验设计（x2=16.348,P＜0.01）、受试者合格标准（x2 =21.667,P＜0.01）、预先设定主要和次要结局指标（x2=53.128,P＜0.01）、产生随机序列方法（x2=187.63,P＜0.01）、受试者流程（x2=16.969,P＜0.01）、描述招募期和随访期（x2=8.938,P＜0.01）、基线资料（x2=140.67,P＜0.01）、讨论局限性（x2=8.581,P＜0.01）、可推广性（x2=146.29,P＜0.01）等主要方面报告比例高.2012年国内发表的关于糖尿病方面RCT文献报告,中华系列和中国糖尿病杂志虽然在多个方面较其他杂志报道比例高,但与CONSORT 2010声明标准差距仍较大,还需在相关期刊推广并采用CONSORT声明.To explore the quality of methodology about the reports of randomized controlled trials about diabetes published in China in 2012 by CONSORT statement.The reporting of randomized controlled trials related to diabetes in Chinese academic journals were searched through WANFANG full-text database.The quality of randomized controlled trials was assessed according to the CONSORT 2010 statement.The series of magazines of Chinese Medical Association and Chinese Journal of diabetes of Chinese Medical Association were compared with other magazines.After screening,2 591 articles among 5 269 articles were included,including 70 articles published in series of magazines of Chinese Medical Association and Chinese Journal of diabetes,and the other 2 521 articles.In 2 591 RCTs,19 （0.7％） RCTs could be identified as a randomized trial in the title,2 574 （99.3％） RCTs had structured abstract,2 510 （96.9％） RCTs described the background,2 053 （79.2％） RCTs described the experimental design,1 936 （74.7％） RCTs described subjects eligibility criteria,2 588 （99.9％） RCTs described interventions,10 （0.39％） RCTs pre-set the primary and secondary outcome measures,120 （4.6％） RCTs elaborated generate a random sequence,0 （0.0％） RCTs described allocation concealment,28 （1.1％） RCTs implementation blinded,308 （1 1.9％） of baseline information described in a table,941 （36.3％） RCTs described the hazards and accidental effects,385 （14.9％） RCTs described limitations,2 505 （96.7％） interpreted the results.Compared with other magazines,there was higher proportion of reports in the description of design of experiment （x2 =16.348,P＜0.01）,eligible subjects standard （x2 =21.667,P＜0.01）,set the primary and secondary outcome indicators （x2 =53.128,P＜0.01）,random sequence generation method （x2 =187.63,P＜ 0.01）,subjects process （x2 =16.969,P＜0.01）,description of recruitment period and the follow-up period （x2 =8.938,p =0.001）,baseline data （x2 =14

关 键 词：随机对照试验 糖尿病 CONSORT声明 

分 类 号：R587.1[医药卫生—内分泌]