【摘要】 目的:评价中医药治疗非小细胞肺癌(NSCLC)临床随机对照试验(RCTs)文献的质量。方法:根据Cochrane中心指南,机检和手检相结合检索1997年1月至2006年12月国内生物医学期刊发表的有关中医药治疗NSCLC的RCTs文献,按照临床流行病学原理及临床试验的统一标准(CONSORT声明),收集所有文献的信息,进行描述性分析和一致性检验。结果:共纳入RCTs文献124篇。其中37篇报告了纳入标准,占29.84%;13篇有排除标准,占10.48%;37篇有明确的中医证候分类,占29.84%;无1篇文献交代样本含量估算情况;24篇报告了随机分配方法,占19.35%;8篇有随机隐藏的简单描述,占6.45%;85篇有基线资料的描述,占68.55%;3篇运用了盲法,占2.42%;77篇有统计学方法描述,占62.10%;6篇文献涉及到知情同意,占4.84%;88篇记录了不良反应,占70.97%;38篇对治疗后病例进行随访,占30.65%。结论:中医药治疗NSCLC的RCTs质量还不够高,有待进一步提高。主要表现为随机分组方法没有说明或应用错误,随机分配隐藏没有得到足够重视,盲法使用过低,受试者的纳入排除标准未充分报告,中医证候分类亟需规范统一,组间基线资料统计不全,样本量低且没有具体的估算方法,疗效评价标准不够统一,忽视依从性、病例脱落、随访情况的分析和伦理规范。更多还原

【Abstract】 Objective:To assess the quality of the published randomized controlled trials (RCTs) of non-small cell lung cancer (NSCLC) treated with TCM therapy. Methods:Guided by the Cochrane Center guidelines, we searched the literatures in Chinese biomedicine journals from 1997 to 2006 by computer retrieval and manual retrieval. The data were extracted according to the principles of clinical epidemiology and Consolidated Standards of Reporting Trials(CONSORT), and were managed by descriptive analysis and uniformity test. Results:124 RCTs were enrolled. The inclusion and exclusion criterias were reported in 37 trials (29.84%) and 13 trials (10.48%) respectively. The type of TCM syndrome was involved in 37 trials (29.84%).None of the trials mentioned the estimation of sample size. The randomization and allocation concealment was reported in 24 trials (19.35%) and 8 trials (6.45%) respectively. The baseline data were described in 85 trials (68.55%). Blind method was mentioned in 3 trials (2.42%). Statistics method was described in 77 trials (62.10%). Informed consent was acquired in 6 trials (4.84%).Adverse events were described in 88 trails (70.97%). Follow-up visit was mentioned in 38 trials (30.65%). Conclusions:There are deficiencies in these RCTs, and it is necessary to improve the quality of clinical research in TCM. Most of the trials do not describe the randomization method or use it in a wrong way,only a few trials use the allocation concealment and blind method, most of the trials do not report the inclusion and exclusion criterias, has no description of comparability of baseline data, and none estimate the sample size, the clinical research assessment standards and syndrome differentiation are not unification, the analysis of treatment compliance, dropped-out case, follow-up visit are neglected.更多还原