非虚假设临床试验引论  

An introduction to the clinical trial under non-null hypothesis

作　　者：赵国龙[1] 孙爱民[1] 陈洪涛[1] 巴秋菊[1] 汲振余[1] 赵翠林[1] 刘欣[1] 

机构地区：[1]河南省医学科学研究所,河南郑州450052

出　　处：《河南医学研究》2005年第1期22-27,34,共7页Henan Medical Research

摘　　要：目的 :提出非虚假设临床试验的基本原理和方法。方法 :将临床试验的每个中心看作一个层 ,以层样本分数为权作加权平均得综合治疗 对照差。将其与最小可识别差量比较以建立非虚假设及其对立假设 ,由此推导非虚假设分层设计基本关系式 ,进而得出非虚假设临床试验所需样本量和检验统计量 ,以抽样试验展示这种临床试验的观测功效。结果 :这些在最小可识别差量取零时还原为虚假设临床试验所需样本量和检验统计量 ,其观测功效和预定功效吻合。结论 :这种临床试验可用于建立试药对有效对照的临床优效性或非劣效性 ,为临床决策提供可靠依据。与经典的虚假设临床试验相比 ,这是一个明显的进步。附有实例描述用法。Objective: To proposes the basic principles and methods of the clinical trial under non_null hypothesis.Methods: Each center in clinical trials is regarded as a stratum.Taking the stratum sample fraction as a weight and making weighted averages yield the summary treatment_control difference.Comparing it with the minimal detectable difference leads to establish the non_null hypothesis versus the alternative.Hence we derive the equations of basic relationship for stratified designs under non_null hypothesis.It further yields the formulae for sample size and the test statistic.The observed power of such trials is exhibited by sampling experiments.Results: They reduce to the formulae for sample size and the test statistic of the classical clinical trial under null hypothesis setting the minimal detectable difference equal to zero.The observed power coincides with the prescribed power.Conclusion: This kind of clinical trials can be applied to establish the clinical superiority or non_inferiority of the tested drug versus active control and to provide reliable evidences for clinical decisions.It is a remarkable progress by contrast with the classical clinical trials under null hypothesis.A worked example illustrates the methodology.

关 键 词：临床试验 对照 临床决策 吻合 试药 药对 综合治疗 假设 差量 虚假 

分 类 号：R282[医药卫生—中药学] G633[医药卫生—中医学]