非虚假设多中心配对临床试验

目的:将配对设计加以扩展,以适应非虚假设和分层。方法:将多中心临床试验的每个中心看作1层,取层样本分数为权计算平均治疗-对照差。将其期望与最小可识别差量比较建立非虚假设,按其方差构造分层配对设计基本关系式,进而推导出样本量公式,检验统计量,和功效函数。将这些用于临床试验的设计、执行和分析。以Monte Carlo方法展示观测功效。结果:这些在最小可识别差量取零时还原为对应经典统计学方法,其观测功效和期望功效吻合,所需样本量小于对应的两组设计。结论:这种临床试验直观高效,适于建立试药对于对照药的临床优效性、非劣效性或等效性,并附有实例描述用法。

The Multi-Center Matched-Pair Clinical Trial under Non-Null Hypothesis

Objective:The matched-pair design is generalized to the situation with non-null hypothesis and stratification.Methods:Each center in multi-center clinical trials is regarded as a stratum.The mean treatment-control difference is obtained by taking the stratum sample fraction as a weight.Comparing its expectation with the minimal detectable difference leads to set up a non-null hypothesis.Its variance is used to construct the equation of the basic relationship for stratified matched-pair designs under non-null hypothesis.Then follow derivations for the formula of sample size,test statistics,and power funcioun.These are used in design,executive,and analysis of clinical trials.The observed power is demonstrated by the Monte Carlo method.Results:These reduce to the relevant classical statistical procedures setting the minimal detectable difference equal to zero.The observed power coincides with the expected power.The required sample size is less than that of corresponding two-sample design.Conclusion:Such a.clinical trial is intuitive and powerful.It is appropriate to establishing the clinical superiority,non-inferiority,or equivalence of tested drug versus control.An example illustrates the methodology.