临床试验期中分析中不同样本空间排序方法矫正效应高估的比较  

Comparison of different sample space ordering method in overestimation of efficacy correction during clinical trial interim analysis

作　　者：李恒 杨永利[1] 贾晓灿[1] 杨超君 施学忠[1] LI Heng;YANG Yongli;JIA Xiaocan;YANG Chaojun;SHI Xuezhong(Department of Computer and Health Statistics,College of Public Health,Zhengzhou University,Zhengzhou 450001)

机构地区：[1]郑州大学公共卫生学院计算机与卫生统计学教研室,郑州450001

出　　处：《郑州大学学报（医学版）》2024年第2期171-177,共7页Journal of Zhengzhou University(Medical Sciences)

基　　金：郑州大学青年人才企业合作创新团队支持计划(32310410);校级教育教学改革研究与实践重点项目(2022ZZUJG063)。

摘　　要：目的:比较逐段(SW)排序法、似然比(LR)排序法和极大似然估计值(MLE)排序法对临床试验中成组序贯设计期中分析效应高估的矫正效果。方法:使用SAS 9.4生成正态分布资料模拟数据集,模拟期中分析,采用中位无偏估计值(MUE)及其95%CI下限评估3种样本空间排序法的矫正效果。采用一项已上市的随机对照临床试验进行验证。结果:与SW排序法相比,LR排序法和MLE排序法的MUE及其95%CI下限更接近真实值。结论:对于正态分布数据,LR和MLE排序法能更好地矫正临床试验期中分析的效应高估。Aim:To compare the performance of stagewise(SW)ordering,likelihood ratio(LR)ordering,and maximum likelihood estimator(MLE)ordering in correcting overestimation of efficacy in different scenarios in group sequential design interim analysis.Methods:Normal distributed quantitative simulation datasets were generated using SAS 9.4.After the simulated interim analysis,the median unbiased estimator(MUE)and its lower limit of 95%confidence interval(CI)were used to assess the performance of SW ordering,LR ordering and MLE ordering.Finally,a randomized controlled trial was used as a validation.Results:Compared with SW ordering,the MUE and its lower limit of 95%CI of LR ordering and MLE ordering were closer to the true intergroup effect.Conclusion:LR ordering and MLE ordering could obtain better correction efficacy than SW ordering during clinical trial interim analysis in normally distributed data.

关 键 词：临床试验 期中分析 样本空间排序 效应高估矫正 

分 类 号：R195.1[医药卫生—卫生统计学]