线性回归模型在等效性、非劣效性中的应用及其SAS实现

Application of Linear Regression Model and Corresponding Program of SAS Software in Equivalence,Non-Inferiority Test

目的针对医学研究中常见的定量资料等效性检验、非劣效性检验存在的问题,探讨应用线性回归模型理论进行定量资料的等效性检验、非劣效性检验,并给予统计软件技术上的支持。方法从线性回归模型的构造出发,根据模型中参数的统计学意义,结合等效性检验、非劣效性检验原理,利用SAS统计软件包,结合实例,编写程序,并计算相应的统计量。结果分别对平行设计和多中心设计的定量资料进行统计学检验,并对主要输出结果进行了解释。结论探索出在临床研究中利用线性回归模型,通过SAS软件包进行统计学检验的途径,并为类似问题的解决提供了理论参考。

Objective To present theory and statistics software support,so as to tackle with the problems in the use of equivalence,non-inferiority and superiority test of quantitative data in medical researches.Methods According to the theory of modeling linear regression and the statistical meaning of the parameters,wrote the program in SAS software package and calculated the relative statistics.Results Equivalence and non-inferiority test were carried on for quantitative data,and the corresponding meaning of the procedure and results were explained at the same time.Conclusion Explored a right way to put forward equivalence,non-inferiority and superiority test using SAS software package,and provided a theoretical reference for similar problems.