摘要
並列摘要
To test the difference of carryover effects, and Grizzle’s two-stage procedure is based on the AB/BA two-period crossover design and linear mixed-effects model with its t-test and analysis of variance. However, carryover effects, group factors and the interactions between treatment factors and period factors cannot be distinguished in the model of two-period crossover design because of alias effect. For the purpose of singling carryover effects out and analysis, building the blocking and mixed-effect model, which eliminated period factors and added dummy variables. Finally, the results of the analysis are obtained from the data of clinical trials using Aspirin with the formulas, comparing the difference between the linear mixed-effects model and both blocking and mixed-effects model the statistical software. And they will be discussed and concluded from using R and SAS.