Nonlinear mixed effects model in the crossover design level data analysis.doc

 PAGE \* MERGEFORMAT 14 Nonlinear mixed effects model in the crossover design level data analysis Author: Chen Wei, YANG Xiao-hong, Chen Chaoqiong, Yangxiang 【Abstract】 Objective To explore the application of nonlinear mixed models in ordered categorical data of cross-over trial analysis and hence provide methodology reference.Method An example was illustrated by modeling nonlinear mixed models using the NLMIXED command of SAS.Results All parameters and their standard error were estimated, so every factor could be intuitionisticatty interpreted.Conclusion When subjects were estimated as random effects, not only the accuracy of the estimation was improved, but also covariates, such as the center, age, gender and baseline could be taken into consideration.The method could also deal with data with missing values and without deleting these observations, it is useful in ordered categorical data of cross-over trial analysis. Keywords: nonlinear mixed models; cross-over trial; ordered categorical data Cross-over trial (cross-over trial) is designed according to pre-trial order, in various periods of the study (subject) each of the implementation of various treatments in order to compare the differences between treatment groups is to compare and between the two groups themselves integrated application design ideas for a design method [1]. It overcomes the experimental observation period before and after the self-control due to a variety of non-experimental factors on the experimental results to the impact of bias, at the same time saving sample size [2], the main source of the more difficult for individual differences in the larger animals (such as cattle, monkey, etc.) or human medicine, psychology and other tests. Can be used for clinical evaluation of the role to relieve symptoms but not cure the effect of drugs such as painkillers, sedatives, anti-rheumatism, lowering blood pressure, anti-insomnia compare the efficacy of drugs. Stati