Application Of The Longitudinal Data Analysis Model In Clinical Trials

Multiple post-treatment measurements are often collected more than two times in the clinical trials. And comparing treatment effects in terms of average change from baseline is a very common practice in these clinical trials.The analysis of covariance model (ANCOVA) is commonly used for handling multiple post-treatment measurements comparison. ANCOVA with the baseline values as the covariate and the change from baseline value as the dependent variable, which can show treatment effects for intend change from baseline adjusted by baseline value.With multiple post-treatment measurements (repeated measurements over time), the longitudinal data analysis (LDA) model is recommended to be used, in the longitu-dinal data analysis (LDA) model the baseline value is specially regarded as a dependent variable in the response vector, along with the post-treatment values.In addition, the great advantage of the LDA model does not exclude the subjects with missing baseline values and the subjects with at least one post-randomization values from analysis, which leads to the more efficient use of all the data.In this essay, we will use common statistical methods and LDA model separately to analyze the longitudinal clinical data separately; and it will demonstrate that longi-tudinal data analysis (LDA) model has a great advantage compared to other models.