Mixed Models for Repeated Measures using Categorical time Effects (MMRM)

Abstract

This chapter provides a brief framework describing the mixed model for repeated measures (MMRM) model and the logistic generalized linear mixed model (GLMM) for binary data, and shows detailed examples of each. It gives a brief introduction of the mathematical mechanics of the mixed linear model and the MMRM as a special case. Before beginning an analysis, it is important to understand the amount and pattern of missing data through relevant descriptive displays. The chapter presents an example to specify the MMRM analysis, show the SAS code to produce the results, review key SAS outputs from the model, and present and interpret the results. The GLMM model takes into account the correlation between the visits for a subject through an estimated variance-covariance pattern.