Finite mixture of nonlinear mixed-effects joint models in the presence of missing and mismeasured covariate, with application to AIDS studies

Abstract

It is a common practice to analyze longitudinal data using nonlinear mixed-effects (NLME) models. However, the following issues may standout. (i) Individuals may be possibly from a heterogeneous population following more than one mean trajectories, while a homogeneous population assumption for model structure may be unrealistically obscuring important features of between- and within-subject variations; (ii) some covariates may be missing and/or measured with errors. There has been few studies concerning both population heterogeneity and covariates measured with errors and missing data features simultaneously in longitudinal data analysis. A finite mixture of NLME joint (FMNLMEJ) models is developed to address simultaneous impact of both features under Bayesian framework, which offers a route to estimate not only model parameters but also probabilities of class membership. An AIDS data set is analyzed to demonstrate the methodologies in comparison of the proposed FMNLMEJ model with a commonly used NLME model.