A nonlinear latent class model for joint analysis of multivariate longitudinal data and a binary outcome

Latent class models have been widely used in longitudinal studies to uncover unobserved heterogeneity in a population and find the characteristics of the latent classes simultaneously using the class allocation probabilities dependent on predictors. However, previous latent class models for longitudinal data suffer from uncertainty in the choice of the number of latent classes. In this study, we propose a Bayesian nonparametric latent class model for longitudinal data, which allows the number of latent classes to be inferred from the data. The proposed model is an infinite mixture model with predictor-dependent class allocation probabilities; an individual longitudinal trajectory is described by the class-specific linear mixed effects model. The model parameters are estimated using Markov chain Monte Carlo methods. The proposed model is validated using a simulated example and a real-data example for characterizing latent classes of estradiol trajectories over the menopausal transition using data from the Study of Women’s Health Across the Nation.

Download Full-text

Joint analysis of multiple longitudinal outcomes: Application of a latent class model

Statistics in Medicine ◽

10.1002/sim.3435 ◽

2008 ◽

Vol 27 (29) ◽

pp. 6228-6249 ◽

Cited By ~ 12

Author(s):

Hein Putter ◽

Tineke Vos ◽

Hanneke de Haes ◽

Hans van Houwelingen

Keyword(s):

Latent Class ◽

Latent Class Model ◽

Joint Analysis ◽

Class Model ◽

Longitudinal Outcomes

Download Full-text

Joint latent class model for longitudinal data and interval‐censored semi‐competing events: Application to dementia

Biometrics ◽

10.1111/biom.12530 ◽

2016 ◽

Vol 72 (4) ◽

pp. 1123-1135 ◽

Cited By ~ 11

Author(s):

Anaïs Rouanet ◽

Pierre Joly ◽

Jean‐François Dartigues ◽

Cécile Proust‐Lima ◽

Hélène Jacqmin‐Gadda

Keyword(s):

Longitudinal Data ◽

Latent Class ◽

Latent Class Model ◽

Class Model ◽

Interval Censored ◽

Competing Events

Download Full-text

Bayesian joint analysis of heterogeneous- and skewed-longitudinal data and a binary outcome, with application to AIDS clinical studies

Statistical Methods in Medical Research ◽

10.1177/0962280217689852 ◽

2017 ◽

Vol 27 (10) ◽

pp. 2946-2963 ◽

Cited By ~ 2

Author(s):

Xiaosun Lu ◽

Yangxin Huang ◽

Jiaqing Chen ◽

Rong Zhou ◽

Shuli Yu ◽

...

Keyword(s):

Longitudinal Data ◽

Latent Class ◽

Joint Modeling ◽

Finite Mixture ◽

Joint Analysis ◽

Binary Outcome ◽

Class Membership ◽

Joint Models ◽

Nonlinear Mixed Effects Models ◽

Skew Distributions

In medical studies, heterogeneous- and skewed-longitudinal data with mis-measured covariates are often observed together with a clinically important binary outcome. A finite mixture of joint models is currently used to fit heterogeneous-longitudinal data and binary outcome, in which these two parts are connected by the individual latent class membership. The skew distributions, such as skew-normal and skew-t, have shown beneficial in dealing with asymmetric data in various applications in literature. However, there has been relatively few studies concerning joint modeling of heterogeneous- and skewed-longitudinal data and a binary outcome. In this article, we propose a joint model in which a flexible finite mixture of nonlinear mixed-effects models with skew distributions is connected with binary logistic model by a latent class membership indicator. Simulation studies are conducted to assess the performance of the proposed models and method, and a real example from an AIDS clinical trial study illustrates the methodology by modeling the viral dynamics to compare potential models with different distribution specifications; the analysis results are reported.

Download Full-text