A general joint model for longitudinal measurements and competing risks survival data with heterogeneous random...

Area of Science:

• Biostatistics
• Longitudinal Data Analysis
• Survival Analysis

Background:

• Longitudinal measurements and competing risks survival data are common in health research.
• Existing joint models often have limitations in handling missing data and complex covariance structures.
• There is a need for flexible models that can jointly analyze these data types.

Purpose of the Study:

• To develop a general joint model for longitudinal measurements and competing risks survival data.
• To address non-ignorable missing data in longitudinal outcomes and informative censoring in survival data.
• To allow for heterogeneous random covariance matrices in joint modeling.

Main Methods:

• A joint model combining a linear mixed effects model for longitudinal data and a proportional cause-specific hazards frailty model for survival data.
• Incorporation of a regression sub-model for the variance-covariance matrix using modified Cholesky decomposition.
• Bayesian Markov Chain Monte Carlo (MCMC) procedure for parameter estimation and inference.

Main Results:

• The proposed model effectively adjusts for non-ignorable missing data and handles informative censoring.
• It enables joint analysis of longitudinal and survival outcomes with time-dependent covariates.
• Simulations demonstrate good performance and frequentist properties of the Bayesian estimation procedure.

Conclusions:

• The developed joint model offers a flexible and powerful framework for analyzing complex longitudinal and competing risks survival data.
• It addresses limitations of existing models by allowing heterogeneous covariance matrices and assessing homogeneity assumptions.
• The approach is illustrated with a real data example, showing its practical utility.