Efficient Inference for Joint Models of Multivariate Longitudinal and Survival Data Using INLAjoint
Clinical research often requires the simultaneous study of longitudinal repeated measurements and time-to-event (i.e., survival) data. Joint models, which can combine these two types of data, are invaluable tools in this context.
Overview
Abstract
Clinical research often requires the simultaneous study of longitudinal repeated measurements and time-to-event (i.e., survival) data. Joint models, which can combine these two types of data, are invaluable tools in this context. A joint model involves multiple regression submodels (one for each longitudinal/survival outcome) usually linked together through correlated or shared random effects. This makes their estimation process rather complex, time-consuming, and sometimes even unfeasible, especially when dealing with many outcomes.
In this context, we introduce INLAjoint, a user-friendly and flexible interface designed to leverage the Integrated Nested Laplace Approximation (INLA) method from the INLA R package. INLAjoint simplifies the application of INLA to fit joint models, ensuring fast and efficient parameter estimation.
Our simulation studies show that INLA reduces the computation time substantially as well as the variability of the parameter estimates compared to alternative strategies such as Bayesian inference via Markov Chain Monte Carlo or maximum likelihood estimation with Monte Carlo expectation maximisation.
We applied INLAjoint to analyze 5 longitudinal markers (3 continuous, 1 count, 1 binary, and 16 random effects) and competing risks of death and transplantation in a clinical trial on primary biliary cholangitis. Given the model's complexity, other estimation strategies couldn't handle it within a reasonable computation time, showcasing INLAjoint's efficiency in the analysis of complex real-world data.
INLAjoint stands out as a highly effective solution for analyzing complex joint models, bridging gaps where traditional methods fall short.
Brief Biography
Denis Rustand obtained his Ph.D. in Public Health, Biostatistics in 2020 where he developed the joint modeling framework for longitudinal and survival data in the context of cancer clinical trials data analysis. He is now a Post-Doctoral fellow at KAUST where he joined the INLA development team. He is the main developer and maintainer of the INLAjoint R package, an user-friendly interface to fit joint longitudinal-survival models with INLA. His research areas include Bayesian computational statistics, survival analysis and applications of statistics to medical research.