J. Gutiérrez Botella, C. Armero, M. Pata, T. Kneib, F. Gude-Sampedro
Survival studies are usually characterized by the existence of two different states, alive and dead, and classical models study the possible transition from the “alive” state to the “dead” one. Multi-state models are special survival models which include more than two states that individuals may experience sequentially.
We present a Bayesian multi-state model to deal with a retrospective study about patients with heart failure (HF) who received cardiac resynchronization therapy (CRT). HF patients can experience atrial fibrillation and congestive heart failure after CRT. These events are considered intermediate disease states that can occur in any order. Death is an absorbing state that can be reached from any of the previous states as well as from CRT, which is considered the initial condition. Bayesian estimation is performed using MCMC methods with JAGS. Posterior distributions associated to transition intensities and probabilities for each transition are obtained and discussed.
Palabras clave: Atrial fibrillation and congestive heart failure, Transition probabilities, Transition intensities
Programado
Pósteres II
9 de noviembre de 2023 11:40
CC: Sala Pausa Café