Bayesian analysis of multi-type recurrent events and dependent termination with nonparametric covariate functions.

Pubmed ID: 26546256

Pubmed Central ID: PMC5061632

Journal: Statistical methods in medical research

Publication Date: Dec. 1, 2017

Affiliation: 1 Department of Biostatistics, The University of Texas School of Public Health, USA.

MeSH Terms: Humans, Cardiovascular Diseases, Bayes Theorem, Longitudinal Studies, Hypertension, Randomized Controlled Trials as Topic, Proportional Hazards Models, Multivariate Analysis, Computer Simulation, Models, Statistical, Statistics, Nonparametric, Anticholesteremic Agents, Markov Chains, Monte Carlo Method, Recurrence, Biostatistics, Likelihood Functions

Grants: KL2 TR000370, R01 NS091307, U01 NS043127

Authors: Davis BR, Lin LA, Luo S, Chen BE

Cite As: Lin LA, Luo S, Chen BE, Davis BR. Bayesian analysis of multi-type recurrent events and dependent termination with nonparametric covariate functions. Stat Methods Med Res 2017 Dec;26(6):2869-2884. Epub 2015 Nov 6.

Studies:

Abstract

Multi-type recurrent event data occur frequently in longitudinal studies. Dependent termination may occur when the terminal time is correlated to recurrent event times. In this article, we simultaneously model the multi-type recurrent events and a dependent terminal event, both with nonparametric covariate functions modeled by B-splines. We develop a Bayesian multivariate frailty model to account for the correlation among the dependent termination and various types of recurrent events. Extensive simulation results suggest that misspecifying nonparametric covariate functions may introduce bias in parameter estimation. This method development has been motivated by and applied to the lipid-lowering trial component of the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial.