Abstract:
|
In clinical trials of hypertension medications, cardiovascular disease events occur frequently and recurrently over the study follow-up time and are often subject to some dependent censoring (e.g. death). The scientific interest in such study may lies in estimation or inference for regression parameters, characterization of dependence between the events, investigating the impact of event incidence, and prediction given a covariate profile. Existing statistical methods focus almost exclusively on the first two of these where various frailty models have been proposed to estimate the effect of risk factors and understanding dependence. However, the incidence of event may elevate the risk of next recurrent event and the effect of risk factors may change over the event history. Moreover, modeling event history that would facilitate prediction is still underdeveloped. We propose a Bayesian semiparametric regression framework for analyzing recurrent events with dependent termination data that permits the simultaneous investigation of all four of the scientific goals.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.