A unified framework for fitting Bayesian semiparametric models to arbitrarily censored survival data, including spatially-referenced data

摘要

A comprehensive, unified approach to modeling arbitrarily censored spatialsurvival data is presented for the three most commonly-used semiparametricmodels: proportional hazards, proportional odds, and accelerated failure time.Unlike many other approaches, all manner of censored survival times aresimultaneously accommodated including uncensored, interval censored,current-status, left and right censored, and mixtures of these. Left-truncateddata are also accommodated leading to models for time-dependent covariates.Both georeferenced (location exactly observed) and areally observed (locationknown up to a geographic unit such as a county) spatial locations are handled;formal variable selection makes model selection especially easy. Model fit isassessed with conditional Cox-Snell residual plots, and model choice is carriedout via LPML and DIC. Baseline survival is modeled with a novel transformedBernstein polynomial prior. All models are fit via a new function which callsefficient compiled C++ in the R package spBayesSurv. The methodology is broadlyillustrated with simulations and real data applications. An important findingis that proportional odds and accelerated failure time models often fitsignificantly better than the commonly-used proportional hazards model.Supplementary materials are available online.