Bayesian modeling of clustered competing risks survival times with spatial random effects

In some studies, survival data are arranged spatially such as geographical regions. Incorporating spatial association in these data not only can increase the accuracy and efficiency of the parameter estimation, but it also investigates the spatial patterns of survivorship. In this paper, we considered a Bayesian hierarchical survival model in the setting of competing risks for the spatially clustered HIV/AIDS data. In this model, a Weibull Parametric distribution with the spatial random effects in the form of county-failure type-level was used. A multivariate intrinsic conditional autoregressive (MCAR) distribution was employed to model the areal spatial random effects. Comparison among competing models was performed by the deviance information criterion and log pseudo-marginal likelihood. We illustrated the gains of our model through the simulation studies and application to the HIV/AIDS data.