Towards joint disease mapping

摘要: This article discusses and extends statistical models to jointly analyse the spatial variation of rates several diseases with common risk factors. We start a review methods for separate analyses diseases, then move ecological regression approaches, where from one enter as surrogate covariates exposure. Finally, we propose general framework modelling two or more some which share latent fields, but possibly different gradients. In our application, consider mortality data on oral, oesophagus, larynx lung cancers males in Germany, all smoking factor. Furthermore, first three are also known be related excessive alcohol consumption. An empirical comparison based formal model criterion well posterior precision relative estimates strongly suggests that joint approach is useful valuable extension over individual analyses.