Bayesian modelling of inseparable space-time variation in disease risk

This paper proposes a unified framework for a Bayesian analysis of incidenc e or mortality data in space and time. We introduce four different types of prior distributions for spacextime interaction in extension of a model wit h only main effects. Each type implies a certain degree of prior dependence for the interaction parameters, and corresponds to the product of one of t he two spatial with one of the two temporal main effects. The methodology i s illustrated by an analysis of Ohio lung cancer data 1968-1988 via Markov chain Monte Carlo simulation. We compare the fit and the complexity of seve ral models with different types of interaction by means of quantities relat ed to the posterior deviance. Our results confirm an epidemiological hypoth esis about the temporal development of the association between urbanization and risk factors for cancer. Copyright (C) 2000 John Wiley & Sons, Ltd.