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.