Statistical issues in the analysis of disease mapping data

In this paper we discuss a number of issues that are pertinent to the analy sis of disease mapping data. As an illustrative example we consider the map ping of larynx cancer across electoral wards in the North West Thames regio n of the U.K. Bayesian hierarchical models are now frequently employed to c arry out such mapping. In a typical situation, a three-stage hierarchical m odel is specified in which the data are modelled as a function of area-spec ific relative risks at stage one; the collection of relative risks across t he study region are modelled at stage two; and at stage three prior distrib utions are assigned to parameters of the stage two distribution. Such model s allow area-specific disease relative risks to be 'smoothed' towards globa l and/or local mean levels across the study region. However, these models c ontain many structural and functional assumptions at different levels of th e hierarchy; we aim to discuss some of these assumptions and illustrate the ir sensitivity. When relative risks are the endpoint of interest, it is com mon practice to assume that, for each of the age-sex strata of a particular area, there is a common multiplier (the relative risk) acting upon each of the stratum-specific risks in that area; we will examine this proportional ity assumption. We also consider the choices of models and priors at stages two and three of the hierarchy, the effect of outlying areas, and an asses sment of the level of smoothing that is being carried out. For inference, w e concentrate on the description of the spatial variability in relative ris ks and on the association between the relative risks of larynx cancer and a n area-level measure of socio-economic status. Copyright (C) 2000 John Wile y & Sons, Ltd.