A shared component model for detecting joint and selective clustering of two diseases

The study of spatial variations in disease rates is a common epidemiologica l approach used to describe the geographical clustering of diseases and to generate hypotheses about the possible 'causes' which could explain apparen t differences in risk. Recent statistical and computational developments ha ve led to the use of realistically complex models to account for overdisper sion and spatial correlation. However, these developments have focused almo st exclusively on spatial modelling of a single disease. Many diseases shar e common risk factors (smoking being an obvious example) and, if similar pa tterns of geographical variation of related diseases can be identified, thi s may provide more convincing evidence of real clustering in the underlying risk surface. We propose a shared component model for the joint spatial an alysis of two diseases. The key idea is to separate the underlying risk sur face for each disease into a shared and a disease-specific component. The v arious components of this formulation are modelled simultaneously by using spatial cluster models implemented via reversible jump Markov chain Monte C arlo methods. We illustrate the methodology through an analysis of oral and oesophageal cancer mortality in the 544 districts of Germany, 1986-1990.