Spatial competing risk models in disease mapping.

Often it is required that the 'health status' of an area must be assessed and this involves the analysis of a range of different diseases within one study window. This often arises, for example, when disease cluster alarms are sounded and there is a need to provide a general overview of health in the vicinity of he cluster area. Our approach leads to the consideration of the joint spatial distribution of a 'basket' of diseases. We examine the use of weighting schemes within our general formulation, and extensions to count data and spatio-temporal modelling.