Probabilistic small area risk assessment using GIS-based data: a case study on Finnish childhood diabetes

A Bayesian hierarchical spatial model is constructed to describe the region al incidence of insulin dependent diabetes mellitus (IDDM) among the under 15-year-olds in Finland. The model exploits aggregated pixel-wise locations for both the cases and the population at risk. Typically such data arise f rom combining geographic information systems (GIS) with large databases. Th e dates of diagnosis and locations of the cases are observed from 1987 to 1 996. The population at risk counts are available for every second year duri ng the same period. A hierarchical model is suggested for the pixel wise ca se counts, including a population model to account for the uncertainty of t he population at risk over the years. The model is applied in the construct ion of disease maps (aggregated 100 km(2) pixels), and spatial posterior pr edictive distributions are computed to study whether there can be found a s tatistically exceptional number of cases in a small area of interest. Copyr ight (C) 2000 John Wiley & Sons, Ltd.