Knorr-Held, Leonhard; Becker, Nikolaus (1999): Bayesian Modelling of Spatial Heterogeneity in Disease Maps with Application to German Cancer Mortality Data. Sonderforschungsbereich 386, Discussion Paper 176




This paper starts with a short overview of basic concepts in disease mapping such as relative risk and age-standardization. Then two recent methods for advanced statistical analysis of areal summary measures of health outcomes are reviewed which overcome difficulties in traditional mapping methods. Both methods account for spatial correlation in an hierarchical Bayesian framework and use computer-intensive Markov chain Monte Carlo methods for statistical inference. The methods are compared through analyses of cancer mortality data from Germany, 1986-1990.