REGIONAL MAPPING OF INCIDENCE RATES USING SPATIAL BAYESIAN MODELS

This study takes a statistical-modeling point of view to the assessmen t of health care services and procedures. The emphasis is on small-are a prediction of incidence rates from spatially contiguous regions, alt hough suitable modifications can also give doctor-level predictions. T he main idea is to recognize the individuality of each region through a spatial Bayesian model for incidence rates. A noise component, becau se of location error and measurement error, is filtered out using empi rical Bayes methods. The resulting smoothed predictors of incidence ra tes provide an accurate picture of the health care service or procedur e under investigation.