Empirical Bayes Estimates of Age-Standardized Relative Risks for Use in Disease Mapping

摘要: There have been many attempts in recent years to map incidence and mortality from diseases such as cancer. Such maps usually display either relative rates each district, measured by a standardized ratio (SMR) or some similar index, the statistical significance level for test of difference between that district elsewhere. Neither these approaches is fully satisfactory we propose new approach using empirical Bayes estimation. The resulting estimators represent weighted compromise SMR, overall mean rate, local rate nearby areas. solution depends on reliability individual SMR estimates amount dispersion over different districts.