ABSTRACT
Areal data is common in disease mapping applications where often, for confidentiality reasons, individual incidence or mortality information is only available as the number of disease cases aggregated in areas. Disease data can be used to construct atlases that show the geographic distribution of aggregated outcomes to understand spatial patterns, identify high-risk areas, and reveal inequalities. Disease risk can be estimated using Standardized Mortality Ratios (SMR) computed as the ratios of the observed to the expected number of mortality cases. This chapter demonstrates how to specify, fit, and interpret a Bayesian spatial model to estimate the risk of lung cancer and assess its relationship with smoking in Pennsylvania, USA, in 2002. Specifically, it shows how to calculate the expected number of counts and SMR values, and how to obtain disease risk estimates and quantify risk factors using R-INLA.
