Figure 1.

Maps of age-adjusted lung cancer mortality rates and the risk estimated using Bayesian and geostatistical methods. The fill color in the left column maps represents the age-adjusted mortality rate per 100,000 person-years recorded over the period 1970–1994 (top graph) or the risk estimated using a Bayesian approach (BYM model) or ATA Poisson kriging. The class boundaries correspond to the deciles of the histogram of original rates. The corresponding prediction variance is mapped in the right column, along with the population at risk that was back-calculated from the rate and count data (lognormal scale).