Impact of modeling approach and risk threshold on the proportion of counties wrongly classified as having low or high cancer mortality risk (realization #50). Fifty realizations of the spatial distribution of cervix cancer mortality rates in Region 2 were simulated and then analyzed using a Bayesian (BYM model) and a geostatistical (point and area-to-area Poisson kriging) approach. Results for the 50th realization are presented. If a county has at least a 0.75 probability to exceed a risk threshold equal to the area-wide rate (3.186), it is flagged as having significantly higher risk, resulting in potential false positives and negatives based on the actual mortality risk (A, B). The proportion of false alarms is computed for a range of risk thresholds expressed as multiples of the area-wide rate (C, D). Receiver Operating Characteristic (ROC) curves plot the probability of false positive versus the probability of detection for two different thresholds (E, F). The average percentage of false positives (FP) is smaller for the Bayesian approach relatively to Poisson kriging for the low threshold but larger for the high threshold.