Figure 10From: Geographical, temporal and racial disparities in late-stage prostate cancer incidence across Florida: A multiscale joinpoint regression analysisImpact of using noise-filtered rates versus raw rates on the results of joinpoint regression. The scatterplots of observed versus modelled rates were created for joinpoint linear regression models fitted to raw rates (A) or rates that were processed using binomial kriging (B). Bottom plots illustrate the impact of noise filtering on the temporal trend of two measures of racial disparities: (C) the number of counties where the 95% confidence intervals of APC for white and black males did not overlap (Equation 8), and (D) the number of counties where APC for Blacks is significantly smaller than for Whites.Back to article page