OBJECTIVES: Differences in cancer incidence and mortality are apparent among various demographic groups. Understanding the underlying determinants that place certain population subgroups at higher incidence and/or mortality of prostate cancer is imperative. Analyzing temporal trends can provide a comprehensive picture of the burden of the disease and generate new insights about the impact of various interventions. This study aims to use advanced geospatial and temporal statistical techniques to model temporal trends in prostate cancer incidence and mortality and their geographical variations across Florida. METHODS: Annual census-tract level rates were computed over the period 1981-2007 for two races (white and black), two categories of age (40-65, >65) and five classes of incomes. They were then smoothed using binomial kriging to filter the noise caused by small population sizes. Joinpoint regression and new disparity statistics were applied to analyze temporal trends and detect potential racial and socio-economic differences. RESULTS: Bivariate analysis of time-series indicated that late-stage diagnosis was generally more prevalent among blacks compared to whites, for age category 40-64 compared to older patients who are covered by Medicare, and among classes of lower socio-economic status. Joinpoint regression showed that the rate of decline in late-stage diagnosis for the two racial groups was similar among older patients (i.e. parallel time series). Both races displayed distinct spatial patterns with higher rates of late-stage diagnosis in the Florida Panhandle for white males whereas high rates clustered in South-eastern Florida for black males. CONCLUSIONS: The observed impact of socioeconomic and demographic factors on temporal trends in health outcomes emphasizes the need for local strategies and cancer control interventions to reduce late-stage diagnosis and improve health outcomes. Furthermore, large variations in the temporal trends in prostate cancer incidence and mortality and geographical variations would have important implications for resource allocation.