Spatial Autocorrelation of Pancreatic Cancer Incidence Across ZIP Codes in Hillsborough County, Florida

Abstract

Pancreatic cancer is one of the most lethal malignancies in the United States, with survival outcomes strongly influenced by early detection and underlying social determinants of health. This study examines the spatial distribution of pancreatic cancer incidence across ZIP codes in Hillsborough County and evaluates the relationship between geographic clustering and sociodemographic characteristics. Using spatial analytical techniques, we tested for geographical autocorrelation and identified statistically significant hotspots and cold spots of pancreatic cancer incidence. Global spatial autocorrelation was assessed using Moran’s I to determine whether incidence rates were spatially clustered across ZIP codes. Local cluster detection was then performed with Getis–Ord Gi* statistic to identify areas with significantly higher or lower incidence relative to neighboring locations. To explore potential drivers of these spatial patterns, regression analyses were conducted using the ZIP code–level sociodemographic variables. Independent variables included racial composition, poverty rate, educational attainment, and indicators of socioeconomic disadvantages. These variables were evaluated for their association with pancreatic cancer incidence rates to determine whether disparities in disease burden corresponded with specific demographic and socioeconomic characteristics. Regression findings indicated that several sociodemographic variables, particularly income and racial composition, were significantly correlated with increased incidence rates. These findings highlight the importance of integrating spatial epidemiology with sociodemographic analysis to identify communities experiencing disproportionate disease burden. Targeted public health interventions, improved screening awareness, and resource allocation in identified hot spot areas may help address disparities in pancreatic cancer outcomes within Hillsborough County.