Creating Small-Area Cancer Risk Estimates to Promote Cancer Control Activities in Rural Areas (306553)Mary E. Charlton, University of Iowa College of Public Health
*Melissa Nicole Jay, University of Iowa College of Public Health
Jacob Oleson, University of Iowa
Keywords: Bayesian hierarchical models, spatial statistics, cancer risk, cancer epidemiology
This research identifies geographic areas in the U.S. that may have increased modifiable cancer risks that can be targeted for intervention, with a particular focus on rural exposures, given that rural populations have unique exposures and risks beyond socioeconomic and lifestyle factors. It is challenging to obtain reliable estimates for diseases such as cancer in rural counties because the number of cases is very small. We create small area estimates for measures of cancer risk in counties across the United States using Bayesian hierarchical Poisson regression models that incorporate spatial and temporal correlation structures to obtain more reliable estimates for rural counties. In particular, we focus on how environmental exposures are related to cancer risk while controlling for socioeconomic variables. We use cancer mortality rate data from the Institute for Health Metrics and Evaluation for 3,108 counties in the U.S. from 2005 to 2014.