B0333
Title: Modeling spatial health disparities using disease maps
Authors: Luca Aiello - University of Milano Bicocca (Italy) [presenting]
Sudipto Banerjee - UCLA (United States)
Abstract: The detection of health disparities across regions through statistical analysis of disease maps is a common goal in epidemiology. Mapping mortality or incidence rates alone may not be sufficient, as it is crucial to identify "difference boundaries" that separate neighbouring regions with significantly distinct effects. This task becomes more challenging when considering multiple outcomes and accounting for interdependence among diseases and regions. We address the problem of multivariate difference boundary detection for correlated diseases by employing Bayesian pairwise multiple comparisons and incorporating adjacency modelling. By estimating the posterior probabilities of diverse spatial effects between neighbouring regions, we utilize a multivariate areally referenced Dirichlet process model that accommodates spatial and inter-disease dependencies through discrete probability distributions. Through simulation studies and application to the detection of difference boundaries for multiple cancers using data from the national cancer institute's surveillance, epidemiology, and end results program, the efficacy of the approach is demonstrated in uncovering health disparities and informing public health decision-making.
