B0846
Title: Bayesian age decomposition modeling of Covid-19 space-time dynamics
Authors: Yao Xin - Medical University of South Carolina (United States)
Andrew Lawson - Medical University of South Carolina (United States) [presenting]
Abstract: Age dependence of COVID-19 infection potential is important, but there is limited public access to age structure in the COVID-19 pandemic progression. A space-time Bayesian model is developed for disease spread with age stratification. Without knowledge of the breakdown of age structure, it is assumed that the age groups have different relevant infection rates and are also conditioned on the observed case counts or deaths. Imputation is used to infer the distribution of age-specific case and death counts. To test the relevance of the approach, age-stratified (anonymized) case and death counts were obtained for the counties of South Carolina during the main waves of the pandemic. This is used to evaluate the method but will also consider whether nowcasting can be used to examine counterfactuals for different counties and hence policy evaluation.
