CMStatistics 2021: Start Registration
View Submission - CMStatistics
B0882
Title: Comparison of Bayesian spatio-temporal models in the infectious disease outbreak surveillance Authors:  Joanne Kim - Medical University of South Carolina (United States) [presenting]
Andrew Lawson - Medical University of South Carolina (United States)
Abstract: COVID-19 pandemic raises the awareness of the public health community for the surveillance of infectious disease outbreaks. Bayesian spatio-temporal models for small area health data have been widely used to analyze infectious disease progression. These models were used for retrospective analysis and prospective surveillance analysis. Retrospective analysis is commonly used to describe the infectious disease outbreak phenomenon; on the other hand, prospective surveillance analysis is more focused on the detection of abnormal patterns of the current data. We compare spatio-temporal models in both retrospective and prospective analysis settings and present and discuss the result for their goodness of fit and prediction ability. The analysis result for both simulation and COVID-19 incidence data of several US states are presented.