Identifying the county clusters based on their temporal trends of preterm birth rates in Georgia, USA using a Bayesian space-time model

Keywords: Preterm Birth, Metropolis-Coupled Markov chain Monte Carlo ((MC)3) simulation; Space-time disease mapping; Trend estimation, Georgia, United States
Session Type: Virtual Paper
Day: Wednesday
Session Start / End Time: 4/7/2021 01:30 PM (Pacific Time (US & Canada)) - 4/7/2021 02:45 PM (Pacific Time (US & Canada))
Room: Virtual 18

Abstract

Previous studies have evaluated both the geographic clusters and temporal trends in rates of adverse birth outcomes including preterm birth (PtB) and low birthweight. This study aims to identify clusters of areal units exhibiting elevated risks of PtB and to identify groups of areal units exhibiting similar temporal trends in PtB in Georgia, USA. County-level PtB rates (K= 159) between 1994 and 2019 (N=25) were collected from Online Analytical Statistical Information System (OASIS) at Georgia Department of Public Health. A Bayesian spatio-temporal mixture model was constructed to identify county clusters based on their temporal trends. The candidate trend functions can have either fixed parametric forms (e.g. linear, step-change) or constrained shapes (e.g. monotonically increasing). The counties exhibit an increase, a decrease, or no change in PtB risk over a 25-year period were identified and mapped.