Evidence-driven Spatiotemporal Covid-19 Hospitalization Prediction With Ising Dynamics

HOIST is a model developed to predict the total number of COVID-19 hospitalization cases in an upcoming 4-week period. By analyzing the weights of external fields, it was found that the booster vaccination rate has a more significant negative correlation with future hospitalization cases compared to the first and second vaccinations in the series. Simulating varying immunization rates, it was found that increasing the vaccination ratio by 10% can reduce the number of current hospitalization cases by 15% on average for all locations. Results from the model suggest that rural vaccination outreach efforts are likely cost-saving endeavors.