A Bayesian spatio-temporal nowcasting model for public health decision-making and surveillance

TL;DR

This paper introduces a Bayesian spatio-temporal nowcasting model for COVID-19 case trends at the county level, improving accuracy and uncertainty quantification in public health surveillance.

Contribution

It develops a novel Bayesian model that accounts for reporting delays and spatial-temporal correlations, outperforming existing methods in Ohio COVID-19 data.

Findings

Enhanced trend assessment accuracy

Better uncertainty quantification

Improved decision support for public health

Abstract

As COVID-19 spread through the United States in 2020, states began to set up alert systems to inform policy decisions and serve as risk communication tools for the general public. Many of these systems, like in Ohio, included indicators based on an assessment of trends in reported cases. However, when cases are indexed by date of disease onset, reporting delays complicate the interpretation of trends. Despite a foundation of statistical literature to address this problem, these methods have not been widely applied in practice. In this paper, we develop a Bayesian spatio-temporal nowcasting model for assessing trends in county-level COVID-19 cases in Ohio. We compare the performance of our model to the current approach used in Ohio and the approach that was recommended by the Centers for Disease Control and Prevention. We demonstrate gains in performance while still retaining interpretability using our model. In addition, we are able to fully account for uncertainty in both the time series of cases and in the reporting process. While we cannot eliminate all of the uncertainty in public health surveillance and subsequent decision-making, we must use approaches that embrace these challenges and deliver more accurate and honest assessments to policymakers.