RAPID: Forecasting & Communicating the US Fall/Winter Respiratory Disease Outlook

点击翻译按钮获取中文摘要

This project will develop new models and forecasts of COVID, influenza, and RSV incidence and severity for the 2023/2024 season. The project will test a new approach to public health forecasting by combining computational models with human judgment. For the first time on a national scale, it will bring together computational modeling and human forecasting, seeking to improve forecasting accuracy and responsiveness. Additionally, it will boost public trust and transparency by introducing innovative communication methods to clearly convey risk and uncertainty. The project will bolster the preparedness of the Center for the 2023/2024 respiratory disease season through the delivery of weekly forecasts, analysis, and communication support. Public communication will be delivered via data dashboards prepared by the CDC and populated with project forecasts, as well as weekly written updates that together will provide a comprehensive, regularly updated, probabilistic outlook on the expected timing, severity, and peaks of COVID, influenza, and RSV. In close collaboration with the CDC, the project will craft tailored visual and verbal communication strategies, emphasizing forecast confidence.This proposal investigates the potential of frequent, probabilistic forecasting and related public communication to improve nationwide public health preparedness during the peak respiratory disease season, building on successful state-level epidemiological response and planning efforts from 2020-2023. The project will operate a public forecasting tournament to provide numerical forecasts for use in communicating and quantifying the expected fall and winter respiratory disease burden of RSV, influenza, and COVID. It will work closely with the CDC to create forecasting questions for the tournament. Questions focused on the timing and magnitude of the peak onset of respiratory illnesses will help chart the expected trajectory, while other questions may track vaccine uptake, week-by-week disease burden, the potential for emerging variants, and the risk of high hospital utilization. In collaboration with the CDC, it also will explore the potential for forecasting questions that could be used to enhance existing CDC modeling through human judgment and parameterization.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.