RAPID: Covid-19 Forecasting Models for Removal of Social Distancing Measures

点击翻译按钮获取中文摘要

Given the dynamic and uncertain nature of the COVID-19 pandemic, predictive models have been used to inform the implementation of harm-reducing social distancing policies in the United States and elsewhere. Relaxing social distancing measures too early or too broadly, particularly in states where there are still many positive cases, will facilitate community transmission with potential to return to exponential resurgence. This proposal will incorporate measures of population density and human mobility patterns into predictive models to project future COVID-19 cases and deaths within US states and the nation as a whole. As locations ease or end prior distancing policies, consideration of population density and careful tracking of human mobility and incorporation of these data into predictive models will provide data-driven evidence for how these actions could potentially affect COVID-19 trajectories. As a broader impact this project will develop, refine, and share COVID-19 modeling tools that will forecast disease trends for different scenarios of relaxed social distancing measures. Results will be made available through an interactive open-access online visualization tool. In addition to the potential to directly inform policy, the results of this project will benefit the modeling community at large by generating data products of utility to all COVID-19 modeling efforts, which will be made available on a website where results, data, models, and other resultant products are available for open access.The new modelling tools and derived data products produced in this work will enhance forecasting and scenarios of COVID-19 infections and mortality in two formative ways. First, data on population density and human mobility will be used to examine (i) how voluntary and mandated social distancing measures affect population-level mobility, and (ii) the relationship of these measures with Covid-19 infection and mortality. Second, a modelling framework will be expanded to generate forecasts of disease reemergence based on different scenarios of relaxation of social distancing measures that are currently in place. This multi-stage hybrid modeling framework combines a statistical model on deaths with a new component quantifying rates at which individuals move from being susceptible to exposed, infected, and then recovered (known as SEIR models). This modeling platform will be flexible to allow regular data updates and to incorporate new types of data on the drivers of coronavirus as they become available.This RAPID award is made by the Ecology and Evolution of Infectious Diseases Program in the Division of Environmental Biology, using funds from the Coronavirus Aid, Relief, and Economic Security (CARES) Act.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.