covidestim: A Vital Tool for Pandemic Response

Project Summary

covidestim was a real-time modeling tool developed by epidemiologists from the Yale School of Public Health and the Harvard T.H. Chan School of Public Health, with computational support from the Yale Center for Research Computing. During its operation, the tool provided up-to-date COVID-19 metrics across the United States on a county-by-county and state-by-state basis, including estimated infection rates, the effective case reproduction number, and estimated seroprevalence.

Designed for both public health officials and the public, covidestim offered critical insights for government officials and health leaders to help make informed decisions about public health interventions.

The tool operated until its shutdown in January 2024 and remains freely available on the covidestim website. All data and estimates over the entire pandemic have been stored and are freely and publicly available in the Harvard Dataverse.

Critically, covidestim focused on "nowcasting"—providing a clearer understanding of the current state of the pandemic rather than projecting future trends, using sophisticated epidemiological models and statistical reasoning. The model addressed several challenges in real-time tracking, such as asymptomatic cases, delays in reporting symptomatic cases, and inconsistencies in cases, hospitalization and death reporting, to develop the most accurate snapshots of the pandemic possible.

During its operation, the covidestim models were initially updated daily (and later weekly), providing a glimpse into the pandemic’s progression through time and across counties and states. The tool revealed key trends, including the geographic heterogeneity of infection rates, with hotspots emerging in different regions at different times. Its findings helped visualize the wave-like behavior of the pandemic.

covidestim had a profound impact on public health planning during the pandemic. The tool was utilized by academic institutions and government bodies, including Yale University, which relied on state-level projections at the start of the academic semester. Some policymakers involved with Connecticut’s reopening plan also used covidestim to track county-level infection trends. The tool’s ability to offer real-time insights helped shape public health interventions and assess their effectiveness.

Though covidestim is no longer being updated, its influence on public health decision-making during the pandemic was significant. The team behind the tool made their code and estimates freely available, encouraging others to build on their work and further advance public health preparedness for future outbreaks.

Project Contributors

This project was supported by Cooperative Agreement NU38OT000297 from the Centers for Disease Control and Prevention (CDC) and the Council of State and Territorial Epidemiologists (CSTE), SHEPheRD Contract 200-2016-91779 from the CDC, and the CDC Broad Agency Announcement Contract 75D30122C14697.