A dynamic modeling tool for estimating healthcare demand from the COVID19 epidemic and evaluating population-wide interventions

Rainisch, Gabriel; Undurraga, Eduardo A.; Sciences, Gerardo Chowell Department of Population Health; Health, School of Public; University, Georgia State; Atlanta,; GA,; USA,; Gobierno, Escuela de; Chile, Pontificia Universidad Cat'olica de; RM,; Resistance, Chile Millennium Initiative for Collaborative Research i; Chile,

Objectives. Public health officials need tools to assist with anticipating the healthcare resources required to confront the SARS-COV-2 pandemic. We built a modeling tool to aid practicing public health officials with estimating healthcare demand from the pandemic in their jurisdictions and to evaluate the potential impacts of population-wide social-distancing interventions. Methods. The tool uses a SEIR compartmental model to project the local spread of the pandemic. Users input case counts, healthcare resources, and select intervention strategies to evaluate. Outputs include the number of infections and deaths with and without intervention, and the demand for hospital and critical care beds and ventilators relative to existing capacity. We illustrate the tool using data from three regions of Chile. Results. Our scenarios indicate a surge in COVID-19 patients could overwhelm Chilean hospitals by June, peaking in July or August at 6 to 50 times the current supply of beds and ventilators. A lockdown strategy or combination of case isolation, home quarantine, social distancing of individuals greater than 70 years, and telework interventions may keep treatment demand below capacity. Conclusions. Aggressive interventions can avert substantial morbidity and mortality from COVID-19. Our tool permits rapid evaluation of locally-applicable policy scenarios and updating of results as new data become available.