Effect of doubling time and social distancing on numbers of beds and ventilators for COVID patients

A model has been developed by Penn Medicine that allows planners to evaluate the projected needs on numbers of beds and ventilators required by a particular hospital for COVID cases using the CHIME tool. As an example, the projected results in the top Figures below show the strong growth in need for beds and ventilators obtained even when the doubling time is 6 days compared to the stronger peak and overwhelming demand when the doubling time is 3 days. The bottom Figures show the dramatic reduction in demand for beds and ventilators accompanying a 25% reduction in person-to-person contact from social distancing. Such models should be useful in guiding planning and decision making on needed beds and ventilators and also can support clear communications on the benefit of interventions and taking individual responsibility. Social distancing clearly saves lives!

Notes: The doubling time reflects the number of days required for infections to double under “status quo” conditions, may vary with region and time, and have typically reported values in recent weeks lying between 3-6 days (https://blogs.scientificamerican.com/observations/flattening-the-covid-19-curves) Social distancing reflects reduction in interpersonal contact and transmission relative to the status quo. Model results depend on numerous factors to be specified for a particular hospital and region (values in Figures in brackets): Currently hospitalised COVID-19 patients (4), Doubling Time (3 and 6 days used in Figs), Social Distancing (0 and 25% used in Figs), Hospitalization (5%), ICU % of total infections (2%), Ventilated % of total infections (1%), Hospital Length of Stay (7 days), ICU Length of Stay (9 days), Ventilation Length of Stay (10 days), Hospital Market Share (15%), Regional Population (4119405) and currently known regional infections to compute detection rate (91). The CHIME model (“Covid-19 Hospital Impact Model for Epidemics”) and tool was developed by the Predictive Healthcare team at Penn Medicine. Source: https://penn-chime.phl.io/