Flattening the curve before it flattens us: hospital critical care capacity limits and mortality from novel coronavirus (SARS-CoV2) cases in US counties

Charles C Branas; Andrew Rundle; Sen Pei; Wan Yang; Brendan G Carr; Sarah Sims; Alexis Zebrowski; Ronan Doorley; Neil Schluger; James W Quinn; Jeffrey Shaman

When considering US regions, the number of counties with critical care beds exceeding their capacity within a month ranged from a high of 38 counties (17.5%) in the Northeast under the very low critical care surge response scenario with no contact reduction, to zero counties in multiple US regions under various scenario combinations. Urban counties were estimated to have greater numbers exceeding their critical care bed capacities within a month, from a maximum of 55 (4.7%) urban counties under the very low critical care surge response scenario with no contact reduction, to 6 (0.5%) urban counties under the high critical care surge response scenario with 50% contact reduction. (Table 2) The 64 counties in the very low critical care surge response scenario with no contact reduction that were at risk of exceeding their bed limits were clustered in various locations across the US -a New York-New Jersey-Connecticut-northeastern Pennsylvania cluster, an eastern Massachusetts cluster, a southeastern Michigan cluster, a southeastern Louisiana cluster, a Colorado cluster, a Washington cluster, a Virginia cluster, and other dispersed counties in five other states. At the other extreme, the 7 counties in the high critical care surge response scenario with 50% contact reduction that were at risk of exceeding their bed limits were clustered in New York, Colorado, and Virginia. (Figure 1) The highest proportion of excess COVID-19 deaths that could have been averted with access to critical care were shown to occur in the Northeast US and urban counties over the month studied. As a measure of actions to flatten the epidemic curve, the difference in excess deaths between a 0% and a 50% contact reduction ranged from an estimated 12,203 to 19,594 excess deaths averted over a month. As a measure of the impact of aggressive critical care surge actions, the difference in excess deaths between the high and the very low critical care surge response scenarios ranged from an estimated 4,029 to 11,420 excess deaths averted over a month. As a measure of the impact of redeploying non-ICU beds for critical care surge response, the difference in excess deaths between the medium and the very low critical care surge response scenarios ranged from an estimated 3,050 to 6,946 excess deaths averted over a month. As a measure of the impact of putting two patients on a single ventilator, the difference in excess deaths between the high and the medium critical care surge response scenarios ranged from an estimated 979 to 4,474 excess deaths averted over a month. (Table 3) The increase in critical care beds that could be achieved under the various surge response scenarios was highly correlated with the number of beds estimated under the baseline critical care bed availability model. Focusing on the medium critical care surge capacity scenario, the . CC-BY-NC-ND 4.0 International license It is made available under a author/funder, who has granted medRxiv a license to display the preprint in perpetuity.