COVID-19 scenario modelling for the mitigation of capacity-dependent deaths in intensive care: computer simulation study

Richard M Wood; Christopher J McWilliams; Matthew J Thomas; Christopher P Bourdeaux; Christos Vasilakis

The modelling undertaken in this study has informed on-the-ground planning of healthcare services in response to the COVID-19 pandemic, both within the healthcare system in question and more widely through public release as an R-based open source tool (hosted on github.com/nhs-bnssg-analytics and promoted via social media). Results have been useful in highlighting the need for additional intensive care beds, where decision-makers must now weigh-up the benefits of converting more beds for such purpose against the opportunity cost of such actions (e.g. if theatre space is used then this may limit the ability to perform emergency surgery). Simulation results have also been useful in understanding workforce implications (through measures relating to estimated continual time at maximum occupancy) and the planning of mortuary capacity (through understanding peak daily deaths). While the number of deaths reported in this modelling study may appear to be large against the estimated 353,862 catchment area of the hospital, they should be viewed in the context of other projections. For example, the agestratified case fatality rates of Verity et al, 2020 produce an estimated 4,349 deaths when no effort is made to explicitly model capacity-dependence in the numbers of deaths.