Title : Population size-driven diversification as a key factor in ICU occupancy peaks during COVID-19
Abstract:
During the COVID-19 pandemic, elevated ICU occupancy levels significantly increased mortality rates for COVID-19 patients. Costly non-pharmaceutical interventions, such as lockdowns, and elective surgery suspensions were implemented to reduce the height of occupancy peaks. The health and financial implications of ICU occupancy levels explain their use as indicators for decision-making during the pandemic and why understanding the behaviour of this occupancy metric could have important implications for healthcare policy. This research investigates the relationship between population size and the maximum weekly COVID ICU occupancy level reached during the 2021-2022 period in 698 Health Service Areas (HSAs) across the United States. Higher population sizes are linked to greater variability and lower quantiles (0.05th, 0.5, and 0.95) of maximum ICU occupancy.
Peak occupancy decreases logarithmically from a median of 74% of ICU beds for a population of 40,000 to 45% for 500,000, stabilizing past this threshold. Through simulation using susceptible-infected-removed (SIR) models and analysis of time series of HSA ICU occupancy data, we find that the observed pattern can be explained by volatility caused by population-size-related stochasticity, local differences in community vulnerability and imperfect synchronization of local epidemics within a population. A parallel is drawn between these three effects and the process of diversification used in the insurance and financial sectors.
Our findings inform bed planning policy for pandemics by highlighting the relative vulnerability to ICU occupancy peaks in bed pools serving different population sizes. Additionally, our results show that healthcare capacity planning for pandemics should be based on consideration of the size of the population served by hospital networks and should maximize capacity sharing between healthcare facilities.
Audience Take Away Notes:
- Our results highlight a significant driver of healthcare capacity risk during pandemics and can be used to inform healthcare planning. Additionally, our results offer a strong theoretical argument to justify the need for regional hospital collaboration during pandemics which can help pandemic preparedness efforts.
- Our results highlight the large influence of population size on healthcare capacity risk during a pandemic which is not yet well-documented in the literature. Gaining a better understanding of this risk can help improve patient care and optimize healthcare capacity allocation for pandemics.
- The empirical curve relating population size and the height of ICU capacity peaks estimated in this study offers an empirical measure of the relative vulnerability of populations of different sizes to ICU capacity peaks. This curve could form the basis of local vulnerability assessments used in pandemic planning.
