Fatality assessment and variant risk monitoring for COVID-19 using three new hospital occupancy related metrics
Ping-Wu Zhang,
Steven H. Zhang,
Wei-Feng Li,
Casey J. Keuthan,
Shuaizhang Li,
Felipe Takaesu,
Cynthia A. Berlinicke,
Jun Wan,
Jing Sun,
Donald J. Zack
Affiliations
Ping-Wu Zhang
Wilmer Eye Institute, Johns Hopkins University School of Medicine; Baltimore, MD, 21231, USA; Corresponding authors at: Wilmer Eye Institute, Johns Hopkins University School of Medicine, 400 N Broadway, Smith Building, Room 3001, Baltimore, MD, 21231, USA.
Steven H. Zhang
Wilmer Eye Institute, Johns Hopkins University School of Medicine; Baltimore, MD, 21231, USA; Department of Computer Science, University of Maryland at College Park (present affiliation)
Wei-Feng Li
Wilmer Eye Institute, Johns Hopkins University School of Medicine; Baltimore, MD, 21231, USA
Casey J. Keuthan
Wilmer Eye Institute, Johns Hopkins University School of Medicine; Baltimore, MD, 21231, USA
Shuaizhang Li
Wilmer Eye Institute, Johns Hopkins University School of Medicine; Baltimore, MD, 21231, USA
Felipe Takaesu
Wilmer Eye Institute, Johns Hopkins University School of Medicine; Baltimore, MD, 21231, USA
Cynthia A. Berlinicke
Wilmer Eye Institute, Johns Hopkins University School of Medicine; Baltimore, MD, 21231, USA
Jun Wan
Department of Medical and Molecular Genetics, Center for Computational Biology and Bioinformatics, Indiana University School of Medicine, Indianapolis, IN, 46202, USA
Jing Sun
Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, MD, 21113, USA
Donald J. Zack
Wilmer Eye Institute, Johns Hopkins University School of Medicine; Baltimore, MD, 21231, USA; Corresponding authors at: Wilmer Eye Institute, Johns Hopkins University School of Medicine, 400 N Broadway, Smith Building, Room 3001, Baltimore, MD, 21231, USA.
Summary: Background: Though case fatality rate (CFR) is widely used to reflect COVID-19 fatality risk, its use is limited by large temporal and spatial variation. Hospital mortality rate (HMR) is also used to assess the severity of COVID-19, but HMR data is not directly available globally. Alternative metrics are needed for COVID-19 severity and fatality assessment. Methods: We introduce new metrics for COVID-19 fatality risk measurements/monitoring and a new mathematical model to estimate average hospital length of stay for deaths (Ldead) and discharges (Ldis). Multiple data sources were used for our analyses. Findings: We propose three, new metrics: hospital occupancy mortality rate (HOMR), ratio of total deaths to hospital occupancy (TDHOR), and ratio of hospital occupancy to cases (HOCR), for dynamic assessment of COVID-19 fatality risk. Estimated Ldead and Ldis for 501,079 COVID-19 hospitalizations in 34 US states between 7 August 2020 and 1 March 2021 were 18·2(95%CI:17·9-18·5) and 14·0(95%CI:13·9-14·0) days, respectively. We found the dramatic changes in COVID-19 CFR observed in 27 countries during early stages of the pandemic were mostly caused by undiagnosed cases. Compared to the first week of November 2021, the week mean HOCRs (mimics hospitalization-to-case ratio) for Omicron variant (58·6% of US new cases as of 25 December 2021) decreased 65·16% in the US as of 16 January 2022. Interpretation: The new and reliable measurements described here could be useful for COVID-19 fatality risk and variant-associated risk monitoring. Funding: No specific funding was associated with the present study.
