COVID-19 Hospitalizations , ICU Admissions and Deaths Associated with the New Variants of Concern

March 29, 2021 | 1 Ashleigh R. Tuite, David N. Fisman, Ayodele Odutayo, Pavlos Bobos, Vanessa Allen, Isaac I. Bogoch, Adalsteinn D. Brown, Gerald A. Evans, Anna Greenberg, Jessica Hopkins, Antonina Maltsev, Douglas G. Manuel, Allison McGeer, Andrew M. Morris, Samira Mubareka, Laveena Munshi, V. Kumar Murty, Samir N. Patel, Fahad Razak, Robert J. Reid, Beate Sander, Michael Schull, Brian Schwartz, Arthur S. Slutsky, Nathan M. Stall, Peter Jüni on behalf of the Ontario COVID-19 Science Advisory Table

COVID-19 Hospitalizations, ICU Admissions and Deaths Associated with the New Variants of Concern Ontario COVID-19 Science Advisory Table   2021 and December 26, 2020 is indicated above the corresponding bars for hospital and ICU occupancy. ICU, intensive care unit.

Summary Background
As of March 28, 2021 new variants of concern (VOCs) account for 67% of all Ontario SARS-CoV-2 infections. The B.1.1.7 variant originally detected in Kent, United Kingdom accounts for more than 90% of all VOCs in Ontario, with emerging evidence that it is both more transmissible and virulent.

Questions
What are the risks of COVID-19 hospitalization, ICU admission and death caused by VOCs as compared with the early variants of SARS-CoV-2?
What is the early impact of new VOCs on Ontario's healthcare system?
A meta-analysis including the Ontario cohort study and additional cohort studies in the United Kingdom and Denmark showed that people infected with VOCs had a 63% higher risk of hospitalization (

Interpretation
The new VOCs will result in a considerably higher burden to Ontario's health care system during the third wave compared to the impact of early SARS-CoV-2 variants during Ontario's second wave.
Since the start of the third wave on March 1, 2021, the number of new cases of SARS-CoV-2 infection, and the COVID-19 hospital and ICU occupancies have surpassed prior thresholds at the start of the province-wide lockdown on December 26, 2020.  framework: grey/red zone = weekly SARS-CoV-2 incidence of ≥40 per 100,000; orange zone = weekly incidence 25 to 39.9 per 100,000; and yellow zone = weekly incidence of 10 to 24.9 per 100,000. VOC, variant of concern. Graph adapted from Ontario COVID-19 Science Advisory Table. 1 The B.1.1.7 variant, which was originally detected in Kent, United Kingdom, currently accounts for more than 90% of all VOCs in Ontario. The B.1.351 and P.1 variants originally detected in South Africa and Brazil, respectively, account for the remaining VOCs. 2 The B.1.1.7 variant which is dominant in Ontario, is at least 40% more transmissible than early variants of SARS-CoV-2, 3 and emerging evidence suggests it may be more virulent. 4

Questions
What are the risks of COVID-19 hospitalization, ICU admission and death caused by VOCs as compared with the early variants of SARS-CoV-2?
What is the early impact of new VOCs on Ontario's healthcare system? Table 1 presents the results of a retrospective cohort study of 26,314 people in Ontario who were PCR-positive for SARS-CoV-2 between February 7 and March 11, 2021, with 9,395 people (35.7%) having an infection caused by a VOC. After adjusting for age, sex and comorbidities, infections due to VOCs were associated with a 62% relative increase in COVID-19 hospitalizations (odds ratio [OR] 1.62, 95% confidence interval [CI] 1.41 to 1.87), a 114% relative increase in ICU admissions (OR 2.14, 95% CI 1.52 to 3.02), and a 40% relative increase in COVID-19 deaths (OR 1.40, 95% CI 1.01 to 1.94). These risk elevations for COVID-19 hospitalization, ICU admission and death were consistent across all age groups.      6 Participants were matched on age, sex, date of specimen collection, ethnicity, geographical location, and index of multiple deprivation, which is a marker for socioeconomic status; estimates were subsequently adjusted for age. VOC, variant of concern; CI, confidence interval; ICU, intensive care unit. Figure 4 shows the time to death observed in a retrospective cohort study by Challen et al. in the United Kingdom involving 109,812 participants with positive PCR tests for SARS-CoV-2 between October 1, 2020, to January 29, 2021, with 54,906 participants infected with new VOCs matched to 54,906 participants infected with the early variant. 6 The curves overlap until day 12 after diagnosis, at which point the curves start to separate, with a higher risk of death among participants infected with new VOCs compared with participants infected with early variants.

Figure 4. Time to Death Following SARS-CoV-2 Infection with New VOCs Compared with Early Variants Curves describing the time to death from first PCR confirmation of SARS-CoV-2 infection among individuals in the United
Kingdom infected with new VOCs versus early SARS-CoV-2 variants. Participants were matched on age, sex, date of specimen collection, ethnicity, geographical location, and index of multiple deprivation, which is a marker for socioeconomic status, and estimates were subsequently adjusted for age. Data from Challen et al. 6 VOC, variant of concern. Figure 5 shows the risk of death associated with new VOCs compared with early SARS-CoV-2 variants from days 0 to 14 and days 15 to 28 after diagnosis of SARS-CoV-2 infection. Results are adapted from the aforementioned retrospective cohort study of people with PCR confirmed SARS-CoV-2 infection in the United Kingdom by Challen et al. 6 Between days 0 and 14 after diagnosis, there was only a minimal difference in the risk of death (RR 1.23, 95% CI 0.92 to 1.64) associated with the new VOCs compared with early variants. However, between days 15 and 28, the risk of death was more than doubled (RR 2.40, 95% CI 1.66 to 3.47).

Figure 5. Risk of COVID-19 Death Associated with new VOCs Compared with Early Variants by Time Since Diagnosis of SARS-CoV-2 Infection
Adapted from a retrospective cohort study by Challen et Table   collection, ethnicity, geographical location, and index of multiple deprivation, which is a marker for socioeconomic status. Each square presents the relative risk for death with the new VOCs versus early variants. The horizontal lines indicate the 95% confidence intervals. The solid vertical line at 1 indicates that there is no difference in prognosis between new VOCs and early SARS-CoV-2 variants. VOC, variant of concern; CI, confidence interval. Figure 6 shows the risk of ICU admission associated with new VOCs versus early SARS-CoV-2 variants at days 1, 5, 10 and 15 after diagnosis of SARS-CoV-2 infection in a retrospective cohort study by Patone et al. The study involved 198,420 individuals in the United Kingdom with PCR-confirmed SARS-CoV-2 infection, of whom 80,494 were infected with new VOCs. 5 On day 1, there was a minimal difference in the risk of ICU admission between people infected with new VOCs and those infected with early variants (RR 1.20, 95% CI 0.58 to 2.48). Subsequently, there was a progressive, lagged increase in the risk of ICU admission associated with new VOCs, with a 58% increase at day 5 and a near fourfold increase at day 10.

Figure 6. Risk of COVID-19 ICU Admission Associated with New VOCs Compared with Early Variants by Time Since Diagnosis of SARS-CoV-2 Infection
Adapted from a retrospective cohort study by Patone et al. 5 The study involved 198,420 individuals in the United Kingdom with PCR-confirmed SARS-CoV-2 infection, of whom 80,494 were infected with new VOCs. Relative risk estimated were adjusted for age, sex, region, socio-demographic factors and comorbidities including asthma, chronic obstructive pulmonary disease, diabetes and hypertension. Each square represents the relative risk for ICU admission with the new VOCs versus early variants. The horizontal lines indicate the 95% confidence intervals. The solid vertical line at 1 indicates that there is no difference in prognosis between new VOCs and early variants. VOC, variant of concern; CI, confidence interval. We project a 2 to 4 week time lag between daily SARS-CoV-2 cases and COVID-19 hospitalizations and ICU admissions, with lagging risk increases due to the new VOCs (see Figures 3 to 5). Therefore, hospital and ICU occupancies due to COVID-19 will continue to increase considerably over time, and would so even if SARS-CoV-2 case numbers were to remain at the current level seen on March 28, 2021.  Table 2 presents a comparison of the number of new SARS-CoV-2 infections and COVID-19 hospital and ICU occupancy for key dates during the third wave compared with the start of the province-wide lockdown on December 26, 2020 during the second wave. As of March 28, 2021, the predicted 7-day average of SARS-CoV-2 infections during the third wave reached the 7-day midpoint average seen at the start of the province-wide lockdown on December 26, 2020.

Interpretation
Compared with early variants of SARS-CoV-2, new VOCs are associated with a 103% increase in the risk of hospitalization, a 63% increase in the risk of ICU admission and a 56% increase in the risk of death due to COVID-19, which will result in a considerably higher burden to the health care system than observed with early variants during the second wave. The risk increase is particularly pronounced 14 to 28 days after a diagnosis of SARS-CoV-2 infection, which in turn will result in delays until the full burden to the health care system becomes apparent.
Since the start of the third wave around March 1, 2021, the number of new cases of SARS-CoV-2 infection, and the COVID-19 hospital and ICU occupancies have surpassed prior thresholds at the start of the province-wide lockdown on December 26, 2020. As of March 28, 2020, hospital occupancy was 21% higher and ICU occupancy 28% higher than at the start of the province-wide lockdown.
Currently, patients aged 59 years and younger make up 46% of new COVID-19 admissions to ICUs, compared with 30% in the week prior to the start of the province-wide lockdown on December 26, 2020.