key: cord-0744883-jupxmegf
authors: Garg, Shikha; Patel, Kadam; Pham, Huong; Whitaker, Michael; O'Halloran, Alissa; Milucky, Jennifer; Anglin, Onika; Kirley, Pam D.; Reingold, Arthur; Kawasaki, Breanna; Herlihy, Rachel; Yousey-Hindes, Kimberly; Maslar, Amber; Anderson, Evan J.; Openo, Kyle P.; Weigel, Andrew; Teno, Kenzie; Ryan, Patricia A.; Monroe, Maya L.; Reeg, Libby; Kim, Sue; Como-Sabetti, Kathryn; Bye, Erica; Shrum Davis, Sarah; Eisenberg, Nancy; Muse, Alison; Barney, Grant; Bennett, Nancy M.; Felsen, Christina B.; Billing, Laurie; Shiltz, Jess; Sutton, Melissa; Abdullah, Nasreen; Talbot, H. Keipp; Schaffner, William; Hill, Mary; Chatelain, Ryan; Wortham, Jonathan; Taylor, Christopher; Hall, Aron; Fry, Alicia M.; Kim, Lindsay; Havers, Fiona P.
title: Clinical Trends Among U.S. Adults Hospitalized With COVID-19, March to December 2020: A Cross-Sectional Study
date: 2021-08-10
journal: Ann Intern Med
DOI: 10.7326/m21-1991
sha: 2026d8d078463992c4b6bb85ae36ee9907d2c84e
doc_id: 744883
cord_uid: jupxmegf

BACKGROUND: The COVID-19 pandemic has caused substantial morbidity and mortality. OBJECTIVE: To describe monthly clinical trends among adults hospitalized with COVID-19. DESIGN: Pooled cross-sectional study. SETTING: 99 counties in 14 states participating in the Coronavirus Disease 2019–Associated Hospitalization Surveillance Network (COVID-NET). PATIENTS: U.S. adults (aged ≥18 years) hospitalized with laboratory-confirmed COVID-19 during 1 March to 31 December 2020. MEASUREMENTS: Monthly hospitalizations, intensive care unit (ICU) admissions, and in-hospital death rates per 100 000 persons in the population; monthly trends in weighted percentages of interventions, including ICU admission, mechanical ventilation, and vasopressor use, among an age- and site-stratified random sample of hospitalized case patients. RESULTS: Among 116 743 hospitalized adults with COVID-19, the median age was 62 years, 50.7% were male, and 40.8% were non-Hispanic White. Monthly rates of hospitalization (105.3 per 100 000 persons), ICU admission (20.2 per 100 000 persons), and death (11.7 per 100 000 persons) peaked during December 2020. Rates of all 3 outcomes were highest among adults aged 65 years or older, males, and Hispanic or non-Hispanic Black persons. Among 18 508 sampled hospitalized adults, use of remdesivir and systemic corticosteroids increased from 1.7% and 18.9%, respectively, in March to 53.8% and 74.2%, respectively, in December. Frequency of ICU admission, mechanical ventilation, and vasopressor use decreased from March (37.8%, 27.8%, and 22.7%, respectively) to December (20.5%, 12.3%, and 12.8%, respectively); use of noninvasive respiratory support increased from March to December. LIMITATION: COVID-NET covers approximately 10% of the U.S. population; findings may not be generalizable to the entire country. CONCLUSION: Rates of COVID-19–associated hospitalization, ICU admission, and death were highest in December 2020, corresponding with the third peak of the U.S. pandemic. The frequency of intensive interventions for management of hospitalized patients decreased over time. These data provide a longitudinal assessment of clinical trends among adults hospitalized with COVID-19 before widespread implementation of COVID-19 vaccines. PRIMARY FUNDING SOURCE: Centers for Disease Control and Prevention.

A minimum set of data are collected for all identified case patients, including site, age, sex, race/ethnicity, hospital admission date, and SARS-CoV-2 testing data, to produce hospitalization rates (https://gis.cdc.gov/grasp /covidnet/covid19_3.html). Rates are calculated using 2019 bridged-race postcensal estimates of the population from the National Center for Health Statistics (NCHS) for the counties included in surveillance (24).

Race/ethnicity is categorized using NCHS classifications as follows: non-Hispanic White, non-Hispanic Black, Hispanic or Latino (here called Hispanic), non-Hispanic Asian or Pacific Islander, and non-Hispanic American Indian or Alaska Native. Data are also collected on persons of more than 1 race/ethnicity. If race was unknown but ethnicity was Hispanic, patients were classified as Hispanic. If race was documented but ethnicity was unknown (6.6%), non-Hispanic ethnicity was assumed.

Trained surveillance staff collect detailed clinical data on a sample of hospitalized case patients through medical record abstractions using a standardized case report form (Supplement Figure, available at Annals.org). Data collected for sampled case patients include underlying medical conditions, ICU status, clinical interventions, and in-hospital outcomes.

To produce random samples of hospitalized patients for medical record abstraction, random numbers (1 through 100) are automatically generated and assigned to each case patient on entry into the surveillance database. The CDC provides sites with lists of random numbers associated with sampled case patients, stratified by time period (March to May 2020; June to September 2020; and monthly thereafter for October, November, and December 2020), site, and age group. Sample sizes vary by surveillance time period, site, and age group and are based on the total number of cases identified in each stratum.

For this analysis, sample sizes were generated to allow monthly estimation of the prevalence of clinical variables in 3 adult age groups (18 to 49 years, 50 to 64 years, and ≥65 years), and separately by 3 racial/ethnic groups (non-Hispanic White, non-Hispanic Black, and Hispanic). Because of small sample sizes, we could not generate monthly estimates for other racial/ethnic groups. Sample sizes were generated to produce relative SEs less than 0.3 for clinical variables, with monthly prevalence estimates of at least 10% (all estimates with relative SE >0.3 are noted in the Supplement Tables, available at Annals.org). Sample weights were applied to reflect the probability of a case being sampled for medical record abstraction within each specified stratum.

We calculated unadjusted monthly rates of hospitalization per 100 000 persons in the population by taking the total number of hospitalized case patients in COVID-NET each month, overall and within each age group, sex, and site, as the numerator and dividing by NCHS population estimates as the denominator.

We also calculated unadjusted monthly rates of ICU admission and death per 100 000 persons in the population. Because ICU admission and death statuses were available only for sampled hospitalized patients, we used weighted numbers of sampled case patients with ICU admission or death as the numerator and NCHS population estimates as the denominator.

To produce robust rates of hospitalization, ICU admission, and death stratified by both age and race/ethnicity, we combined data from several months (that is, March to May, June to September, and October to December) and calculated rates as described in the previous paragraphs.

Among the sample of hospitalized case patients with detailed medical record abstractions, we examined monthly trends in the proportion with selected characteristics, interventions, in-hospital outcomes, and median length of hospital stay.

Facility residence at admission was defined as residence in a rehabilitation facility, assisted living or residential care facility, group home, nursing home, skilled-nursing facility, long-term care facility, long-term acute care hospital, alcohol or drug treatment center, or psychiatric facility. Health care workers were defined as all paid and unpaid persons serving in health care settings who have the potential for direct or indirect exposure to patients or infectious materials. Case patients were defined as having 0, 1, 2, or 3 or more underlying condition categories rather than individual conditions (Supplement Figure) . Any medication prescribed for treatment of SARS-CoV-2 infection, including investigational ORIGINAL RESEARCH Clinical Trends in COVID-19 Hospitalizations agents or treatment through a clinical trial, was classified as a COVID-19-associated treatment. Invasive mechanical ventilation, bilevel positive airway pressure, continuous positive airway pressure, and high-flow nasal cannula were defined on the basis of the highest level of respiratory support received. Vasopressor use was included only if administered as a continuous infusion. Renal replacement therapy during hospitalization was described regardless of its receipt before hospitalization. In-hospital death was defined as death from any cause among case patients hospitalized with COVID-19.

We used data from all case patients who were aged 18 years or older and were hospitalized with COVID-19 during 1 March to 31 December 2020 to describe demographics of hospitalized case patients and hospitalization rates by age, sex, race/ethnicity, and site. All other analyses were limited to sampled hospitalized patients for whom medical record abstractions were completed and a discharge disposition was known.

Data for sampled case patients were analyzed using SAS survey procedures to account for sampling weights. We present proportions by month with 95% CIs for binary measures and medians with interquartile ranges for continuous measures. The Taylor series linearization method was used for variance estimation. For analyses using sampled data, unweighted case counts and weighted percentages were presented. All analyses were done using SAS, version 9.4 (SAS Institute).

This activity was reviewed by CDC, and its conduct was consistent with applicable federal law and CDC policy (see, for example, 45 C.F.R. part 46.102(l)(2), 21 C.F.R. part 56, 42 U.S.C. §241(d), 5 U.S.C. §552a, and 44 U.S.C. §3501 et seq.). Sites participating in COVID-NET obtained approval from their state or local institutional review board, as applicable.

This work was funded by the CDC, which was involved in the design of the study; the collection, analysis, and 

During 1 March to 31 December 2020, a total of 116 743 adults in COVID-NET aged 18 years or older were hospitalized with COVID-19. Among these 116 743 case patients, 26.4% were aged 18 to 49 years, 27.7% were aged 50 to 64 years, and 46.0% were aged 65 years or older ( Table 1) . Overall, 50.7% were male, 40.8% were non-Hispanic White, 26.9% were non-Hispanic Black, and 20.0% were Hispanic; race/ethnicity was unknown in 6.8% of cases. The distribution of hospitalized case patients varied by site.

The median age of hospitalized case patients was 62.0 years; median age decreased from 61 years in March to 55 years in June but subsequently increased to 65 years in December (Supplement Table 1 , available at Annals.org).

Monthly rates of COVID-19-associated hospitalization peaked in April, July, and December 2020 ( Figure 1, A) ; overall rates ranged from 20.2 (March) to 105.3 (December) per 100 000 persons in the population (Supplement Table 2 [A], available at Annals.org) and were highest among adults aged 65 years or older ( Figure 1 [B]; Supplement Table 2 [A]). During most months, hospitalization rates were higher (rate ratio, 1.0 to 1.3) among men than women (Supplement Table 2 [B], available at Annals.org). Hospitalization rates varied by site and month; most sites had peak rates in November or December 2020 (Supplement Table 2 [C], available at Annals.org).

Of the 116 743 adults hospitalized with COVID-19 during March to December 2020, we included 18 508 

Clinical Trends in COVID-19 Hospitalizations sampled case patients with completed medical record abstractions and a discharge disposition in clinical analyses. Demographic characteristics of sampled patients were similar to those of all hospitalized patients overall ( Table 1 ) and across months (Supplement Table 1 ).

Peaks in monthly rates of ICU admission and in-hospital death associated with COVID-19 mirrored that of hospitalization ( Figure 1 [A]; Supplement Table 3 [A], available at Annals.org). Although rates of ICU admission and death increased with age, trends over time were similar across all 3 age groups (Figure 1 [C Table 5 , available at Annals.org). Tocilizumab was used infrequently (2.5% overall); use peaked at 7.8% in April and decreased to 0.4% in December.

The median length of stay for hospitalized patients (including those discharged alive and those who died in the hospital) decreased from 6.4 days in March to 4.7 days in July, then remained relatively unchanged from July through December ( Figure 2 [A]; Supplement Table  6 , available at Annals.org). The percentage of hospitalized patients admitted to the ICU decreased from 37.8% in March to 20.5% in December (Figure 2 [B]; Supplement Table 6 ), with decreases across all age groups. Vasopressor use similarly decreased from 22.7% in March to 12.8% in December (Figure 2 [C]; Supplement Table 6 ). Overall, 11.4% of case patients hospitalized with COVID-19 died in the hospital; the percentage of deaths decreased from a high of 17.1% in April to a low of 7.1% in June but remained stable or increased from July through December (Figure 2 [D]; Supplement Table 6 ). Among case patients who were discharged alive, 3.1% were discharged to hospice (range, 1.2% in March to 4.1% in December) (data not shown). Among adults aged 65 years or older, 6.7% were discharged to hospice.

The percentage of hospitalized case patients receiving invasive mechanical ventilation as the highest level of respiratory support decreased from 27.8% in March to 12.3% in December (Figure 3 [A] ; Supplement Table 6 ); trends were similar across age groups (Figure 3, B) . The 

Clinical Trends in COVID-19 Hospitalizations percentage receiving oxygen via a high-flow nasal cannula increased from 5.6% in March to 8.7% in December (Figure 3 , A) but varied over time and across age groups (Figure 3, C) . The percentage receiving bilevel positive airway pressure or continuous positive airway pressure similarly increased from 2.5% in March to 7.5% in December (Figure 3 , A) and varied across age groups (Figure 3, D) .

Using a large, geographically diverse, population-based surveillance network for COVID-19-associated hospitalizations during March to December 2020, we found that monthly rates of COVID-19-associated hospitalization, ICU admission, and in-hospital death were highest in December, corresponding with the third peak of the U.S. pandemic. Rates of all 3 outcomes were higher among older than younger adults, men than women, and Hispanic and non-Hispanic Black persons than non-Hispanic White persons across the entire surveillance period. Among hospitalized case patients, monthly trends in the percentage receiving clinically effective interventions, such as systemic steroids, remdesivir, and noninvasive respiratory support, increased.

Over the same period, the percentage requiring intensive interventions, such as ICU-level care, invasive mechanical ventilation, and vasopressor support, decreased. Although the median age of hospitalized case patients and the percentage of in-hospital deaths decreased during the first 4 months of surveillance, median age subsequently increased, and the percentage of in-hospital deaths remained stable or increased during July through December. These data provide a longitudinal assessment of clinical trends among adults hospitalized with COVID-19 before widespread implementation of COVID-19 vaccines.

Community-level incidence of COVID-19 was likely the largest contributor to observed trends in rates of hospitalization, ICU admission, and in-hospital death associated with COVID-19 (14); monthly changes in COVID-19 incidence (https://covid.cdc.gov/covid-data-tracker/#datatracker-home) corresponded with monthly changes in COVID-NET hospitalization rates. Rates of all 3 outcomes varied by age, sex, and race/ethnicity. Similar to other studies, we found that older age (3-10) and male sex (25) (26) (27) were associated with higher rates of severe COVID-19 throughout the surveillance period. Although we did not adjust for such factors as age and underlying conditions, we found that men had up to 2 times the rate of ICU admission and 3 times the rate of in-hospital death 

Annals.org Annals of Internal Medicine of women. A recent global meta-analysis found that although men and women had a similar number of SARS-CoV-2 infections, male sex was associated with increased odds of admission to an intensive therapy unit (odds ratio, 2.84) and death (odds ratio, 1.39) (26) . Although potential reasons for the association between male sex and severe COVID-19 are still being elucidated, theories include sex-related differences in adaptive immunity to viral infections; differences in expression of angiotensin-converting enzyme, which facilitates SARS-CoV-2 viral entry; and sex-based differences in the prevalence of comorbid conditions and risk behaviors (such as tobacco and alcohol use) that may be associated with severe COVID-19 (26, 27) . Our analysis also contributes to an expanding body of literature on racial and ethnic disparities related to COVID-19 (5, 8, (28) (29) (30) (31) (32) (33) . Rates of hospitalization, ICU admission, and in-hospital death were highest among Hispanic persons across all age groups, followed by non-Hispanic Black persons. Possible contributing factors, which we did not account for in this analysis, include geographic differences in COVID-19 incidence and care practices that may correlate with the geographic distribution of racial/ethnic groups. Inequities in social determinants of health with downstream effects on overall health of racial/ethnic minority groups (34) likely also contributed to these disparities. Other studies have similarly found increased morbidity and mortality associated with COVID-19 among non-Hispanic Black (33) and Hispanic (35, 36) persons compared with non-Hispanic White persons.

Despite the increase in rates of hospitalization, ICU admission, and death during the winter peak of the pandemic, the percentage of hospitalized case patients who required intensive interventions decreased substantially over the surveillance period, as in other studies (11, 12, 14, 15) . Evolving practices of health care providers, guided by increasing knowledge and experience in managing patients with COVID-19, likely contributed to improved patient outcomes (11, 12, 15). Although the percentage of patients receiving invasive mechanical ventilation decreased over time across all age groups, use of noninvasive respiratory support, such as bilevel positive airway pressure, continuous positive airway pressure, and high-flow nasal cannula, increased. Several studies have shown improved outcomes with prone positioning (37) and use of noninvasive ventilatory support in lieu of invasive mechanical ventilation (11, 15, 38) for patients with acute respiratory failure associated with COVID-19. Decreasing use of invasive mechanical ventilation may have also lessened the need for ICU-level care because other types of respiratory support could be delivered in non-ICU settings (39).

Increasing use of more effective treatments of COVID-19 may also have improved outcomes among patients hospitalized with COVID-19. Dexamethasone has been shown to reduce 28-day mortality among patients requiring supplemental oxygen or invasive mechanical ventilation (20) ; use of systemic corticosteroids in our analysis increased from less than 20% in March to almost 75% in December 2020. Remdesivir, which has been shown to reduce time to clinical recovery in patients with severe COVID-19 (19), was received by fewer than 2% of patients in March, but this percentage increased to more than 50% by December 2020. Meanwhile, the use of medications found to be ineffective for treatment of COVID-19 (40), such as hydroxychloroquine, decreased substantially by June 2020. Use of other immunomodulatory agents was infrequent during the period covered by our analysis. Recent data among hospitalized patients with COVID-19 pneumonia have shown reduced time to recovery and improved clinical status with baricitinib (41) and reduced likelihood of progression to mechanical ventilation or death with tocilizumab (42) ; accordingly, use of these agents in combination with dexamethasone alone or dexamethasone plus remdesivir may increase.

Observed trends in clinical interventions and outcomes may in part reflect a complex interplay between COVID-19 incidence and health care system capacity (43) . As hospitalization rates peaked in different phases and geographic locations, ICU admission rates likely also increased. However, during peaks, the percentage of hospitalized patients receiving ICU admission, mechanical ventilation, or renal replacement therapy (44) could have decreased or been capped because of the scarcity of resources. An analysis from CDC's National Healthcare Safety Network found that from April to July 2020, many hospitals reported shortages of supplies for ventilators and other resources (43) . Changes in timing of discharge and discharge disposition might also be influenced by bed capacity or the ability to discharge patients to subacute or long-term care facilities. In our analysis, almost 7% of patients aged 65 years or older were discharged to hospice. Among all case patients, the percentage discharged to hospice increased from 1% in March to 4% in December 2020. Patterns in mortality might also be influenced by health care system capacity. One study done among 88 Department of Veterans Affairs hospitals found that strains on ICU capacity were correlated with higher ICU mortality associated with COVID-19 (45) . Linkage of COVID-NET data to facility-level data could allow for more direct examination of the effect of resource availability on COVID-19-associated interventions and outcomes. This analysis has several limitations. Because COVID-NET covers approximately 10% of the U.S. population, our findings may not be generalizable to the entire country. This analysis presented data for the entire network and did not account for differences across sites; several factors, including peaks in hospitalization rates and the racial/ethnic distribution of case patients, varied by site. Because SARS-CoV-2 testing was done at the discretion of health care providers, COVID-NET may not have captured all COVID-19-associated hospitalizations. Changes in testing practices over the course of the pandemic may have influenced trends because sicker patients were more likely to be tested early in the pandemic when testing capacity was limited. This analysis included only data on in-hospital outcomes; thus, we could not characterize the burden of COVID-19-associated deaths outside the hospital. The relative SEs for some estimates were greater than 0.3, particularly for interventions or outcomes with low prevalence, such as renal replacement therapy or in-hospital deaths among patients aged 18 to ORIGINAL RESEARCH Clinical Trends in COVID-19 Hospitalizations 49 years. Last, sample sizes were not sufficient to fully explore differences in interventions and outcomes by race/ethnicity or to produce estimates for all racial/ethnic groups, including non-Hispanic American Indian or Alaska Native and Asian or Pacific Islander patients.

In this analysis, done during the first 10 months of the U.S. pandemic and before the large-scale availability of COVID-19 vaccines, we found that monthly rates of COVID-19-associated hospitalization, ICU admission, and in-hospital death peaked in December 2020 and were highest among older adults, men, and racial/ethnic minority groups. We saw encouraging trends in the percentage of hospitalized patients receiving effective treatments of COVID-19 and in the percentage requiring intensive interventions. These trends likely reflected increasing knowledge, experience, and tools to combat COVID-19 among health care systems and providers. However, this analysis also highlights ongoing racial and ethnic disparities related to outcomes associated with severe COVID-19. It will be important to follow trends in patient characteristics, interventions, and outcomes over time to monitor the effects of clinical and public health interventions, including vaccination, and to ensure that progress is being made in closing the gap in racial and ethnic disparities related to COVID-19 in the United States.

Estimated incidence of coronavirus disease 2019 (COVID-19) illness and hospitalization-United States

Estimated disease burden of COVID-19

Covid-19 in critically ill patients in the Seattle region -case series

Epidemiology, clinical course, and outcomes of critically ill adults with COVID-19 in New York City: a prospective cohort study

Hospitalization rates and characteristics of patients hospitalized with laboratory-confirmed coronavirus disease 2019 -COVID-NET, 14 states

Clinical characteristics of Covid-19

Risk factors for intensive care unit admission and in-hospital mortality among hospitalized adults identified through the US Coronavirus Disease 2019 (COVID-19)-Associated Hospitalization Surveillance Network (COVID-NET)

COVID-NET Surveillance Team. Risk factors for coronavirus disease 2019 (COVID-19)-associated hospitalization: COVID-19-Associated Hospitalization Surveillance Network and Behavioral Risk Factor Surveillance System

Factors associated with hospital admission and critical illness among 5279 people with coronavirus disease 2019 in New York City: prospective cohort study

Characteristics, outcomes, and trends of patients with COVID-19-related critical illness at a learning health system in the United States

Trends in COVID-19 risk-adjusted mortality rates

Outcomes from intensive care in patients with COVID-19: a systematic review and meta-analysis of observational studies

Quality and Clinical Research Collaborative. Declines in mortality over time for critically ill adults with coronavirus disease

Changing age distribution of the COVID-19 pandemic -United States

Decline in COVID-19 hospitalization growth rates associated with statewide mask mandates -10 states

ACTT-1 Study Group Members. Remdesivir for the treatment of Covid-19 -final report

Dexamethasone in hospitalized patients with Covid-19

ACTT-2 Study Group Members. Baricitinib plus remdesivir for hospitalized adults with Covid-19

Association of treatment dose anticoagulation with in-hospital survival among hospitalized patients with COVID-19

Census populations with bridged race categories. National Vital Statistics System

Mortality among US patients hospitalized with SARS-CoV-2 infection in 2020

Male sex identified by global COVID-19 meta-analysis as a risk factor for death and ITU admission

Sex differences in susceptibility, severity, and outcomes of coronavirus disease 2019: crosssectional analysis from a diverse US metropolitan area

Racial disparities in COVID-19 testing and outcomes. Retrospective cohort study in an integrated health system

Characteristics and clinical outcomes of adult patients hospitalized with COVID-19 -Georgia

Assessment of COVID-19 hospitalizations by race/ethnicity in 12 states

Racial and ethnic disparities in COVID-19-related infections, hospitalizations, and deaths. A systematic review

Racial disparities in incidence and outcomes among patients with COVID-19

Centers for Disease Control and Prevention. Introduction to COVID-19 racial and ethnic health disparities

Characteristics and factors associated with COVID-19 infection, hospitalization, and mortality across race and ethnicity

Disproportionate incidence of COVID-19 infection, hospitalizations, and deaths among persons identifying as Hispanic or Latino

STOP-COVID Investigators. Prone positioning and survival in mechanically ventilated patients with coronavirus disease 2019-related respiratory failure

Helmet CPAP to treat hypoxic pneumonia outside the ICU: an observational study during the COVID-19 outbreak

ACTT-2 Study Group Members. Baricitinib plus remdesivir for hospitalized adults with Covid-19

Tocilizumab in patients hospitalized with Covid-19 pneumonia

National Healthcare Safety Network. Hospital capacities and shortages of healthcare resources among US hospitals during the coronavirus disease 2019 (COVID-19) pandemic, National Healthcare Safety Network (NHSN)

Estimating shortages in capacity to deliver continuous kidney replacement therapy during the COVID-19 pandemic in the United States

Association of intensive care unit patient load and demand with mortality rates in US Department of Veterans Affairs hospitals during the COVID-19 pandemic

Clinical Trends in COVID-19 Hospitalizations ORIGINAL RESEARCH

California Emerging Infections Program, 360 22nd Street

Communicable Disease Epidemiology Program, Colorado Department of Public Health and Environment, 4300 Cherry Creek South Drive

Yale Emerging Infections Program, One Church Street, 7th Floor

Georgia Emerging Infections Program

Iowa Department of Public Health, 321 East 12th Street

Bureau of Epidemiology and Population Health, 333 South Grand Avenue

Vaccine Preventable Disease Section, Minnesota Department of Health, PO Box 64975

New Mexico Department of Health, Infectious Disease Epidemiology Bureau, Epidemiology and Response Division, 1190 St. Francis Drive/ PO Box 26110

Room 651 Corning Tower

URMC -Center for Community Health, Emerging Infections Program, 46 Prince Street

Ohio Department of Health

Office of Disease Prevention and Epidemiology, Oregon Public Health Division, 800 NE Oregon Street, Suite 772

Vanderbilt University School of Medicine, 1161 21st Avenue South

Village at Vanderbilt, Suite 2100, 1500 21st Avenue South

Critical revision of the article for important intellectual content

Administrative, technical, or logistic support

Appendix Figure. Rates (with 95% CIs) of COVID-19-associated hospitalization, ICU admission, and in-hospital death, by age and race/ethnicity.