Risk Factors Associated With SARS-CoV-2 Infections, Hospitalization, and Mortality Among US Nursing Home Residents | Geriatrics | JAMA Network Open | JAMA Network
[Skip to Navigation]
Sign In
Table 1.  Intraclass Correlation Coefficient at County and Nursing Home Facility Level From 3-Level Logistic Regression Models Estimating SARS-CoV-2 Infection, Hospitalization, and Mortality Within 30 Days of Diagnosis
Intraclass Correlation Coefficient at County and Nursing Home Facility Level From 3-Level Logistic Regression Models Estimating SARS-CoV-2 Infection, Hospitalization, and Mortality Within 30 Days of Diagnosis
Table 2.  Results From a Competing Risk Regression Model and a Conditional Competing Risk Regression Model Conditioned on Nursing Home for Resident Characteristics Associated With SARS-Cov-2 Infection
Results From a Competing Risk Regression Model and a Conditional Competing Risk Regression Model Conditioned on Nursing Home for Resident Characteristics Associated With SARS-Cov-2 Infection
Table 3.  Results from a Conditional Competing Risk Model Conditioned on Facility for Resident Characteristics Associated With Acute Hospitalization 30 Days After SARS-CoV-2 Infection
Results from a Conditional Competing Risk Model Conditioned on Facility for Resident Characteristics Associated With Acute Hospitalization 30 Days After SARS-CoV-2 Infection
Table 4.  Results from a Conditional Competing Risk Model Conditioned on Facility for Resident Characteristics Associated With Mortality 30 Days After SARS-CoV-2 Infection
Results from a Conditional Competing Risk Model Conditioned on Facility for Resident Characteristics Associated With Mortality 30 Days After SARS-CoV-2 Infection
1.
Johns Hopkins University and Medicine. COVID-19 dashboard by the Center for Systems Science and Engineering at Johns Hopkins University. Accessed November 15, 2020. https://coronavirus.jhu.edu/map.html
2.
The New York Times. More than one-third of U.S. coronavirus deaths are linked to nursing homes. The New York Times. Updated February 26, 2021. Accessed November 15, 2020. https://www.nytimes.com/interactive/2020/us/coronavirus-nursing-homes.html
3.
Kaiser Family Foundation. State COVID-19 data and policy actions. Updated March 2, 2021. Accessed November 15, 2020. https://www.kff.org/coronavirus-covid-19/issue-brief/state-covid-19-data-and-policy-actions/
4.
Williamson  EJ, Walker  AJ, Bhaskaran  K,  et al.  Factors associated with COVID-19-related death using OpenSAFELY.   Nature. 2020;584(7821):430-436. doi:10.1038/s41586-020-2521-4 PubMedGoogle ScholarCrossref
5.
Ioannou  GN, Locke  E, Green  P,  et al.  Risk factors for hospitalization, mechanical ventilation, or death among 10131 US veterans with SARS-CoV-2 infection.   JAMA Netw Open. 2020;3(9):e2022310. doi:10.1001/jamanetworkopen.2020.22310 PubMedGoogle Scholar
6.
Holman  N, Knighton  P, Kar  P,  et al.  Risk factors for COVID-19–related mortality in people with type 1 and type 2 diabetes in England: a population-based cohort study.   Lancet Diabetes Endocrinol. 2020;8(10):823-833. doi:10.1016/S2213-8587(20)30271-0 PubMedGoogle ScholarCrossref
7.
Gorges  RJ, Konetzka  RT.  Staffing levels and COVID-19 cases and outbreaks in U.S. nursing homes.   J Am Geriatr Soc. 2020;68(11):2462-2466. doi:10.1111/jgs.16787 PubMedGoogle ScholarCrossref
8.
Li  Y, Cen  X, Cai  X, Temkin-Greener  H.  Racial and ethnic disparities in COVID-19 infections and deaths across U.S. nursing homes.   J Am Geriatr Soc. 2020;68(11):2454-2461. doi:10.1111/jgs.16847 PubMedGoogle ScholarCrossref
9.
Li  Y, Temkin-Greener  H, Shan  G, Cai  X.  COVID-19 infections and deaths among Connecticut nursing home residents: facility correlates.   J Am Geriatr Soc. 2020;68(9):1899-1906. doi:10.1111/jgs.16689 PubMedGoogle ScholarCrossref
10.
Figueroa  JF, Wadhera  RK, Papanicolas  I,  et al.  Association of nursing home ratings on health inspections, quality of care, and nurse staffing with COVID-19 cases.   JAMA. 2020;324(11):1103-1105. doi:10.1001/jama.2020.14709 PubMedGoogle ScholarCrossref
11.
De Smet  R, Mellaerts  B, Vandewinckele  H,  et al.  Frailty and mortality in hospitalized older adults with COVID-19: retrospective observational study.   J Am Med Dir Assoc. 2020;21(7):928-932.e1. doi:10.1016/j.jamda.2020.06.008PubMedGoogle ScholarCrossref
12.
Mendes  A, Serratrice  C, Herrmann  FR,  et al.  Predictors of in-hospital mortality in older patients with COVID-19: The COVIDAge Study.   J Am Med Dir Assoc. 2020;21(11):1546-1554.e3. doi:10.1016/j.jamda.2020.09.014PubMedGoogle ScholarCrossref
13.
Centers for Medicare & Medicaid Services. Minimum Data Set (MDS) 3.0 for nursing homes and swing bed providers. Accessed February 10, 2020. https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/NursingHomeQualityInits/NHQIMDS30
14.
Goodwin  JS, Li  S, Zhou  J, Graham  JE, Karmarkar  A, Ottenbacher  K.  Comparison of methods to identify long term care nursing home residence with administrative data.   BMC Health Serv Res. 2017;17(1):376. doi:10.1186/s12913-017-2318-9 PubMedGoogle ScholarCrossref
15.
Intrator  O, Hiris  J, Berg  K, Miller  SC, Mor  V.  The residential history file: studying nursing home residents’ long-term care histories(*).   Health Serv Res. 2011;46(1 Pt 1):120-137. doi:10.1111/j.1475-6773.2010.01194.x PubMedGoogle ScholarCrossref
16.
Centers for Disease Control and Prevention. International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM). Accessed March 2, 2021. https://www.cdc.gov/nchs/icd/icd10cm.htm
17.
Thomas  KS, Dosa  D, Wysocki  A, Mor  V.  The Minimum Data Set 3.0 Cognitive Function Scale.   Med Care. 2017;55(9):e68-e72. doi:10.1097/MLR.0000000000000334 PubMedGoogle ScholarCrossref
18.
Kroenke  K, Spitzer  RL, Williams  JB.  The PHQ-9: validity of a brief depression severity measure.   J Gen Intern Med. 2001;16(9):606-613. doi:10.1046/j.1525-1497.2001.016009606.x PubMedGoogle ScholarCrossref
19.
Saliba  D, DiFilippo  S, Edelen  MO, Kroenke  K, Buchanan  J, Streim  J.  Testing the PHQ-9 interview and observational versions (PHQ-9 OV) for MDS 3.0.   J Am Med Dir Assoc. 2012;13(7):618-625. doi:10.1016/j.jamda.2012.06.003 PubMedGoogle ScholarCrossref
20.
Perlman  CM, Hirdes  JP.  The aggressive behavior scale: a new scale to measure aggression based on the minimum data set.   J Am Geriatr Soc. 2008;56(12):2298-2303. doi:10.1111/j.1532-5415.2008.02048.x PubMedGoogle ScholarCrossref
21.
Fine  JP, Gray  RJ.  A proportional hazards model for subdistribution of a competing risk.   J Am Stat Assoc. 1999;94:496-509. doi:10.1080/01621459.1999.10474144 Google ScholarCrossref
22.
Zhou  B, Latouche  A, Rocha  V, Fine  J.  Competing risks regression for stratified data.   Biometrics. 2011;67(2):661-670. doi:10.1111/j.1541-0420.2010.01493.x PubMedGoogle ScholarCrossref
23.
Muff  S, Held  L, Keller  LF.  Marginal or conditional regression models for correlated non-normal data?   Methods Ecol Evol. 2016;7(12):1514-1524. doi:10.1111/2041-210X.12623 Google ScholarCrossref
24.
Breslow  NE, Day  NE, Halvorsen  KT, Prentice  RL, Sabai  C.  Estimation of multiple relative risk functions in matched case-control studies.   Am J Epidemiol. 1978;108(4):299-307. doi:10.1093/oxfordjournals.aje.a112623 PubMedGoogle ScholarCrossref
25.
Singh  S, Lin  YL, Kuo  YF, Nattinger  AB, Goodwin  JS.  Variation in the risk of readmission among hospitals: the relative contribution of patient, hospital and inpatient provider characteristics.   J Gen Intern Med. 2014;29(4):572-578. doi:10.1007/s11606-013-2723-7 PubMedGoogle ScholarCrossref
26.
Figueroa  JF, Wadhera  RK, Lee  D, Yeh  RW, Sommers  BD.  Community-level factors associated with racial And ethnic disparities in COVID-19 rates In Massachusetts.   Health Aff (Millwood). 2020;39(11):1984-1992. doi:10.1377/hlthaff.2020.01040 PubMedGoogle ScholarCrossref
27.
Abrams  HR, Loomer  L, Gandhi  A, Grabowski  DC.  Characteristics of U.S. nursing homes with COVID-19 cases.   J Am Geriatr Soc. 2020;68(8):1653-1656. doi:10.1111/jgs.16661 PubMedGoogle ScholarCrossref
28.
Chen  MK, Chevalier  JA, Long  EF.  Nursing Home Staff Networks and COVID-19. National Bureau of Economic Research; 2020;898-2937.
29.
Ritchie  H, Ortiz-Ospina  E, Beltekian  D,  et al. Coronavirus disease 2019 (COVID-19) deaths. Our World in Data. Updated March 2, 2021. Accessed February 10, 2020. https://ourworldindata.org/covid-deaths?country=USA
30.
Centers for Medicare & Medicaid Services. COVID-19 nursing home dataset. Accessed February 10, 2021. https://data.cms.gov/Special-Programs-Initiatives-COVID-19-Nursing-Home/COVID-19-Nursing-Home-Dataset/s2uc-8wxp
31.
Shi  SM, Bakaev  I, Chen  H, Travison  TG, Berry  SD.  Risk factors, presentation, and course of Coronavirus Disease 2019 in a large, academic long-term care facility.   J Am Med Dir Assoc. 2020;21(10):1378-1383.e1. doi:10.1016/j.jamda.2020.08.027PubMedGoogle ScholarCrossref
32.
Petrilli  CM, Jones  SA, Yang  J,  et al.  Factors associated with hospital admission and critical illness among 5279 people with coronavirus disease 2019 in New York City: prospective cohort study.   BMJ. 2020;369:m1966. doi:10.1136/bmj.m1966 PubMedGoogle ScholarCrossref
33.
Price-Haywood  EG, Burton  J, Fort  D, Seoane  L.  Hospitalization and mortality among Black patients and White patients with COVID-19.   N Engl J Med. 2020;382(26):2534-2543. doi:10.1056/NEJMsa2011686 PubMedGoogle ScholarCrossref
34.
Czaja  CA, Miller  L, Alden  N,  et al.  Age-related differences in hospitalization rates, clinical presentation, and outcomes among older adults hospitalized with influenza—U.S: Influenza Hospitalization Surveillance Network (FluSurv-NET).   Open Forum Infect Dis. 2019;6(7):ofz225. doi:10.1093/ofid/ofz225 PubMedGoogle Scholar
35.
Sze  S, Pan  D, Nevill  CR,  et al.  Ethnicity and clinical outcomes in COVID-19: a systematic review and meta-analysis.   EClinicalMedicine. 2020;29:100630. doi:10.1016/j.eclinm.2020.100630 PubMedGoogle Scholar
36.
Yan  BW, Ng  F, Chu  J, Tsoh  J, Nguyen  T. Asian Americans facing high COVID-19 case fatality. Health Affairs Blog. July 13, 2020. Accessed February 10, 2021. https://www.healthaffairs.org/do/10.1377/hblog20200708.894552/full/
37.
Li  Y, Yin  J, Cai  X, Temkin-Greener  J, Mukamel  DB.  Association of race and sites of care with pressure ulcers in high-risk nursing home residents.   JAMA. 2011;306(2):179-186. doi:10.1001/jama.2011.942 PubMedGoogle ScholarCrossref
38.
Hendren  NS, de Lemos  JA, Ayers  C,  et al.  Association of body mass index and age with morbidity and mortality in patients hospitalized with COVID-19: results from the American Heart Association COVID-19 Cardiovascular Disease Registry.   Circulation. 2021;143(2):135-144. doi:10.1161/CIRCULATIONAHA.120.051936 PubMedGoogle ScholarCrossref
39.
Klang  E, Kassim  G, Soffer  S, Freeman  R, Levin  MA, Reich  DL.  Severe obesity as an independent risk factor for COVID-19 mortality in hospitalized patients younger than 50.   Obesity (Silver Spring). 2020;28(9):1595-1599. doi:10.1002/oby.22913 PubMedGoogle ScholarCrossref
40.
Recalde  M, Pistillo  A, Fernandez-Bertolin  S,  et al  Body mass index and risk of COVID-19 diagnosis, hospitalisation, and death: a population-based multi-state cohort analysis including 2,524,926 people in Catalonia, Spain.   medRxiv. Preprint posted online November 28, 2020. doi:10.1101/2020.11.25.20237776Google Scholar
41.
Tartof  SY, Qian  L, Hong  V,  et al.  Obesity and mortality among patients diagnosed with COVID-19: results from an integrated health care organization.   Ann Intern Med. 2020;173(10):773-781. doi:10.7326/M20-3742 PubMedGoogle ScholarCrossref
42.
Livingston  G, Rostamipour  H, Gallagher  P,  et al.  Prevalence, management, and outcomes of SARS-CoV-2 infections in older people and those with dementia in mental health wards in London, UK: a retrospective observational study.   Lancet Psychiatry. 2020;7(12):1054-1063. doi:10.1016/S2215-0366(20)30434-X PubMedGoogle ScholarCrossref
43.
Reyes-Bueno  JA, Mena-Vázquez  N, Ojea-Ortega  T,  et al.  Case fatality of COVID-19 in patients with neurodegenerative dementia.   Neurologia. 2020;35(9):639-645. doi:10.1016/j.nrl.2020.07.005 PubMedGoogle ScholarCrossref
44.
Atkins  JL, Masoli  JAH, Delgado  J,  et al.  Preexisting comorbidities predicting COVID-19 and mortality in the UK Biobank Community cohort.   J Gerontol A Biol Sci Med Sci. 2020;75(11):2224-2230. doi:10.1093/gerona/glaa183 PubMedGoogle ScholarCrossref
45.
Harrison  SL, Fazio-Eynullayeva  E, Lane  DA, Underhill  P, Lip  GYH.  Comorbidities associated with mortality in 31,461 adults with COVID-19 in the United States: a federated electronic medical record analysis.   PLoS Med. 2020;17(9):e1003321. doi:10.1371/journal.pmed.1003321 PubMedGoogle Scholar
46.
Kaiser Family Foundation. A dozen facts about Medicare Advantage in 2020. Updated January 13, 2021. Accessed December 7, 2020. https://www.kff.org/medicare/issue-brief/a-dozen-facts-about-medicare-advantage-in-2020/
47.
Shioda  K, Weinberger  DM, Mori  M.  Navigating through health care data disrupted by the COVID-19 pandemic.   JAMA Intern Med. 2020;180(12):1569-1570. doi:10.1001/jamainternmed.2020.5542 PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Views 4,952
    Citations 0
    Original Investigation
    Geriatrics
    March 31, 2021

    Risk Factors Associated With SARS-CoV-2 Infections, Hospitalization, and Mortality Among US Nursing Home Residents

    Author Affiliations
    • 1Center for Drug Safety and Effectiveness, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland
    • 2Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland
    • 3Sealy Center on Aging, Department of Internal Medicine, The University of Texas Medical Branch, Galveston
    JAMA Netw Open. 2021;4(3):e216315. doi:10.1001/jamanetworkopen.2021.6315
    Key Points

    Question  What risk factors are associated with SARS-CoV-2 infections, hospitalization, and mortality among nursing home residents?

    Findings  In this cohort study among 482 323 long-stay residents, risk of SARS-CoV-2 infections were associated with geographic area and the specific facility, not by characteristics of the residents. Among residents diagnosed with SARS-CoV-2 infections, the risk of hospitalization associated with individual resident characteristics differed from the risk of death.

    Meaning  These findings suggest that decisions on hospitalization of nursing home residents with SARS-CoV-2 were inconsistently associated with risk of death.

    Abstract

    Importance  Nursing home residents account for approximately 40% of deaths from SARS-CoV-2.

    Objective  To identify risk factors for SARS-CoV-2 incidence, hospitalization, and mortality among nursing home residents in the US.

    Design, Setting, and Participants  This retrospective longitudinal cohort study was conducted in long-stay residents aged 65 years or older with fee-for-service Medicare residing in 15 038 US nursing homes from April 1, 2020, to September 30, 2020. Data were analyzed from November 22, 2020, to February 10, 2021.

    Main Outcomes and Measures  The main outcome was risk of diagnosis with SARS-CoV-2 (per International Statistical Classification of Diseases, Tenth Revision, Clinical Modification [ICD-10-CM] codes) by September 30 and hospitalization or death within 30 days after diagnosis. Three-level (resident, facility, and county) logistic regression models and competing risk models conditioned on nursing home facility were used to determine association of patient characteristics with outcomes.

    Results  Among 482 323 long-stay residents included, the mean (SD) age was 82.7 (9.2) years, with 326 861 (67.8%) women, and 383 838 residents (79.6%) identifying as White. Among 137 119 residents (28.4%) diagnosed with SARS-CoV-2 during follow up, 29 204 residents (21.3%) were hospitalized, and 26 384 residents (19.2%) died within 30 days. Nursing homes explained 37.2% of the variation in risk of infection, while county explained 23.4%. Risk of infection increased with increasing body mass index (BMI; calculated as weight in kilograms divided by height in meters squared) (eg, BMI>45 vs BMI 18.5-25: adjusted hazard ratio [aHR], 1.19; 95% CI, 1.15-1.24) but varied little by other resident characteristics. Risk of hospitalization after SARS-CoV-2 increased with increasing BMI (eg, BMI>45 vs BMI 18.5-25: aHR, 1.40; 95% CI, 1.28-1.52); male sex (aHR, 1.32; 95% CI, 1.29-1.35); Black (aHR, 1.28; 95% CI, 1.24-1.32), Hispanic (aHR, 1.20; 95% CI, 1.15-1.26), or Asian (aHR, 1.46; 95% CI, 1.36-1.57) race/ethnicity; impaired functional status (eg, severely impaired vs not impaired: aHR, 1.15; 95% CI, 1.10-1.22); and increasing comorbidities, such as renal disease (aHR, 1.21; 95% CI, 1.18-1.24) and diabetes (aHR, 1.16; 95% CI, 1.13-1.18). Risk of mortality increased with age (eg, age >90 years vs 65-70 years: aHR, 2.55; 95% CI, 2.44-2.67), impaired cognition (eg, severely impaired vs not impaired: aHR, 1.79; 95% CI, 1.71-1.86), and functional impairment (eg, severely impaired vs not impaired: aHR, 1.94; 1.83-2.05).

    Conclusions and Relevance  These findings suggest that among long-stay nursing home residents, risk of SARS-CoV-2 infection was associated with county and facility of residence, while risk of hospitalization and death after SARS-CoV-2 infection was associated with facility and individual resident characteristics. For many resident characteristics, there were substantial differences in risk of hospitalization vs mortality. This may represent resident preferences, triaging decisions, or inadequate recognition of risk of death.

    Introduction

    While 5% of US SARS-CoV-2 infections have occurred in nursing home residents, they account for almost 40% of deaths.1-3 The case fatality rates are 5 times higher in long-stay nursing home residents than the national mean.2

    Large cohort studies conducted in community-dwelling adults have identified important risk factors for SARS-CoV-2–related hospitalization and deaths, such as advanced age, male sex, and comorbidities.4-6 Nursing home residents typically are very old and frail, have more comorbidities and cognitive dysfunction, and are dependent in activities of daily living. Therefore, risk factors for SARS-CoV-2 outcomes may differ for nursing home residents.

    Ecological studies conducted at the nursing home level have explored the role of resident and nursing home characteristics associated with SARS-CoV-2 outcomes.7-10 Mixed evidence exists that facilities with higher percentages of racial/ethnic minorities, such as Black and Hispanic individuals,8 lower nurse staffing,7,10 and lower ratings for quality were associated with higher rates of SARS-CoV-2 cases and deaths.9 Individual resident characteristics, such as cognitive and functional status, were not evaluated in prior studies. Patients with impaired cognition or functional status may be at increased risk of SARS-CoV-2 infection because they require more assistance from staff members. There is a lack of large-scale resident-level studies among nursing homes to comprehensively describe risk factors for SARS-CoV-2 infection and outcomes.11,12

    We used national data from long-stay nursing home residents in the US to identify risk factors for SARS-CoV-2 infections and for hospitalization and mortality after SARS-CoV-2 infections. We were interested in whether the factors associated with hospitalization and death from SARS-CoV-2 among community-dwelling populations were similar to those in long-term care.

    Methods

    The cohort study was approved by the University of Texas Medical Branch institutional review board complies with the Centers for Medicare & Medicaid Services (CMS) Data Use Agreement requirements, which waived the need for informed consent for use of the study data because data were deidentified. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies.

    Data Source

    We used the Minimum Data Set (MDS) version 3.0, a federally mandated standardized resident assessment13 and linked Medicare claims data for 100% of US nursing home residents from January 1 to December 31, 2020, last updated on January 31, 2021. The MDS contains information on resident clinical, psychosocial, and functional characteristics. We used the Medicare Beneficiary Summary File for demographic and enrollment information, the Carrier File for physician claims, the Outpatient Standard Analytic File (SAF) for outpatient claims, the Hospital SAF for hospitalization, and skilled nursing facility claims.

    Study Population

    We identified long stay residents aged 65 years and older residing in nursing homes as of April 1, 2020 (eFigure in the Supplement). Using a previously validated approach,14,15 we identified nursing home stays based on the MDS data and excluded any skilled nursing facility care during that stay. The Center for Disease Control and Prevention established the International Statistical Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM)16 code U07.1 for SARS-CoV-2 on April 1, 2020. We excluded residents if they were diagnosed with SARS-CoV-2 before April 1, 2020 using ICD-10-CM codes of J12.89, J20.8, J40, J22 J98.8, J80 combined with B97.29, or U07.1 to identify SARS-CoV-2. We restricted nursing home residents to those with continuous enrollment in Medicare Parts A and B with no enrollment in health maintenance organizations from April 1, 2020, until SARS-CoV-2 diagnosis, death, or the study end date on September 30, 2020. Most SARS-CoV-2 claims came from the Carrier File, followed by outpatient, inpatient, and skilled nursing facility files (eTable 1 in the Supplement). Nearly half of the patients with SARS-CoV-2 infection had 1 or 2 claims, and one-fourth had 7 or more claims (eTable 2 in the Supplement).

    Outcomes

    We examined 3 outcomes: new diagnosis of SARS-CoV-2 infection until September 30, 2020, hospitalization within 30 days of SARS-CoV-2 diagnosis, and death within 30 days of SARS-CoV-2 diagnosis. For hospitalization and mortality outcomes, we followed-up patients until October 31, 2020, using claims last updated on January 31, 2021. We identified the first diagnosis of SARS-CoV-2 from the Carrier File, Outpatient SAF, Hospital SAF, or nursing facility files using ICD-10-CM code U07.1. Hospitalization was identified from the Hospital SAF file, and death was identified from the Medicare Beneficiary Summary File.

    Resident Characteristics

    We included characteristics if there were a priori reasons why they might be associated with increased risk of SARS-CoV-2 infection, such as a condition that might necessitate more physical contact by staff or that might interfere with following instructions on social distancing. We also included characteristics associated with risk of hospitalization or death in prior studies. We identified resident characteristics from the resident’s most recent MDS assessment prior to April 1, 2020, including resident age, sex, race/ethnicity, body mass index (BMI; calculated as weight in kilograms divided by height in meters squared), cognitive function (categorized as cognitively intact and mild, moderate, and severe impairment),17 mood (categorized as no depression, minimal to mild depression, and moderate to severe depression),18,19 hallucinations or delusions or aggressive behavior (yes or no),20 activities of daily living score (sum of 0-4 scores for 8 activities of daily living items: 0-8 indicated no dependence; 9-16, mild dependence; 17-24, moderate dependence; and 25-32, severe dependence), use of tube or catheter (yes or no), physician prognosis of life expectancy of less than 6 months, as well as diagnosis for respiratory disease, cancer, heart disease, diabetes, renal disease, malnutrition, and neurologic conditions.

    Statistical Analysis

    SARS-CoV-2 infections vary widely across geographic regions and among nursing homes.1-3,9 To describe the variation, for each outcome we constructed a 3-level logistic regression model based on resident, nursing home facility, and county level that estimated the intraclass correlation coefficient at the county and nursing home levels with and without controlling for resident characteristics.

    To estimate the association of resident characteristics with SARS-CoV-2 diagnosis between April 1, 2020, and September 30, 2020, we first constructed a proportional hazards competing risk model with death as a competing risk.21 To account for geography and facility, we then constructed a conditional competing risk model conditioned on facility. In essence, individuals are then compared within the same facility. This controls for differences between facilities and differences between geographic areas, because each facility is nested in a geographic area.22-25

    For hospitalization and death, the day of diagnosis was day 0, and all residents were followed-up for 30 days. We constructed a conditional competing risk model conditioned on nursing facility for hospitalization while treating death as a competing risk and conditional Cox proportional hazards regression models for death. All resident characteristics, plus the month of diagnosis of SARS-CoV-2, were included in the models. In addition, the 3-level logistic regression model provided an alternative way to control for variation due to geography and facility.

    All analyses were performed using SAS Enterprise statistical software version 7.12 (SAS Institute). P values were 2-sided, and statistical significance was set at P < .05. Data were analyzed from November 22, 2020, to February 10, 2021.

    Results

    This cohort study included 482 323 long-stay residents at 15 038 nursing homes. The mean (SD) age was 82.7 (9.2) years, with 326 861 (67.8%) women and 383 838 residents (79.6%) identifying as White. The SARS-CoV-2 infection rate was 28.4%. Among 137 119 residents diagnosed with COVID-19, 29 206 residents (21.3%) were hospitalized, and 26 382 residents (19.2%) died within 30 days.

    Variation in Incidence, Hospitalization, and Death Among Counties and Facilities

    Table 1 shows the amount of variation in SARS-Cov-2 infection, hospitalization, and death attributed to the county and individual facilities, generated from 3-level multivariable logistic regressions models in which residents are nested in facilities and facilities nested in counties. In the null models and the models including all resident characteristics, approximately 23% of the variation in incidence was attributable to the county and 37% to the facility. County accounted for approximately 7% of the variation in risk of hospitalization and 2% in risk of death, while facility accounted for 17% of the variation in risk of hospitalization and 9% in risk of death (Table 1).

    Characteristics Associated With SARS-CoV-2 Infections

    Table 2 summarizes the resident characteristics associated with risk of SARS-CoV-2 diagnosis during follow up. Unadjusted SARS-Cov-2 infection rates and adjusted hazard ratios (aHRs) from a competing risk model and a conditional competing risk model that controlled for differences among nursing homes are presented.

    The magnitude of the HR for SARS-CoV-2 infection changed substantially for some variables in the conditional competing risk models that controlled for differences among nursing homes, compared with the competing risk model. For example, Black residents (aHR, 1.56; 95% CI, 1.54-1.58), Hispanic residents (aHR, 1.72; 95% CI, 1.59-1.76) and Asian residents (aHR, 1.63; 95% CI, 1.58-1.69) had higher risk of SARS-CoV-2 infections in the competing risk model that did not control for the variation in risk among facilities or counties. In the model conditioned on facility, which controlled for differences among facilities and also (indirectly) among counties, the aHR was reduced to 1.04 (95% CI, 1.03-1.06) for Black residents, 1.07 (95% CI, 1.05-1.10) for Hispanic residents, and 1.07 (1.03-1.11) for Asian residents.

    In the conditional model, there was a monotonic association between BMI and risk of infection (eg, compared with BMI 18.1-25, BMI<18.5: aHR, 0.88; 95% CI, 0.86-0.90 vs BMI>45: aHR, 1.19; 95% CI, 1.15-1.24). Characteristics that might indicate need for more care and staff contact, such as severe cognitive or functional impairment or having a feeding tube, intravenous line, or catheter, were associated with increased risk of infection. An estimated poor life expectancy (aHR, 0.55; 95% CI, 0.53-0.56) was the only characteristic other than BMI and race/ethnicity that was associated with a greater than 6% difference in risk of SARS-CoV-2 diagnosis.

    Characteristics Associated With Hospitalization After a Diagnosis of SARS-CoV-2

    As shown in Table 3, the unadjusted rates and the adjusted risk of hospitalization from a conditional competing risk model increased with increasing BMI (eg, BMI>45 vs BMI 18.5-25: aHR, 1.40; 95% CI, 1.28-1.52). Men (aHR, 1.32; 95% CI, 1.29-1.35), Black residents (aHR, 1.28; 95% CI, 1.24-1.32), Hispanic residents (aHR, 1.20; 95% CI, 1.15-1.26), and Asian residents (aHR, 1.46; 95% CI, 1.36-1.57) diagnosed with SARS-CoV-2 had higher risks of hospitalization. Hospitalization risk was higher with increasing cognitive impairment or functional impairment. Several comorbidities, including renal disease (aHR, 1.21; 95% CI, 1.18-1.24), diabetes (aHR, 1.16; 95% CI, 1.13-1.18), and respiratory conditions (aHR, 1.14; 95% CI, 1.11-1.16), were associated with increased risk of hospitalization. The adjusted hazard of hospitalization after SARS-CoV-2 diagnosis declined from April through September (eg, compared with April, May: aHR, 0.50; 95% CI, 0.48-0.52; September: aHR, 0.21; 95% CI, 0.20-0.23).

    Results from the 3-level logistic regression models assessing resident characteristics associated with risk of hospitalization were in accordance with the main analysis (eTable 3 in the Supplement).

    Characteristics Associated With Mortality After SARS-CoV-2

    As shown in Table 4, risk of death in the 30 days after SARS-CoV-2 infection increased with age (age >90 years vs 65-70 years: aHR, 2.55; 95% CI, 2.44-2.67). There was no association of mortality with high BMI (BMI>45: aHR, 1.05; 95% CI, 0.95-1.16), while mortality was increased with BMI less than 18.5 (aHR, 1.19; 95% CI, 1.14-1.24) compared with BMI of 18.5 to 25. Men were at higher risk of death (aHR, 1.57; 95% CI, 1.53-1.61). Black (aHR, 0.99; 95% CI, 0.95-1.02) and Hispanic (aHR, 0.97; 95% CI, 0.93-1.02) race/ethnicity were not associated with mortality after a SARS-CoV-2 diagnosis, while Asian race/ethnicity (aHR, 1.19; 95% CI, 1.10-1.28) had higher risk. Risk of mortality increased with increasing cognitive dysfunction (eg, compared with no impairment, moderately impaired: aHR, 1.45; 95% CI, 1.41-1.50; severely impaired: aHR, 1.79; 95% CI, 1.71-1.86) and impaired functional status (eg, compared with functional independence, moderately dependent: aHR, 1.55; 95% CI, 1.47-1.64; fully dependent: aHR, 1.94; 95% CI, 1.83-2.05). Similar to the findings with risk of hospitalization, risk of mortality after a diagnosis of SARS-CoV-2 declined by half from April to May, then continued to decline through September. Results from 3-level logistic regression models showed similar findings to the main analysis for the association of patient characteristics with mortality after a SARS-CoV-2 diagnosis (eTable 3 in the Supplement).

    Discussion

    In this cohort study of more than 480 000 US long-stay nursing home residents, we found that the risk of infection was primarily associated with geography and the particular nursing home facility, with minimal contribution of individual characteristics of residents. To our knowledge, this is the first national study of long-stay nursing home residents. Previous studies have reported that community factors, such as percentage of foreign-born, household size, and job type, were associated with increased risk of SARS-CoV-2 infection.26 Also, nursing home characteristics, such as size, percentage of Black, Hispanic, or Asian population, percentage of residents enrolled in Medicaid, and lower staffing, were associated with increased SARS-CoV-2 infections.7-10,27,28 Risk of hospitalization and risk of death in the 30 days after a SARS-CoV-2 diagnosis were highest in April, with steep declines thereafter, consistent with prior reports.29,30

    Resident characteristics played a large role in risk of hospitalization and death. With individual characteristics, such as age, race/ethnicity, cognitive status, and functional status, there was a large divergence between the magnitude of their association with hospitalization vs the magnitude of association with mortality. With race/ethnicity and BMI, the magnitude of risk of hospitalization were considerable higher than magnitude of risk for mortality. With other characteristics, the magnitude of risk for mortality was substantially higher than the magnitude of risk for hospitalization.

    The interpretation of the divergence between risks of hospitalization vs risks of mortality is complex. In studies of community populations, the risks of hospitalization and the risks of death associated with increasing age tend to parallel each other, consistent with the concept that decisions for hospitalization are influenced by clinical judgement on risk of death.5,31-33 A similar pattern is seen with other diseases, such as influenza.34 In contrast, there was an inconsistent association of risk of hospitalization with increasing age among US nursing home residents diagnosed with SARS-CoV-2. This may represent resident or family preference to avoid hospitalization, or triaging decisions in areas where hospital beds were scarce, or lack of appropriate clinical evaluation and prognostication in facilities overwhelmed by the pandemic, or perhaps some combination of these or other explanations. Understanding this phenomenon may require a qualitative approach.

    In contrast, residents who were Black, Hispanic, or Asian had substantially higher risk of hospitalization after SARS-CoV-2 diagnosis, but the risk for mortality was very close to that of White residents. For example, after controlling for differences among nursing homes, Black residents were at 28% higher risk of hospitalization but not at increased risk of death. Studies of community residents have also found higher hospitalization rates for Black individuals,5,32,33 with little or no differences in mortality.5,33 Consistent with prior studies, we found that Asian residents were at increased risk of hospitalization and mortality.35,36 A higher prevalence of heart disease, as well as language barriers, may contribute to disparity among Asian individuals.36,37

    We should emphasize that all of these results, including ours, are from analyses that controlled for other resident characteristics and that also controlled for the individual nursing home facility using conditional models. Other investigators have reported an overall increase in SARS-CoV-2 mortality among Black nursing home residents that is explained partly by geographic location (ie, areas of the US with high initial infection rates were often also areas with higher Black populations) and partly by nursing home quality (eg, residents in facilities with a high proportion of Black residents have worse outcomes from SARS-CoV-2 and other conditions, independent of resident race).8,9

    There was a similar divergence between hospitalization risk and risk of mortality with increasing BMI. An association of BMI with death after SARS-CoV-2 infections was found in community populations, while we found no association of mortality risk with increasing BMI in nursing home residents.38-41 With Black, Hispanic, or Asian race/ethnicity and elevated BMI, there were early reports suggesting elevated mortality risks. This may have contributed to an increased clinical sensitivity to those factors. Indeed, the increase in hospitalization rates in Black, Hispanic, or Asian nursing home residents and in residents who were obese may have contributed to the lower mortality rates.

    We found an association of increasing cognitive impairment with death after SARS-CoV-2 diagnosis, similar to studies in other settings.4,42,43 However, there were inconsistent associations of cognitive function with hospitalization. Residents with cognitive impairment may be unable to make decisions for their treatment and may lack family member support owing to nursing facility and hospital restrictions.42,43 These residents also may be more likely to have advance directives precluding invasive measures. The prevalence of comorbidities was much greater in nursing home residents compared with the general population, but the associations of comorbidities with SARS-CoV-2 infection, hospitalization, and mortality were similar to community studies.44,45

    Limitations

    The study has several limitations. First, we evaluated rate of clinical diagnosis of SARS-CoV-2 infections, not the rate of actual infections. The intensity of diagnostic evaluation for SARS-CoV-2 may differ by patient characteristics, leading to underrecognition of disease. This may be the reason why residents with an estimated life expectancy of less than 6 months had a lower risk of diagnosis with SARS-CoV-2. Conversely, the ICD-10-CM code for SARS-CoV-2 may have been used in residents without the disease. Second, the study findings cannot be generalized to residents without Medicare or with Medicare Advantage. In 2020, 36% of Medicare beneficiaries were enrolled in a Medicare Advantage plan, and the share of beneficiaries in Medicare Advantage plans ranges widely across states.46 Third, we did not explore what nursing home characteristics were associated with outcomes. Fourth, there was no official ICD-10-CM code for SARS-CoV-2 prior to April 1, 2020, and that code may not have been universally used, particularly in April. Fifth, the SARS-CoV-2 pandemic may have disrupted data submission, and there may be delays in submission of Medicare claims.47 However, we analyzed Medicare data updated as of January 31, 2021, while the hospitalization and death outcomes were through October 31, 2020. Sixth, we had no data on completion and content of advance directives, which presumably influenced the extent of treatment received.

    Conclusions

    In this national cohort study of long-stay nursing home residents, risk of SARS-CoV-2 was associated with the geographic area and particularly by facility, while risk of hospitalization and death after SARS-CoV-2 was associated with individual resident characteristics. We identified novel risk factors, such as impaired cognition and physical functioning. For many resident characteristics, there are substantial differences in risk of hospitalization vs mortality. This may represent resident preferences, triaging decisions, or inadequate assessment of risk of death.

    Back to top
    Article Information

    Accepted for Publication: February 25, 2021.

    Published: March 31, 2021. doi:10.1001/jamanetworkopen.2021.6315

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Mehta HB et al. JAMA Network Open.

    Corresponding Author: James S. Goodwin, MD, Sealy Center on Aging, Department of Internal Medicine, University of Texas Medical Branch, 301 University Blvd, Galveston, TX 77555-0177 (jsgoodwi@utmb.edu).

    Author Contributions: Dr Li had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Mehta, Goodwin.

    Acquisition, analysis, or interpretation of data: All authors.

    Drafting of the manuscript: Mehta, Goodwin.

    Critical revision of the manuscript for important intellectual content: All authors.

    Statistical analysis: Mehta, Li.

    Obtained funding: Goodwin.

    Supervision: Goodwin.

    Conflict of Interest Disclosures: None reported.

    Funding/Support: This work was supported by grant No. K05-CA134923 from the National Cancer Institute, grant No. P30-AG024832-12 from the Claude D. Pepper Older Americans Independence Center, and Clinical and Translational Science Award No. UL1TR001439.

    Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    References
    1.
    Johns Hopkins University and Medicine. COVID-19 dashboard by the Center for Systems Science and Engineering at Johns Hopkins University. Accessed November 15, 2020. https://coronavirus.jhu.edu/map.html
    2.
    The New York Times. More than one-third of U.S. coronavirus deaths are linked to nursing homes. The New York Times. Updated February 26, 2021. Accessed November 15, 2020. https://www.nytimes.com/interactive/2020/us/coronavirus-nursing-homes.html
    3.
    Kaiser Family Foundation. State COVID-19 data and policy actions. Updated March 2, 2021. Accessed November 15, 2020. https://www.kff.org/coronavirus-covid-19/issue-brief/state-covid-19-data-and-policy-actions/
    4.
    Williamson  EJ, Walker  AJ, Bhaskaran  K,  et al.  Factors associated with COVID-19-related death using OpenSAFELY.   Nature. 2020;584(7821):430-436. doi:10.1038/s41586-020-2521-4 PubMedGoogle ScholarCrossref
    5.
    Ioannou  GN, Locke  E, Green  P,  et al.  Risk factors for hospitalization, mechanical ventilation, or death among 10131 US veterans with SARS-CoV-2 infection.   JAMA Netw Open. 2020;3(9):e2022310. doi:10.1001/jamanetworkopen.2020.22310 PubMedGoogle Scholar
    6.
    Holman  N, Knighton  P, Kar  P,  et al.  Risk factors for COVID-19–related mortality in people with type 1 and type 2 diabetes in England: a population-based cohort study.   Lancet Diabetes Endocrinol. 2020;8(10):823-833. doi:10.1016/S2213-8587(20)30271-0 PubMedGoogle ScholarCrossref
    7.
    Gorges  RJ, Konetzka  RT.  Staffing levels and COVID-19 cases and outbreaks in U.S. nursing homes.   J Am Geriatr Soc. 2020;68(11):2462-2466. doi:10.1111/jgs.16787 PubMedGoogle ScholarCrossref
    8.
    Li  Y, Cen  X, Cai  X, Temkin-Greener  H.  Racial and ethnic disparities in COVID-19 infections and deaths across U.S. nursing homes.   J Am Geriatr Soc. 2020;68(11):2454-2461. doi:10.1111/jgs.16847 PubMedGoogle ScholarCrossref
    9.
    Li  Y, Temkin-Greener  H, Shan  G, Cai  X.  COVID-19 infections and deaths among Connecticut nursing home residents: facility correlates.   J Am Geriatr Soc. 2020;68(9):1899-1906. doi:10.1111/jgs.16689 PubMedGoogle ScholarCrossref
    10.
    Figueroa  JF, Wadhera  RK, Papanicolas  I,  et al.  Association of nursing home ratings on health inspections, quality of care, and nurse staffing with COVID-19 cases.   JAMA. 2020;324(11):1103-1105. doi:10.1001/jama.2020.14709 PubMedGoogle ScholarCrossref
    11.
    De Smet  R, Mellaerts  B, Vandewinckele  H,  et al.  Frailty and mortality in hospitalized older adults with COVID-19: retrospective observational study.   J Am Med Dir Assoc. 2020;21(7):928-932.e1. doi:10.1016/j.jamda.2020.06.008PubMedGoogle ScholarCrossref
    12.
    Mendes  A, Serratrice  C, Herrmann  FR,  et al.  Predictors of in-hospital mortality in older patients with COVID-19: The COVIDAge Study.   J Am Med Dir Assoc. 2020;21(11):1546-1554.e3. doi:10.1016/j.jamda.2020.09.014PubMedGoogle ScholarCrossref
    13.
    Centers for Medicare & Medicaid Services. Minimum Data Set (MDS) 3.0 for nursing homes and swing bed providers. Accessed February 10, 2020. https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/NursingHomeQualityInits/NHQIMDS30
    14.
    Goodwin  JS, Li  S, Zhou  J, Graham  JE, Karmarkar  A, Ottenbacher  K.  Comparison of methods to identify long term care nursing home residence with administrative data.   BMC Health Serv Res. 2017;17(1):376. doi:10.1186/s12913-017-2318-9 PubMedGoogle ScholarCrossref
    15.
    Intrator  O, Hiris  J, Berg  K, Miller  SC, Mor  V.  The residential history file: studying nursing home residents’ long-term care histories(*).   Health Serv Res. 2011;46(1 Pt 1):120-137. doi:10.1111/j.1475-6773.2010.01194.x PubMedGoogle ScholarCrossref
    16.
    Centers for Disease Control and Prevention. International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM). Accessed March 2, 2021. https://www.cdc.gov/nchs/icd/icd10cm.htm
    17.
    Thomas  KS, Dosa  D, Wysocki  A, Mor  V.  The Minimum Data Set 3.0 Cognitive Function Scale.   Med Care. 2017;55(9):e68-e72. doi:10.1097/MLR.0000000000000334 PubMedGoogle ScholarCrossref
    18.
    Kroenke  K, Spitzer  RL, Williams  JB.  The PHQ-9: validity of a brief depression severity measure.   J Gen Intern Med. 2001;16(9):606-613. doi:10.1046/j.1525-1497.2001.016009606.x PubMedGoogle ScholarCrossref
    19.
    Saliba  D, DiFilippo  S, Edelen  MO, Kroenke  K, Buchanan  J, Streim  J.  Testing the PHQ-9 interview and observational versions (PHQ-9 OV) for MDS 3.0.   J Am Med Dir Assoc. 2012;13(7):618-625. doi:10.1016/j.jamda.2012.06.003 PubMedGoogle ScholarCrossref
    20.
    Perlman  CM, Hirdes  JP.  The aggressive behavior scale: a new scale to measure aggression based on the minimum data set.   J Am Geriatr Soc. 2008;56(12):2298-2303. doi:10.1111/j.1532-5415.2008.02048.x PubMedGoogle ScholarCrossref
    21.
    Fine  JP, Gray  RJ.  A proportional hazards model for subdistribution of a competing risk.   J Am Stat Assoc. 1999;94:496-509. doi:10.1080/01621459.1999.10474144 Google ScholarCrossref
    22.
    Zhou  B, Latouche  A, Rocha  V, Fine  J.  Competing risks regression for stratified data.   Biometrics. 2011;67(2):661-670. doi:10.1111/j.1541-0420.2010.01493.x PubMedGoogle ScholarCrossref
    23.
    Muff  S, Held  L, Keller  LF.  Marginal or conditional regression models for correlated non-normal data?   Methods Ecol Evol. 2016;7(12):1514-1524. doi:10.1111/2041-210X.12623 Google ScholarCrossref
    24.
    Breslow  NE, Day  NE, Halvorsen  KT, Prentice  RL, Sabai  C.  Estimation of multiple relative risk functions in matched case-control studies.   Am J Epidemiol. 1978;108(4):299-307. doi:10.1093/oxfordjournals.aje.a112623 PubMedGoogle ScholarCrossref
    25.
    Singh  S, Lin  YL, Kuo  YF, Nattinger  AB, Goodwin  JS.  Variation in the risk of readmission among hospitals: the relative contribution of patient, hospital and inpatient provider characteristics.   J Gen Intern Med. 2014;29(4):572-578. doi:10.1007/s11606-013-2723-7 PubMedGoogle ScholarCrossref
    26.
    Figueroa  JF, Wadhera  RK, Lee  D, Yeh  RW, Sommers  BD.  Community-level factors associated with racial And ethnic disparities in COVID-19 rates In Massachusetts.   Health Aff (Millwood). 2020;39(11):1984-1992. doi:10.1377/hlthaff.2020.01040 PubMedGoogle ScholarCrossref
    27.
    Abrams  HR, Loomer  L, Gandhi  A, Grabowski  DC.  Characteristics of U.S. nursing homes with COVID-19 cases.   J Am Geriatr Soc. 2020;68(8):1653-1656. doi:10.1111/jgs.16661 PubMedGoogle ScholarCrossref
    28.
    Chen  MK, Chevalier  JA, Long  EF.  Nursing Home Staff Networks and COVID-19. National Bureau of Economic Research; 2020;898-2937.
    29.
    Ritchie  H, Ortiz-Ospina  E, Beltekian  D,  et al. Coronavirus disease 2019 (COVID-19) deaths. Our World in Data. Updated March 2, 2021. Accessed February 10, 2020. https://ourworldindata.org/covid-deaths?country=USA
    30.
    Centers for Medicare & Medicaid Services. COVID-19 nursing home dataset. Accessed February 10, 2021. https://data.cms.gov/Special-Programs-Initiatives-COVID-19-Nursing-Home/COVID-19-Nursing-Home-Dataset/s2uc-8wxp
    31.
    Shi  SM, Bakaev  I, Chen  H, Travison  TG, Berry  SD.  Risk factors, presentation, and course of Coronavirus Disease 2019 in a large, academic long-term care facility.   J Am Med Dir Assoc. 2020;21(10):1378-1383.e1. doi:10.1016/j.jamda.2020.08.027PubMedGoogle ScholarCrossref
    32.
    Petrilli  CM, Jones  SA, Yang  J,  et al.  Factors associated with hospital admission and critical illness among 5279 people with coronavirus disease 2019 in New York City: prospective cohort study.   BMJ. 2020;369:m1966. doi:10.1136/bmj.m1966 PubMedGoogle ScholarCrossref
    33.
    Price-Haywood  EG, Burton  J, Fort  D, Seoane  L.  Hospitalization and mortality among Black patients and White patients with COVID-19.   N Engl J Med. 2020;382(26):2534-2543. doi:10.1056/NEJMsa2011686 PubMedGoogle ScholarCrossref
    34.
    Czaja  CA, Miller  L, Alden  N,  et al.  Age-related differences in hospitalization rates, clinical presentation, and outcomes among older adults hospitalized with influenza—U.S: Influenza Hospitalization Surveillance Network (FluSurv-NET).   Open Forum Infect Dis. 2019;6(7):ofz225. doi:10.1093/ofid/ofz225 PubMedGoogle Scholar
    35.
    Sze  S, Pan  D, Nevill  CR,  et al.  Ethnicity and clinical outcomes in COVID-19: a systematic review and meta-analysis.   EClinicalMedicine. 2020;29:100630. doi:10.1016/j.eclinm.2020.100630 PubMedGoogle Scholar
    36.
    Yan  BW, Ng  F, Chu  J, Tsoh  J, Nguyen  T. Asian Americans facing high COVID-19 case fatality. Health Affairs Blog. July 13, 2020. Accessed February 10, 2021. https://www.healthaffairs.org/do/10.1377/hblog20200708.894552/full/
    37.
    Li  Y, Yin  J, Cai  X, Temkin-Greener  J, Mukamel  DB.  Association of race and sites of care with pressure ulcers in high-risk nursing home residents.   JAMA. 2011;306(2):179-186. doi:10.1001/jama.2011.942 PubMedGoogle ScholarCrossref
    38.
    Hendren  NS, de Lemos  JA, Ayers  C,  et al.  Association of body mass index and age with morbidity and mortality in patients hospitalized with COVID-19: results from the American Heart Association COVID-19 Cardiovascular Disease Registry.   Circulation. 2021;143(2):135-144. doi:10.1161/CIRCULATIONAHA.120.051936 PubMedGoogle ScholarCrossref
    39.
    Klang  E, Kassim  G, Soffer  S, Freeman  R, Levin  MA, Reich  DL.  Severe obesity as an independent risk factor for COVID-19 mortality in hospitalized patients younger than 50.   Obesity (Silver Spring). 2020;28(9):1595-1599. doi:10.1002/oby.22913 PubMedGoogle ScholarCrossref
    40.
    Recalde  M, Pistillo  A, Fernandez-Bertolin  S,  et al  Body mass index and risk of COVID-19 diagnosis, hospitalisation, and death: a population-based multi-state cohort analysis including 2,524,926 people in Catalonia, Spain.   medRxiv. Preprint posted online November 28, 2020. doi:10.1101/2020.11.25.20237776Google Scholar
    41.
    Tartof  SY, Qian  L, Hong  V,  et al.  Obesity and mortality among patients diagnosed with COVID-19: results from an integrated health care organization.   Ann Intern Med. 2020;173(10):773-781. doi:10.7326/M20-3742 PubMedGoogle ScholarCrossref
    42.
    Livingston  G, Rostamipour  H, Gallagher  P,  et al.  Prevalence, management, and outcomes of SARS-CoV-2 infections in older people and those with dementia in mental health wards in London, UK: a retrospective observational study.   Lancet Psychiatry. 2020;7(12):1054-1063. doi:10.1016/S2215-0366(20)30434-X PubMedGoogle ScholarCrossref
    43.
    Reyes-Bueno  JA, Mena-Vázquez  N, Ojea-Ortega  T,  et al.  Case fatality of COVID-19 in patients with neurodegenerative dementia.   Neurologia. 2020;35(9):639-645. doi:10.1016/j.nrl.2020.07.005 PubMedGoogle ScholarCrossref
    44.
    Atkins  JL, Masoli  JAH, Delgado  J,  et al.  Preexisting comorbidities predicting COVID-19 and mortality in the UK Biobank Community cohort.   J Gerontol A Biol Sci Med Sci. 2020;75(11):2224-2230. doi:10.1093/gerona/glaa183 PubMedGoogle ScholarCrossref
    45.
    Harrison  SL, Fazio-Eynullayeva  E, Lane  DA, Underhill  P, Lip  GYH.  Comorbidities associated with mortality in 31,461 adults with COVID-19 in the United States: a federated electronic medical record analysis.   PLoS Med. 2020;17(9):e1003321. doi:10.1371/journal.pmed.1003321 PubMedGoogle Scholar
    46.
    Kaiser Family Foundation. A dozen facts about Medicare Advantage in 2020. Updated January 13, 2021. Accessed December 7, 2020. https://www.kff.org/medicare/issue-brief/a-dozen-facts-about-medicare-advantage-in-2020/
    47.
    Shioda  K, Weinberger  DM, Mori  M.  Navigating through health care data disrupted by the COVID-19 pandemic.   JAMA Intern Med. 2020;180(12):1569-1570. doi:10.1001/jamainternmed.2020.5542 PubMedGoogle ScholarCrossref
    ×