Factors Associated with Severe Outcomes Among Immunocompromised Adults Hospitalized for COVID-19 — COVID-NET, 10 States, March 2020–February 2022

Immunocompromised persons are at increased risk for severe COVID-19–related outcomes, including intensive care unit (ICU) admission and death (1). Data on adults aged ≥18 years hospitalized with laboratory-confirmed COVID-19 from 10 U.S. states in the COVID-19–Associated Hospitalization Surveillance Network (COVID-NET) were analyzed to assess associations between immunocompromise and ICU admission and in-hospital death during March 1, 2020–February 28, 2022. Associations of COVID-19 vaccination status with ICU admission and in-hospital death were also examined during March 1, 2021–February 28, 2022. During March 1, 2020–February 28, 2022, among a sample of 22,345 adults hospitalized for COVID-19, 12.2% were immunocompromised. Among unvaccinated patients, those with immunocompromise had higher odds of ICU admission (adjusted odds ratio [aOR] = 1.26; 95% CI = 1.08–1.49) and in-hospital death (aOR = 1.34; 95% CI = 1.05–1.70) than did nonimmunocompromised patients. Among vaccinated patients,* those with immunocompromise had higher odds of ICU admission (aOR = 1.40; 95% CI = 1.01–1.92) and in-hospital death (aOR = 1.87; 95% CI = 1.28–2.75) than did nonimmunocompromised patients. During March 1, 2021–February 28, 2022, among nonimmunocompromised patients, patients who were vaccinated had lower odds of death (aOR = 0.58; 95% CI = 0.39–0.86) than did unvaccinated patients; among immunocompromised patients, odds of death between vaccinated and unvaccinated patients did not differ. Immunocompromised persons need additional protection from COVID-19 and using multiple known COVID-19 prevention strategies,^† including nonpharmaceutical interventions, up-to-date vaccination of immunocompromised persons and their close contacts,^§ early testing, and COVID-19 prophylactic (Evusheld) and early antiviral treatment,^¶ can help prevent hospitalization and subsequent severe COVID-19 outcomes among immunocompromised persons.

COVID-NET is a CDC-funded collaboration for population-based surveillance of laboratory-confirmed COVID-19–associated hospitalization in 99 U.S. counties in 14 states. A COVID-NET case is defined as a positive real-time reverse transcription–polymerase chain reaction or rapid antigen test result for SARS-CoV-2 (the virus that causes COVID-19) within 14 days before or during hospitalization in a person who lived in the surveillance catchment area. Medical chart abstraction and representative sampling methods have been described previously (2). Data collected on sampled adults hospitalized during March 1, 2020–February 28, 2022, across 10 participating states** were examined. Patients whose hospitalization was not likely related to COVID-19^†† and those without a completed chart review were excluded. Immunocompromised patients were defined as those having one or more predefined immunocompromising conditions.^§§ COVID-19 vaccination definitions for immunocompromised persons changed during the study period^¶¶; in this analysis, a vaccinated patient was defined as one who had received both doses of a 2-dose COVID-19 vaccination series or 1 dose of a single-dose COVID-19 vaccine with or without additional or booster doses ≥14 days before their positive SARS-CoV-2 test result, per state immunization information system records. Vaccinated patients with additional or booster doses were not analyzed separately. Patients were considered unvaccinated if no COVID-19 vaccination was recorded before the positive test result; patients who were documented to have received only the first dose of a 2-dose series or their last vaccination series dose <14 days before receiving a positive SARS-CoV-2 test result were excluded.

Demographic and clinical characteristics of hospitalized patients were assessed; Pearson’s chi-square tests were used to compare differences between immunocompromised and nonimmunocompromised patients. Bivariate and multivariable logistic regression analyses were used to assess associations between immunocompromise and both ICU admission and in-hospital death among vaccinated and unvaccinated patients in separate models. Associations between each individual immunocompromising condition and death were assessed using multivariable analyses, adjusting only for age and sex to improve model convergence. Bivariate and multivariable analyses were used to assess the association between vaccination status and both ICU admission and in-hospital death among immunocompromised and nonimmunocompromised patients in separate models, using data beginning March 1, 2021, when immunocompromised patients first reported receiving vaccine doses, through February 28, 2022. Multivariable analyses were adjusted for age, sex, site (entered as a fixed effect), SARS-CoV-2 variant–predominant period,*** and other factors with documented or potential association and a p-value <0.10 in bivariate analyses. Statistical analyses used SAS (version 9.4; SAS Institute) survey procedures to account for sampling weights, with statistical significance set at alpha = 0.05. This activity was reviewed by CDC and conducted consistent with applicable federal law and CDC policy.^†††

During March 1, 2020–February 28, 2022, a representative sample of 24,625 (11.0%, unweighted) of 223,069 COVID-NET cases had complete chart review, including 22,345^§§§ (90.7%, unweighted) that met inclusion criteria. Among the 22,345 patients included, 12.2% were immunocompromised, including 11.1%, 10.9%, and 17.3% of patients hospitalized during the pre-Delta, Delta, and Omicron variant–predominant periods, respectively. Overall, immunocompromised patients were more likely to be older and to be non-Hispanic White (Table 1). Compared with nonimmunocompromised patients, those with immunocompromise had a statistically significantly higher prevalence of all underlying medical conditions except diabetes and neurologic disease.

Among unvaccinated patients, those who were immunocompromised had higher odds of ICU admission (aOR = 1.26) and death (aOR = 1.34) than did nonimmunocompromised patients^¶¶¶ (Table 2). Similarly, among vaccinated patients, those who were immunocompromised also had higher odds of ICU admission (aOR = 1.40) and in-hospital death (aOR = 1.87) compared with nonimmunocompromised patients.**** Among patients with a specific immunocompromising condition compared with patients without that condition (irrespective of immunocompromise status), the odds of in-hospital death were higher for those with AIDS or low CD4+ count (aOR = 2.03), immunosuppressive therapy use (aOR = 1.65), multiple myeloma (aOR = 5.28), or solid organ transplant (aOR = 2.12) and lower for patients with immunoglobulin deficiency (aOR = 0.16) (Supplementary Table 1, https://stacks.cdc.gov/view/cdc/118606).

Among immunocompromised patients, those who were vaccinated did not have statistically significantly different odds of ICU admission or in-hospital death^†††† compared with unvaccinated patients (Table 3). Among nonimmunocompromised patients, those who were vaccinated had lower odds of death (aOR = 0.58) than did unvaccinated patients.^§§§§

During the pre-Delta and Delta variant–predominant periods, immunocompromised patients generally had higher odds of death, irrespective of vaccination status compared with nonimmunocompromised patients, and nonimmunocompromised patients who were vaccinated had lower odds of death compared with unvaccinated patients (Supplementary Table 2, https://stacks.cdc.gov/view/cdc/118607). However, in the Omicron variant–predominant period, odds of death, irrespective of immunocompromise or vaccination status, were not statistically significantly different.

Acknowledgments

Onika Anglin, Kadam Patel, CDC and General Dynamics Information Technology, Atlanta, Georgia; Jennifer Milucky, Michael Whitaker, CDC; Susan Brooks, Ashley Coates, Monica Napoles, Jeremy Roland, Maria Rosales, Lena Tayo, Tiffany Tsukuda, California Emerging Infections Program; Amanda Kamali, John Openshaw, California Department of Public Health; Sharon Emmerling, Breanna Kawasaki, Madelyn Lensing, Sarah McLafferty, Jordan Surgnier, Millen Tsegaye, Colorado Department of Public Health and Environment; Ann Basting, Tessa Carter, Maria Correa, Daewi Kim, Carol Lyons, Amber Maslar, Julie Plano, Hazhia Sorosinidi, Connecticut Emerging Infections Program, Yale School of Public Health; Taylor Eisenstein, Emily Fawcett, Annabel Patterson, Foundation for Atlanta Veterans Education and Research; Atlanta Veterans Affairs Medical Center; Emory University School of Medicine; Marina Bruck, Rayna Ceaser, Sabrina Hendrick, Johanna Hernandez, Asmith Joseph, Grayson Kallas, Stephanie Lehman, Jana Manning, Allison Roebling, Suzanne Segler, Chandler Surell, Katelyn Ward, Hope Wilson, Emory University School of Medicine; Georgia Emerging Infections Program, Georgia Department of Public Health; Veterans Affairs Medical Center; Chloe Brown, Jim Collins, Shannon Johnson, Justin Henderson, Sue Kim, Alexander Kohrman, Lauren Leegwater, Sanchitha Meda, Alyanna Melicor, Val Tellez Nunez, Sierra Peguies-Khan, Michigan Department of Health and Human Services; Kayla Bilski, Natalie Bullis, Kathy Como-Sabetti, Emma Contestabile, Kristen Ehresmann, Claire Henrichsen, Grace Hernandez, Jackie Johnson, Tenzin Kunkyi, Erica Mumm, Kieu My Phi, Lisa Nguyen, Elizabeth Snyderman, Minnesota Department of Health; Cory Cline, Adrienne Domen, Melissa Judson, Sunshine Martinez, Florent Nkouaga, Kelly Plymesser, Jasmyn Sanchez, Chad Smelser, New Mexico Department of Health; Kathy M. Angeles, Molly Bleecker, Sarah Shrum Davis, Nancy Eisenberg, Emily B. Hancock, Sarah A. Khanlian, Sarah Lathrop, Francesca Pacheco, Mayvilynne Poblete, Dominic Rudin, Yadira Salazar-Sanchez, New Mexico Emerging Infections Program; Jennifer Akpo, Celina Chavez, Murtada Khalifa, Alesia Reed, Yassir Talha, CDC Foundation and New Mexico Department of Health; Kerianne Engesser, Adam Rowe, Nancy Spina, New York State Department of Health; Sophrena Bushey, Virgina Cafferky, Maria Gaitan, Christine Long, Thomas Peer, University of Rochester School of Medicine and Dentistry; Sam Hawkins, Oregon Health Authority; Kathy Billings, Katie Dyer, Anise Elie, Gail Hughett, Karen Leib, Tiffanie Markus, Terri McMinn, Danielle Ndi, Manideepthi Pemmaraju, Vanderbilt University Medical Center.

Corresponding author: Jason Robert C. Singson, [email protected].

¹California Emerging Infections Program, Oakland California; ²Council of State and Territorial Epidemiologists, Atlanta, Georgia; ³CDC COVID-19 Emergency Response Team; ⁴Colorado Department of Public Health and Environment; ⁵Connecticut Emerging Infections Program, Yale School of Public Health, New Haven, Connecticut; ⁶Georgia Emerging Infections Program, Georgia Department of Health; ⁷Emory University School of Medicine, Atlanta, Georgia; ⁸Atlanta Veterans Affairs Medical Center, Atlanta, Georgia; ⁹Michigan Department of Health and Human Services; ¹⁰Minnesota Department of Health; ¹¹New Mexico Department of Health; ¹²New York State Department of Health; ¹³University of Rochester School of Medicine and Dentistry, Rochester, New York; ¹⁴Public Health Division, Oregon Health Authority; ¹⁵Vanderbilt University Medical Center, Nashville, Tennessee; ¹⁶Career Epidemiology Field Officer Program, CDC.

References

1. Kim L, Garg S, O’Halloran A, et al. 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). Clin Infect Dis 2021;72:e206–14. https://doi.org/10.1093/cid/ciaa1012 PMID:32674114
2. Garg S, Patel K, Pham H, et al. Clinical trends among U.S. adults hospitalized with COVID-19, March to December 2020: a cross-sectional study. Ann Intern Med 2021;174:1409–19. https://doi.org/10.7326/M21-1991 PMID:34370517
3. Tenforde MW, Patel MM, Gaglani M, et al.; IVY Network. Effectiveness of a third dose of Pfizer-BioNTech and Moderna vaccines in preventing COVID-19 hospitalization among immunocompetent and immunocompromised adults—United States, August–December 2021. MMWR Morb Mortal Wkly Rep 2022;71:118–24. https://doi.org/10.15585/mmwr.mm7104a2 PMID:35085218
4. Tenforde MW, Self WH, Gaglani M, et al.; IVY Network. Effectiveness of mRNA vaccination in preventing COVID-19–associated invasive mechanical ventilation and death—United States, March 2021–January 2022. MMWR Morb Mortal Wkly Rep 2022;71:459–65. https://doi.org/10.15585/mmwr.mm7112e1 PMID:35324878
5. Harpaz R, Dahl RM, Dooling KL. Prevalence of immunosuppression among US adults, 2013. JAMA 2016;316:2547–8. https://doi.org/10.1001/jama.2016.16477 PMID:27792809
6. Chow N, Fleming-Dutra K, Gierke R, et al.; CDC COVID-19 Response Team. Preliminary estimates of the prevalence of selected underlying health conditions among patients with coronavirus disease 2019—United States, February 12–March 28, 2020. MMWR Morb Mortal Wkly Rep 2020;69:382–6. https://doi.org/10.15585/mmwr.mm6913e2 PMID:32240123
7. Andrejko KL, Pry JM, Myers JF, et al.; California COVID-19 Case-Control Study Team. Effectiveness of face mask or respirator use in indoor public settings for prevention of SARS-CoV-2 infection—California, February–December 2021. MMWR Morb Mortal Wkly Rep 2022;71:212–6. https://doi.org/10.15585/mmwr.mm7106e1 PMID:35143470
8. Sami S, Horter L, Valencia D, et al. Investigation of SARS-CoV-2 transmission associated with a large indoor convention—New York City, November–December 2021. MMWR Morb Mortal Wkly Rep 2022;71:243–8. https://doi.org/10.15585/mmwr.mm7107a4 PMID:35176005
9. Wiltz JL, Feehan AK, Molinari NM, et al. Racial and ethnic disparities in receipt of medications for treatment of COVID-19—United States, March 2020–August 2021. MMWR Morb Mortal Wkly Rep 2022;71:96–102. https://doi.org/10.15585/mmwr.mm7103e1 PMID:35051133
10. Fu J, Reid SA, French B, et al.; COVID-19 and Cancer Consortium (CCC19). Racial disparities in COVID-19 outcomes among black and white patients with cancer. JAMA Netw Open 2022;5:e224304. https://doi.org/10.1001/jamanetworkopen.2022.4304 PMID:35344045

Abbreviations: A/PI = Asian or Pacific Islander; AI/AN = American Indian or Alaska Native; COVID-NET = COVID-19–Associated Hospitalization Surveillance Network; NA = not applicable.
* Vaccinated patients were defined as those with a positive SARS-CoV-2 test result from a specimen collected ≥14 days after either the second dose of a 2-dose vaccination series or after 1 dose of a single dose vaccine. When not otherwise specified, vaccinated patients include those who might have received additional or booster doses. Vaccinated patients without additional or booster doses include both those eligible and not yet eligible for an additional or booster dose. Vaccinated patients with additional booster doses received additional or booster doses on or after August 13, 2021, with a positive SARS-CoV-2 test result from a specimen collected ≥14 days after receipt of the additional or booster dose.
^† Selected counties in California, Colorado, Connecticut, Georgia, Michigan, Minnesota, New Mexico, New York, Oregon, and Tennessee.
^§ Representative sample of all cases reported to COVID-NET, stratified by patient age and COVID-NET site. Percentages were weighted to account for the probability of selection for sampled cases.
^¶ White, Black, AI/AN, and A/PI persons were non-Hispanic; Hispanic persons could be of any race.
** Relative SE >30. Estimates might be unstable; results should be interpreted with caution.
^†† Includes patients who were classified as multiracial, non-Hispanic. Non-Hispanic ethnicity was assumed for patients with unknown ethnicity.
^§§ Pre-Delta variant–predominant period = March 1, 2020–June 26, 2021; Delta variant–predominant period = June 27–December 18, 2021; Omicron variant–predominant period = December 19, 2021–February 28, 2022. https://www.cdc.gov/mmwr/volumes/71/wr/mm7116e1.htm
^††† Defined as one or more of the following: chronic lung disease (including asthma), chronic metabolic disease, diabetes mellitus, blood disorder/hemoglobinopathy, cardiovascular disease, neurologic disease, renal disease, gastrointestinal/liver disease, rheumatologic/autoimmune condition, obesity, feeding tube dependence, or wheelchair dependence.

Abbreviations: aOR = adjusted odds ratio; COVID-NET = COVID-19–Associated Hospitalization Surveillance Network; ICU = intensive care unit; Ref = referent group.
* Vaccinated patients were defined as those with a positive SARS-CoV-2 test result from a specimen collected ≥14 days after the second dose of a 2-dose vaccination series or after 1 dose of a single dose vaccine. When not otherwise specified, vaccinated patients include those who might have received additional or booster doses.
^† Selected counties in California, Colorado, Connecticut, Georgia, Michigan, Minnesota, New Mexico, New York, Oregon, and Tennessee.
^§ Representative sample of all cases reported to COVID-NET, stratified by age and COVID-NET site. Percentages were weighted to account for the probability of selection for sampled cases.
^¶ ICU admission model among unvaccinated patients was adjusted for age, sex, site, variant predominant period, race/ethnicity, hypertension, diabetes, chronic lung disease, cardiovascular disease, renal disease, and obesity. ICU status was not known for 66 nonimmunocompromised patients. Death model among unvaccinated patients was adjusted for age, sex, site, variant predominant period, race/ethnicity, long-term care facility, hypertension, diabetes, chronic lung disease, cardiovascular disease, renal disease, blood disorders, neurologic disease, and rheumatologic/autoimmune condition. Death outcome was unknown for one immunocompromised patient and 38 nonimmunocompromised patients.
** ICU admission model among vaccinated patients was adjusted for age, sex, site, variant predominant period, race/ethnicity (American Indian or Alaska Native and Asian or Pacific Islander were reclassified as other/unknown because of small numbers as well as patients who identified as multiracial or unknown race), chronic metabolic disease, liver disease, and rheumatologic/autoimmune condition. ICU status was not known for seven nonimmunocompromised patients. Death model among vaccinated patients was adjusted for age, sex, site, variant predominant period, race/ethnicity, cardiovascular disease, renal disease, and rheumatologic/autoimmune condition. Death outcome was unknown for one immunocompromised patient and seven nonimmunocompromised patients.
^†† p-value <0.01.
^§§ p-value <0.05.

Abbreviations: aOR = adjusted odds ratio; COVID-NET = COVID-19–Associated Hospitalization Surveillance Network; ICU = intensive care unit; Ref = referent group.
* Vaccinated patients were defined as those with a positive SARS-CoV-2 test result from a specimen collected ≥14 days after either the second dose of a 2-dose vaccination series or after 1 dose of a single dose vaccine. When not otherwise specified, vaccinated patients include those who might have received additional or booster doses.
^† Selected counties in California, Colorado, Connecticut, Georgia, Michigan, Minnesota, New Mexico, New York, Oregon, and Tennessee.
^§ Representative sample of all cases reported to COVID-NET, stratified by age and COVID-NET site. Percentages were weighted to account for the probability of selection for sampled cases.
^¶ ICU admission model among immunocompromise patients was adjusted for age, sex, site, variant predominant period, rheumatologic/autoimmune condition, and obesity. Death model among immunocompromised patients was adjusted for age, sex, site, variant predominant period hypertension, renal disease, and rheumatologic/autoimmune condition. Death outcome among immunocompromised patients was not known for one unvaccinated patient and one vaccinated patient without additional or booster doses.
** ICU admission model among nonimmunocompromised patients was adjusted for age, sex, site, variant predominant period, diabetes, and obesity. Death model among nonimmunocompromised patients was adjusted for age, sex, site, variant predominant period, long-term care facility residence, hypertension, diabetes mellitus, chronic metabolic disease, cardiovascular disease, renal disease, and neurologic disease. Death outcome among nonimmunocompromised patients was not known for 11 unvaccinated patients, three patients vaccinated without additional or booster doses, and three patients vaccinated with additional or booster doses.
^†† p-value <0.05.