Estimates of Bivalent mRNA Vaccine Durability in Preventing COVID-19–Associated Hospitalization and Critical Illness Among Adults with and Without Immunocompromising Conditions — VISION Network, September 2022–April 2023

On September 1, 2022, CDC’s Advisory Committee on Immunization Practices (ACIP) recommended a single bivalent mRNA COVID-19 booster dose for persons aged ≥12 years who had completed at least a monovalent primary series. Early vaccine effectiveness (VE) estimates among adults aged ≥18 years showed receipt of a bivalent booster dose provided additional protection against COVID-19–associated emergency department and urgent care visits and hospitalizations compared with that in persons who had received only monovalent vaccine doses (1); however, insufficient time had elapsed since bivalent vaccine authorization to assess the durability of this protection. The VISION Network* assessed VE against COVID-19–associated hospitalizations by time since bivalent vaccine receipt during September 13, 2022–April 21, 2023, among adults aged ≥18 years with and without immunocompromising conditions. During the first 7–59 days after vaccination, compared with no vaccination, VE for receipt of a bivalent vaccine dose among adults aged ≥18 years was 62% (95% CI = 57%–67%) among adults without immunocompromising conditions and 28% (95% CI = 10%–42%) among adults with immunocompromising conditions. Among adults without immunocompromising conditions, VE declined to 24% (95% CI = 12%–33%) among those aged ≥18 years by 120–179 days after vaccination. VE was generally lower for adults with immunocompromising conditions. A bivalent booster dose provided the highest protection, and protection was sustained through at least 179 days against critical outcomes, including intensive care unit (ICU) admission or in-hospital death. These data support updated recommendations allowing additional optional bivalent COVID-19 vaccine doses for certain high-risk populations. All eligible persons should stay up to date with recommended COVID-19 vaccines.

The VISION Network evaluated VE of bivalent vaccines against COVID-19–associated hospitalization by length of time since receipt of the most recent dose during September 13, 2022–April 21, 2023, across five sites in seven states. VE methods used by the VISION Network have been previously described (2). For this analysis, adults aged ≥18 years with and without immunocompromising conditions who were hospitalized with COVID-19–like illness^† were included if the patient received molecular testing (e.g., real-time reverse transcription–polymerase chain reaction) for SARS-CoV-2 during the 14 days preceding or up to 72 hours after hospital admission. Patients were categorized as immunocompromised or not based on International Classification of Diseases, Tenth Revision (ICD-10) discharge codes.^§ Patients were classified on the index date^¶ as unvaccinated (no COVID-19 vaccine doses received), vaccinated with monovalent doses only, or vaccinated with one mRNA bivalent booster dose (regardless of number of previous monovalent doses received). Patients who received only monovalent doses were included if they received any combination of 1–4 doses (or 1–5 doses if immunocompromised) monovalent mRNA (Moderna or Pfizer-BioNTech), Janssen (Johnson & Johnson), or Novavax vaccine doses; recipients of a single monovalent mRNA dose or a single Novavax dose were excluded. In addition, patients were excluded if any vaccine dose was received <7 days before the index date, if a bivalent dose was received ≥180 days before the index date, or if >1 bivalent dose was received.** Patients aged <50 years without documented immunocompromising conditions were excluded if they had received >3 monovalent doses. Patients were considered to have critical illness if they were admitted to an ICU, died, or both.^††

Absolute VE was estimated using a test-negative case-control design comparing the odds of vaccination (either bivalent booster or monovalent doses only versus being unvaccinated) among case- and control patients. Relative VE was calculated by comparing those who received a bivalent booster with those who received monovalent doses only. A combined model was generated and included patients who had only received monovalent vaccination ≥7 days before their index date, or a bivalent mRNA booster dose at 7–59, 60–119, or 120–179 days before their index date, compared with an unvaccinated reference group. Odds ratios and 95% CIs were estimated using multivariable logistic regression controlling for age, race and ethnicity, sex, calendar day (days since January 1, 2021), and geographic region. Age and calendar day were modeled as natural cubic splines. VE was modeled separately for persons with and without immunocompromising conditions, by age group (18–64 and ≥65 years), and for each outcome (hospitalization and critical illness).^§§ Analyses were conducted using R (version 4.2.2; The R Foundation). This study was conducted consistent with applicable federal law and CDC policy and was reviewed and approved by Institutional Review Boards at participating sites or under a reliance agreement with the Institutional Review Board of Westat, Inc.^¶¶

Among 66,141 hospitalized patients without immunocompromising conditions who met inclusion criteria, 6,907 (10.4%) were case-patients and 59,234 (89.6%) were control patients (Table 1). Median age of case- and control patients was 76 years and 71 years, respectively. Among case- and control patients, 25.9% and 23.2% were unvaccinated, respectively; a bivalent vaccine dose had been received by 16.3% of case-patients and 20.6% of control patients. VE against COVID-19–associated hospitalization was similar across age groups, but waned over time, from 62% during the first 7–59 days after the bivalent dose to 24% by 120–179 days among adults aged ≥18 years (Table 2). Among those who received monovalent doses only, VE was 21% a median 376 days after the last dose. VE against critical illness was 69% during the 7–59 days after receipt of a bivalent dose and was more sustained (50% at 120–179 days after bivalent vaccination) than VE against hospitalization.

Among 18,934 hospitalized patients with immunocompromising conditions who met inclusion criteria, 1,834 (9.7%) were case-patients and 17,100 (90.3%) were control patients (Table 3); these persons represented 22.3% of the overall hospitalized population who met inclusion criteria. Median age of case- and control patients was 73 years and 70 years, respectively. Within this group, 17.1% of case-patients and 16.3% of control patients were unvaccinated; 21.0% of case-patients and 25.1% of control patients had received a bivalent dose. Among patients aged ≥18 years with immunocompromising conditions, VE against COVID-19–associated hospitalization was 28% during the first 7–59 days after receipt of the bivalent dose and declined to 13% by 120–179 days. VE for those who received monovalent doses only was 3% (median 355 days after the last dose). Estimates of relative and absolute VE were similar (Supplementary Table, https://stacks.cdc.gov/view/cdc/128421).

Corresponding author: Ruth Link-Gelles, media@cdc.gov.

¹Coronavirus and Other Respiratory Viruses Division, National Center for Immunization and Respiratory Diseases, CDC; ²Westat, Rockville, Maryland; ³Section of Pediatric Infectious Diseases, Department of Pediatrics, Baylor Scott & White Health, Temple, Texas; ⁴Department of Medical Education, Texas A&M University College of Medicine, Temple, Texas; ⁵Influenza Division, National Center for Immunization and Respiratory Diseases, CDC; ⁶Children’s Minnesota, Minneapolis, Minnesota; ⁷Department of Biomedical Informatics, Columbia University Irving Medical Center, New York, New York; ⁸NewYork-Presbyterian Hospital, New York, New York; ⁹HealthPartners Institute, Minneapolis, Minnesota; ¹⁰Division of Infectious Diseases and Clinical Epidemiology, Intermountain Healthcare, Salt Lake City, Utah; ¹¹Kaiser Permanente Center for Health Research, Portland, Oregon; ¹²Kaiser Permanente Vaccine Study Center, Kaiser Permanente Northern California Division of Research, Oakland, California; ¹³Center for Biomedical Informatics, Regenstrief Institute, Indianapolis, Indiana; ¹⁴School of Medicine, Indiana University, Indianapolis, Indiana; ¹⁵School of Medicine, University of Colorado Anschutz Medical Campus, Aurora, Colorado; ¹⁶Vanderbilt University Medical Center, Nashville, Tennessee; ¹⁷Fairbanks School of Public Health, Indiana University, Indianapolis, Indiana.

References

1. Tenforde MW, Weber ZA, Natarajan K, et al. Early estimates of bivalent mRNA vaccine effectiveness in preventing COVID-19–associated emergency department or urgent care encounters and hospitalizations among immunocompetent adults—VISION Network, nine states, September–November 2022. MMWR Morb Mortal Wkly Rep 2023;71:1637–46. https://doi.org/10.15585/mmwr.mm7153a1 PMID:36921274
2. Thompson MG, Natarajan K, Irving SA, et al. Effectiveness of a third dose of mRNA vaccines against COVID-19–associated emergency department and urgent care encounters and hospitalizations among adults during periods of Delta and Omicron variant predominance—VISION Network, 10 states, August 2021–January 2022. MMWR Morb Mortal Wkly Rep 2022;71:139–45. https://doi.org/10.15585/mmwr.mm7104e3 PMID:35085224
3. DeCuir J, Surie D, Zhu Y, et al.; IVY Network. Effectiveness of monovalent mRNA COVID-19 vaccination in preventing COVID-19–associated invasive mechanical ventilation and death among immunocompetent adults during the Omicron variant period—IVY Network, 19 U.S. states, February 1, 2022–January 31, 2023. MMWR Morb Mortal Wkly Rep 2023;72:463–8. https://doi.org/10.15585/mmwr.mm7217a3 PMID:37104244
4. Britton A, Embi PJ, Levy ME, et al. Effectiveness of COVID-19 mRNA vaccines against COVID-19–associated hospitalizations among immunocompromised adults during SARS-CoV-2 Omicron predominance—VISION Network, 10 states, December 2021–August 2022. MMWR Morb Mortal Wkly Rep 2022;71:1335–42. https://doi.org/10.15585/mmwr.mm7142a4 PMID:36264840

Abbreviations: BV = bivalent; ICU = intensive care unit; KPCHR = Kaiser Permanente Center for Health Research; KPNC = Kaiser Permanente Northern California; MV = monovalent; NA = not applicable; SMD = standardized mean or proportion difference.
* Hospitalizations with a discharge code consistent with COVID-19–like illness were included. COVID-19–like illness diagnoses included acute respiratory illness, respiratory signs or symptoms or febrile signs or symptoms using diagnosis codes from the International Classification of Diseases, Tenth Revision. Clinician-ordered molecular assays (e.g., real-time reverse transcription–polymerase chain reaction) for SARS-CoV-2 occurring ≤14 days before to <72 hours after the date of admission were included.
^† California (September 13, 2022–April 21, 2023), Indiana (September 13, 2022–April 12, 2023), Minnesota and Wisconsin (September 13, 2022–April 21, 2023), Oregon and Washington (September 13, 2022–April 14, 2023), and Utah (September 13, 2022–April 21, 2023).
^§ An absolute SMD >0.20 indicates a nonnegligible difference in variable distributions between hospitalizations for vaccinated versus unvaccinated patients, or for patients with a positive SARS-CoV-2 test result versus patients with a negative SARS-CoV-2 test result. For mRNA COVID-19 vaccination status, a single SMD was calculated by averaging the absolute SMDs obtained from pairwise comparisons of each vaccinated category versus unvaccinated. Specifically, it was calculated as the average of the absolute value of the SMDs for 1) vaccinated with only MV doses, ≥7 days earlier versus unvaccinated, 2) vaccinated with an mRNA BV dose, 7–59 days earlier versus unvaccinated, 3) vaccinated with an mRNA BV dose, 60–119 days earlier versus unvaccinated, and 4) vaccinated with an mRNA BV dose, 120–179 days earlier versus unvaccinated.
^¶ Vaccination was defined as having received the last MV or BV dose within the specified range of days before the index date, which was the date of respiratory specimen collection associated with the most recent positive or negative SARS-CoV-2 test result before the admission date, or the admission date if testing only occurred after the admission.
** Includes persons who received a single dose of Janssen (Johnson & Johnson) vaccine. Persons who received a single dose of Pfizer-BioNTech, Moderna, or Novavax vaccine were excluded from the analysis.
^†† Variant predominance was defined as the period when a variant accounted for ≥50% of all sequenced specimens in the U.S. Department of Health and Human Services region where the site is located. XBB-related sublineages predominated at Intermountain Healthcare beginning January 28, 2023; at HealthPartners, KPNC, and Regenstrief Institute beginning February 4, 2023; and at KPCHR beginning February 11, 2023.
^§§ Other race includes American Indian or Alaska Native, Asian, Native Hawaiian or other Pacific Islander, other races not listed, and multiple races. Because of small numbers, these categories were combined.
^¶¶ Because persons might have received vaccine from more than one manufacturer, columns might sum to >100%.
*** In-hospital death was identified at each individual site and was defined as a death while hospitalized or ≤28 days after admission.

Abbreviations: BV = bivalent; MV = monovalent; NA = not applicable; Ref = referent group; VE = vaccine effectiveness.
* VE was calculated as (1 − odds ratio) x 100%, estimated using a test-negative case-control design, adjusted for age, sex, race and ethnicity, geographic region, and calendar time (days since January 1, 2021).
^† Patients were considered to have critical illness if they were admitted to an intensive care unit or died. Death was identified at each individual site and was defined as a death while hospitalized or ≤28 days after admission.
^§ California (September 13, 2022–April 21, 2023), Indiana (September 13, 2022–April 12, 2023), Minnesota and Wisconsin (September 13, 2022–April 21, 2023), Oregon and Washington (September 13, 2022–April 14, 2023), and Utah (September 13, 2022–April 21, 2023).
^¶ These estimates are imprecise, which might be because of a relatively small number of persons in each level of vaccination or case status. This imprecision indicates the actual VE could be substantially different from the point estimate shown, and estimates should therefore be interpreted with caution. Additional data accrual could increase precision and allow appropriate interpretation.
** For VE against critical illness, case-patients were persons admitted to an intensive care unit or who experienced death associated with COVID-19, and control patients were persons hospitalized without COVID-19.

Abbreviations: BV = bivalent; ICU = intensive care unit; KPCHR = Kaiser Permanente Center for Health Research; KPNC = Kaiser Permanente Northern California; MV = monovalent; NA = not applicable; SMD = standardized mean or proportion difference.
* Hospitalizations with a discharge code consistent with COVID-19–like illness were included. COVID-19–like illness diagnoses included acute respiratory illness, respiratory signs or symptoms or febrile signs or symptoms using diagnosis codes from the International Classification of Diseases, Tenth Revision. Clinician-ordered molecular assays (e.g., real-time reverse transcription–polymerase chain reaction) for SARS-CoV-2 occurring ≤14 days before to <72 hours after the date of admission were included.
^† California (September 13, 2022–April 21, 2023), Indiana (September 13, 2022–April 12, 2023), Minnesota and Wisconsin (September 13, 2022–April 21, 2023), Oregon and Washington (September 13, 2022–April 14, 2023), and Utah (September 13, 2022–April 21, 2023).
^§ An absolute SMD >0.20 indicates a nonnegligible difference in variable distributions between hospitalizations for vaccinated versus unvaccinated patients or for patients with a positive SARS-CoV-2 test result versus patients with a negative SARS-CoV-2 test result. For mRNA COVID-19 vaccination status, a single SMD was calculated by averaging the absolute SMDs obtained from pairwise comparisons of each vaccinated category versus unvaccinated. Specifically, it was calculated as the average of the absolute value of the SMDs for 1) vaccinated with only MV doses, ≥7 days earlier versus unvaccinated, 2) vaccinated with an mRNA BV dose, 7–59 days earlier versus unvaccinated, 3) vaccinated with an mRNA BV dose, 60–119 days earlier versus unvaccinated, and 4) vaccinated with an mRNA BV dose, 120–179 days earlier versus unvaccinated.
^¶ Vaccination was defined as having received the last MV or BV dose within the specified range of days before the index date, which was the date of respiratory specimen collection associated with the most recent positive or negative SARS-CoV-2 test result before the admission date or the admission date if testing only occurred after the admission.
** Includes persons who received a single dose of Janssen (Johnson & Johnson) vaccine. Persons who received a single dose of Pfizer-BioNTech, Moderna, or Novavax vaccine were excluded from the analysis.
^†† Variant predominance was defined as the period when a variant accounted for ≥50% of all sequenced specimens in the U.S. Department of Health and Human Services region where the site is located. XBB-related sublineages predominated at Intermountain Healthcare beginning January 28, 2023; at HealthPartners, KPNC, and Regenstrief Institute beginning February 4, 2023; and at KPCHR beginning February 11, 2023.
^§§ Other race includes American Indian or Alaska Native, Asian, Native Hawaiian or other Pacific Islander, other races not listed, and multiple races. Because of small numbers, these categories were combined.
^¶¶ Because persons might have received vaccine from more than one manufacturer, columns might sum to >100%.
*** In-hospital death was identified at each individual site and was defined as a death while hospitalized or ≤28 days after admission.