Importance: Recent sublineages of the SARS–CoV–2 Omicron variant, including BA.4 and BA.5, may be associated with greater immune evasion and less protection against COVID–19 following vaccination.
Objective: To evaluate the association between COVID–19 mRNA vaccination with 2, 3, or 4 doses among immunocompetent adults and the risk of medically attended COVID–19 illness during a period of BA.4/BA.5 predominant circulation; to evaluate the relative severity of COVID–19 in hospitalized cases across Omicron BA.1, BA.2/BA.2.12.1, and BA.4/BA.5 sublineage periods.
Setting, Design and Participants: Test-negative study of adults with COVID–19–like illness (CLI) and molecular testing for SARS–CoV–2 conducted in 10 states from December 16, 2021, to August 20, 2022.
Exposure: mRNA COVID–19 vaccination.
Main Outcomes and Measures: Emergency department/urgent care encounters, hospitalizations, and admission to the intensive care unit (ICU) or in-hospital death. The adjusted odds ratio (OR) for the association between prior vaccination and medically attended COVID-19 was used to estimate VE, stratified by care setting and vaccine doses (2, 3, or 4 doses vs 0 doses as reference group). Among hospitalized case-patients, demographic and clinical characteristics and in-hospital outcomes including ICU admission and death were compared across sublineage periods.
Results: Between June 19 – August 20, 2022, 82,229 ED/UC and 21,007 hospital encounters were included for the BA.4/BA.5 vaccine effectiveness analysis. Among adults hospitalized with CLI, the adjusted odds ratio (OR) was 0.75 (95% CI: 0.68-0.83) for receipt of 2 vaccine doses at ≥150 days after receipt, 0.32 (95% CI: 0.20–0.50) for a third dose 7–119 days after receipt, and 0.64 (95% CI: 0.58–0.71) for a third dose ≥120 days (median 235 days) after receipt for cases vs controls. For COVID-19-associated hospitalization, among patients ages ≥65 years 7-59 and ≥60 days (median 88 days) after a fourth dose, ORs were 0.34 (95% CI: 0.25–0.47) and 0.43 (95% CI: 0.34–0.56), respectively. Among hospitalized cases, ICU admission and/or in-hospital death occurred in 21.4% during the BA.1 vs 14.7% during the BA.4/BA.5 period (standardized mean difference: 0.17).
Conclusion: VE against medically attended COVID-19 illness decreased over time since last dose; receipt of one or two booster doses increased effectiveness over a primary series alone.