Abstract

ABSTRACT Background Assessments of COVID‐19 vaccine effectiveness are needed to monitor the protection provided by updated vaccines against severe COVID‐19. We evaluated the effectiveness of original monovalent and bivalent (ancestral strain and Omicron BA.4/5) COVID‐19 vaccination against COVID‐19‐associated hospitalization and severe in‐hospital outcomes. Methods During September 8, 2022 to August 31, 2023, adults aged ≥ 18 years hospitalized with COVID‐19‐like illness were enrolled at 26 hospitals in 20 US states. Using a test‐negative case–control design, we estimated vaccine effectiveness (VE) with multivariable logistic regression adjusted for age, sex, race/ethnicity, admission date, and geographic region. Results Among 7028 patients, 2924 (41.6%) were COVID‐19 case patients, and 4104 (58.4%) were control patients. Compared to unvaccinated patients, absolute VE against COVID‐19‐associated hospitalization was 6% (−7%–17%) for original monovalent doses only (median time since last dose [IQR] = 421 days [304–571]), 52% (39%–61%) for a bivalent dose received 7–89 days earlier, and 13% (−10%–31%) for a bivalent dose received 90–179 days earlier. Absolute VE against COVID‐19‐associated invasive mechanical ventilation or death was 51% (34%–63%) for original monovalent doses only, 61% (35%–77%) for a bivalent dose received 7–89 days earlier, and 50% (11%–71%) for a bivalent dose received 90–179 days earlier. Conclusion Bivalent vaccination provided protection against COVID‐19‐associated hospitalization and severe in‐hospital outcomes within 3 months of receipt, followed by a decline in protection to a level similar to that remaining from previous original monovalent vaccination by 3–6 months. These results underscore the benefit of remaining up to date with recommended COVID‐19 vaccines.