(AU) Real-World Effectiveness of mRNA COVID-19 Vaccines Among Elderly U.S. Nursing Home Residents

Sunday, August 28, 2022

8:00 AM – 1:15 PM CEST

Location: Hall E South

Publication Number: 1045

Presenting Author(s)

Hector S. Izurieta, MPH, MD
– US Food and Drug Administration, Silver Spring, Maryland, United States

Co-Author(s)

Yun Lu
– US Food and Drug Administration, MD, United States
Arnstein Lindaas
– Acumen LLC, United States
Kathryn Matuska, BA
– Data and Policy Analyst, Acumen LLC, San Francisco, CA, United States
Xiangyu Shi
– Acumen LLC, Cupertino, CA, United States
Mikhail Menis, PharmD
– US Food and Drug Administration, Olney, Maryland, United States
Whitney Steele
– US Food and Drug Administration, Silver Spring, MD, United States
Michael Wernecke
– Acumen LLC, United States
Yoganand Chillarige, MPA
– Research Manager, Acumen LLC, Burlingame, California, United States
Hui Lee Wong, PhD
– Associate Director for Innovation and Development, US Food and Drug Administration, Silver Spring, Maryland, United States
Jeffrey Kelman
– CMS, United States
Richard A. Forshee
– US Food and Drug Administration, United States

Background: COVID-related hospitalization/death disproportionately affect older adults, especially nursing home (NH) residents.

Objectives: To investigate the real-world vaccine effectiveness (VE) of mRNA COVID-19 vaccines among elderly U.S. NH long-term stay residents in the pre-Delta and Delta variant periods.

Methods: Retrospective cohort study using Medicare claims data among beneficiaries ages ≥65 years residing in U.S. NHs December 2020–November 2021. The study allowed for time-varying exposures: residents contributed time to unvaccinated, one, two (fully vaccinated), and three/booster-dose vaccinated cohorts. Primary outcomes were severe COVID-related outcomes (hospitalizations and deaths). Marginal Structural Cox regression models were used to estimate hazard ratios for outcomes in the vaccinated and unvaccinated cohorts. Confounding was adjusted for by using propensity score weighting and by including time-varying factors in the outcome model (census-tract level COVID-19 circulation, proportion Delta).

Results: The source population consisted of >1,015,000 Medicare beneficiaries residing in >15,000 NHs as of December 2020. After applying inclusion criteria, the eligible population included >336,000 elderly NH long-term stay Medicare beneficiaries. Over 50% of NH residents received >1 dose of a COVID-19 vaccine between December 27, 2020–January 16, 2021, and >70% received >1 dose by January 31, 2021.

We conducted a preliminary analysis using data accrued prior to August 14, 2021. The unvaccinated and fully vaccinated cohorts were well-balanced after propensity score weighting except for level of COVID-19 circulation, prior influenza vaccination, and prior COVID-19 diagnosis. The crude rate of post-weighting COVID-19 hospitalization per 100,000 person-weeks was 145.32 and 18.82 among the unvaccinated and fully vaccinated, respectively; and 139.41 and 39.95, respectively, for the death outcome. During the high COVID-19 circulation (5,000 COVID-19 cases over 4 weeks per 100,000 persons), pre-Delta period VE was 74% (95% CI 70–78%) against COVID-related death and 72% (95% CI 66–77%) against COVID-related hospitalization, 99% Delta period VE was 64% (95% CI 54–72%) against death and 43% (95% CI 24–58%) against hospitalization.

Conclusions: COVID-19 mRNA vaccines were effective against severe COVID-related outcomes among elderly NH residents. Limitations of this study include potential under-ascertainment of vaccinations received outside of the Medicare system, which could lead to an underestimate of VE.