Healthy vaccinee effect in the evaluation of updated COVID-19 vaccines in elderly populations
TL;DR
This Swedish population study reveals that healthy vaccinee bias inflates estimates of the effectiveness of updated COVID-19 vaccines in adults aged 65 years or older, highlighting the need to account for frailty and overall health status.
Abstract
Established determinants of health in the elderly help guide routines for indicated vaccine administration, while unmeasured frailty may limit vaccine access. We evaluate the performance of the 2024-2025 COVID-19 vaccine adapted to the Omicron JN.1 lineage in a Swedish population aged ≥65 years (N = 245 696). Vaccine effectiveness (VE) on COVID-19-related hospitalization and a negative control outcome (NCO; all-cause mortality) are assessed in various cohorts between October 1, 2024 to March 31, 2025. The VE was 75% (95% CI 70%–79%) overall, and 84% (95% CI 80%–87%) and 65% (95% CI 34%–82%) in individuals with and without vaccination with the prior updated COVID-19 vaccine in 2023-2024, respectively. The NCO in individuals exposed to the study vaccine in 2024-2025 was half of that seen in those not exposed to the vaccine (HR 0.43 [95% CI 0.40–0.45]). Removing individuals hospitalized with COVID-19 from this population did not change the difference in NCO (HR 0.43 [95% CI 0.41-0.46]). These findings suggest the presence of selection effects arising from under-provision of health services to elderly individuals with frailty. The healthy vaccinee effect should be considered in observational studies of the effectiveness of updated COVID-19 vaccines in elderly populations. This Swedish population study reveals that healthy vaccinee bias inflates estimates of the effectiveness of updated COVID-19 vaccines in adults aged 65 years or older, highlighting the need to account for frailty and overall health status.