ABSTRACT
The COVID-19 vaccination campaign in Belgium aimed to reduce disease spread and severity. We quantified the observed vaccine effectiveness (VE) against symptomatic infection (VEi) and hospitalization (VEh). Exhaustive data on testing and vaccination was combined with a clinical hospital survey. We estimated VEi using a test negative design and VEh using a proportional hazard analysis. We controlled for prior infection, age, sex, province of residence and calendar week of sampling. Variant of concern specific VE-estimates were obtained by time since vaccination from July 2021 to April 2022. We included 1,433,135 persons. VEi against Delta waned from an initial estimate of 81% (95%CI 80-82) to 56% (95%CI 56-57) 100-150 days after primary-vaccination. Booster-vaccination increased initial VEi to 84% (95%CI 83-85). Against Omicron, an initial VEi of 37% (95%CI 34-40) waned to 18% (95%CI 17-20) 100-150 days after primary-vaccination. Booster-vaccination increased VEi to 52% (95%CI 51-53) and waned to 25% (95%CI 24-27) 100-150 days after vaccination. Hybrid immunity conferred by prior infection and booster-vaccination outperformed booster-vaccination only even if the infection was over one year ago, 67% (95%CI 66-68). Initial VEh for booster-vaccination decreased from 93% (95%CI 93-94) against Delta to 87% (95%CI 85-89) against Omicron. VEh for Omicron waned to 66% (95%CI 63-70) 100-150 days after booster-vaccination. In conclusion, we report significant immune-escape by Omicron. VEh was less affected than VEi and immune-escape was attenuated by booster-vaccination. Waning further reduced VEi- and VEh-estimates. Infection-acquired immunity offered additional protection against symptomatic infection in vaccinated persons which lasted at least one year.