Real-world effectiveness of influenza vaccine over a decade during the 2011–2021 seasons—Implications of vaccine mismatch

BACKGROUND: Influenza imposes a significant healthcare burden in Korea, leading the government to initiate a national immunization program. Previous studies on vaccine effectiveness (VE) were limited to single-season estimation in Korea. METHODS: This multicenter prospective cohort study enrolled patients with influenza-like illnesses at 10 medical centers in Korea from 2011 to 2021. The demographic and clinical data were collected from questionnaire surveys and electronic medical records. Using a test-negative design, we aimed to investigate the effectiveness of a seasonal influenza vaccine for antigenic matching of the vaccine and circulating viral strains over 10 seasons. RESULTS: Overall, 5322 adults aged ≥65 years were enrolled. Only three (33.3 %) of nine seasons showed >70 % antigenic match between vaccine and circulating strains. Influenza VE was significantly variable by season, ranging from -46.9 % (95 %confidence interval CI: -127.6-5.2) in the 2011/12 season to 47.7 % (95 %CI: 22.6-64.7) in the 2016/17 season. A significant difference was observed in the VE depending on whether the vaccine strains matched with epidemic strains: 28.8 % (95 %CI: 8.8-44.8) in matched seasons versus -12.0 % (95 %CI: -30.0-3.7) in mismatched seasons. Across the study period, influenza-related hospitalizations were reduced by 13.6 % (95 %CI: 0.7-24.8) with vaccination. In a subgroup analysis, the VE against influenza-related hospitalization was 48.4 % (95 %CI 29.6-62.2) in A/H3N2 dominant seasons and 53.8 % (95 %CI: -73.4-87.7) in A/H1N1 dominant seasons, respectively. CONCLUSION: Influenza vaccine mismatch was frequent over the study period, leading to negligibly low VE in mismatched seasons. Influenza vaccination reduces the risk of influenza-related hospitalizations.