BACKGROUND
Adult studies have demonstrated within-season declines in influenza vaccine effectiveness (VE); data in children are limited.
METHODS
We conducted a prospective, test-negative study of children 6 months-17 years hospitalized with acute respiratory illness at 7 pediatric medical centers during the 2015-2016 through 2019-2020 influenza seasons. Case-patients were children with an influenza-positive molecular test matched by illness onset to influenza-negative control-patients. We estimated VE [100% x (1 - odds ratio)] by comparing the odds of receipt of ≥1 dose of influenza vaccine ≥14 days before illness onset among influenza-positive children to influenza-negative children. Changes in VE over time between vaccination date and illness onset date were estimated using multivariable logistic regression.
RESULTS
Of 8,430 children, 4,653 (55%) received ≥1 dose of influenza vaccine. On average, 48% were vaccinated through October and 85% through December each season. Influenza vaccine receipt was lower in case-patients than control-patients (39% vs. 57%, p < 0.001); overall VE against hospitalization was 53% (95% CI: 46%-60%). Pooling data across 5 seasons, the odds of influenza-associated hospitalization increased 4.2% (-3.2%-12.2%) per month since vaccination, with an average VE decrease of 1.9% per month (n = 4,000, p = 0.275). Odds of hospitalization increased 2.9% (95% CI: -5.4%-11.8%) and 9.6% (95% CI: -7.0%-29.1%) per month in children ≤8 years (n = 3,084) and 9-17 years (n = 916), respectively. These findings were not statistically significant.
CONCLUSIONS
We observed minimal, not statistically significant within-season declines in VE. Vaccination following current ACIP guidelines for timing of vaccine receipt remains the best strategy for preventing influenza-associated hospitalizations in children.