Assess whether the commonly cited 40%–60% influenza vaccine effectiveness (VE) range reflects performance across historical US seasonal conditions captured by the CDC Flu VE Network.
Approach:
Evidence set: Meta-analysis of peer-reviewed Flu VE Network end-of-season estimates for 2004–2005 through 2023–2024; estimates were unavailable for two seasons.
Methods: Pooled 18 season-level estimates representing 96,335 participants using a random-effects inverse-variance model; assessed heterogeneity and compared seasons with and without reported antigenic mismatch.
Key Findings:
Overall pooled VE was 41.3% (95% CI, 35.6%–46.5%), with substantial between-season heterogeneity (I² = 82.8%; P < .001).
Season-specific VE ranged from 10% to 60%; 9 of 18 estimates were below 40%, 9 were within 40%–60%, and none exceeded 60%.
The prediction interval for VE in one future season was 15.6%–59.3%.
Pooled VE was 36.2% (95% CI, 29.9%–42.0%) in seasons with reported mismatch and 50.4% (95% CI, 44.2%–56.0%) in seasons without reported mismatch; the difference was significant (P = .001).
Antigenic mismatch in at least one vaccine component was reported in 11 of 18 seasons.
Interpretation:
The authors report that the 40%–60% heuristic reflects performance under favorable biological conditions, while historical Flu VE Network estimates indicate VE of approximately 40% in a typical season. VE varied substantially across seasons, and pooled VE was lower in seasons with reported antigenic mismatch.
Limitations:
The analysis was restricted to VE estimates from one US network.
The mismatch comparison was associational; seasons also differed in population immunity, prior vaccination, and other factors.
Participant-level characteristics were unavailable, and formal quality scoring was not performed.
Conclusion:
The authors state that influenza VE expectations should reflect historical performance across the full range of seasons, rather than only the most favorable seasons.
