How did 2025–2026 vaccine effectiveness vary by influenza strain (A/H1N1, A/H3N2, B/Victoria, B/Yamagata)?
Executive summary
Interim U.S. network data for the 2024–25 season indicate vaccines provided measurable protection overall and substantial protection against hospitalization in children (up to 78%), but effectiveness varied by subtype and by data source [1] [2]. Multiple studies and reviews note that VE commonly ranges from about 10%–60% across seasons and that mismatch and evolving subclades—particularly H3N2—drive large variation in subtype-specific VE [3] [4].
1. What the U.S. networks measured: overall and by subtype
Four U.S. vaccine-effectiveness networks produced interim 2024–25 VE estimates showing the 2024–25 vaccine reduced the likelihood of medically attended influenza and hospitalizations; VE values differed by age and outcome, with outpatient VE in children/adolescents reported at roughly 32%–60% and hospitalization VE up to 63%–78% in some pediatric networks [1] [5] [2]. Those reports emphasize that most detections that season were influenza A (about 97% of positive specimens), meaning subtype performance—A/H1N1 vs A/H3N2—strongly influenced the season’s overall VE numbers [5].
2. A/H1N1 — generally better-matched and higher VE in some reports
CDC sequencing noted many A(H1N1)pdm09 detections matched vaccine-related HA clades (including viruses similar to the A/Wisconsin vaccine reference), and some network estimates showed protection against H1N1 contributing to the positive VE signal [1]. Broader syntheses and historical ranges place seasonal VE against H1N1 often toward the higher end of the typical 10%–60% band, reflecting both better antigenic stability and closer matches in many seasons [3]. Available sources do not provide a single definitive numeric VE for A/H1N1 across all U.S. networks for 2024–25 beyond the CDC network descriptions [1] [5].
3. A/H3N2 — the wild card: lower VE and subclade mismatch risk
Experts and surveillance reporting stress that H3N2 tends to yield lower VE and more severe seasons, particularly for older adults, because H3N2 evolves rapidly and can be antigenically mismatched to vaccine strains; recent emergence of H3N2 subclade K raised concerns about reduced vaccine effectiveness and potential mismatch with the vaccine reference strains [4] [6]. Some early external analyses and international preprints suggested reduced effectiveness against emergent H3N2 variants and warned that H3N2-dominant seasons historically depress overall VE [4]. Available sources do not give a single pooled VE percentage for A/H3N2 from the U.S. networks for 2024–25 beyond statements of concern and variability [1] [4].
4. Influenza B (Victoria and Yamagata) — limited circulation, different considerations
Reports note that nearly all detections in 2024–25 were influenza A, so influenza B viruses contributed little to overall VE estimates that season [5]. Public-health authorities and reviews also record that B/Yamagata detections have been effectively absent in recent years and WHO recommended removing B/Yamagata from some vaccine formulations; thus, B/Victoria has been the B lineage of practical interest [7]. Laboratory data cited in context of 2025 vaccine composition indicate good recognition of B/Victoria by vaccine strains, but the surveillance-based VE for B/Victoria in U.S. midseason networks for 2024–25 is not summarized as a single number in the available reporting [8] [7].
5. Conflicting studies and why numbers differ: Cleveland Clinic preprint vs network data
A large Cleveland Clinic observational analysis claimed a negative VE (vaccinated people had higher risk) for working-aged adults in 2024–25, estimating −26.9% effectiveness, but that finding contrasts with CDC network interim analyses that concluded vaccination reduced medically attended influenza and hospitalizations across age groups [9] [1]. Public-health summaries and professional outlets framed the Cleveland Clinic preprint cautiously and highlighted methodological and population differences; the Immunization Managers note summarized the preprint’s findings but also put them in context of broader evidence that vaccination prevented hospitalizations, particularly in children [10] [2]. These disagreements show how study design, population (healthcare workers vs general community), outcomes measured (any infection vs medically attended illness or hospitalization), and timing produce divergent VE estimates.
6. What this means for 2025–26 expectations and practical takeaways
Regulators and advisory groups adjusted 2025–26 vaccine composition (including changes relevant to manufacturing method) based on surveillance and lab data; analysts warn that if H3N2 subclade K or other mismatches dominate, subtype-specific VE—especially against H3N2—could be lower, while vaccines still generally reduce severe outcomes like hospitalization [11] [8] [4]. Reviews emphasize the usual seasonal VE range (about 10%–60% historically) and that even in imperfect years vaccination reduces hospitalizations and severe disease; these are the consistent messages across CDC, medical summaries, and reviews [3] [2].
Limitations and what’s not found in current reporting
- Available sources do not provide a single harmonized VE percentage for each subtype (A/H1N1, A/H3N2, B/Victoria, B/Yamagata) from a pooled 2024–25 U.S. analysis; network reports give age- and outcome-specific numbers and note that most detections were influenza A [1] [5].
- Available sources document absence of B/Yamagata circulation and related policy decisions but do not report specific 2024–25 VE estimates against B/Yamagata in the U.S. because that lineage was essentially absent [7].