
New evidence backs pediatric flu vaccination efficacy as CDC drops it from childhood schedule
Key Takeaways
- In each of 5 influenza seasons from 2016 to 2023, children aged 2 to 5 years with fall birthdays, who were more likely to be vaccinated due to well-child visit timing, had consistently lower rates of diagnosed influenza than those with summer birthdays, providing quasi-experimental evidence of vaccine efficacy.
- Wald estimates indicated that for every 100 children vaccinated due to their fall birthday who would not otherwise have been vaccinated, 9 to 14 diagnosed influenza cases were averted per season, a magnitude consistent with older randomized trial data.
New research using a natural experiment design found consistent reductions in diagnosed influenza among vaccinated young children in every season studied from 2016 to 2023.
As the AAP continues to recommend annual influenza vaccination for all children 6 months and older, in direct opposition to the CDC's January 2026 schedule overhaul, a research letter recently published in JAMA Pediatrics provides timely quasi-experimental evidence that the vaccine reduces clinically meaningful influenza infections in children aged 2 to 5 years across multiple seasons.¹
The CDC's revised schedule, released under the direction of HHS Secretary Robert F. Kennedy Jr., downgraded flu vaccination from a universal recommendation to a shared clinical decision-making category, citing concerns about a lack of contemporaneous randomized data and potential confounding in the historical observational evidence base.¹ The AAP responded swiftly, calling the decision "dangerous and unnecessary."²
"At a time when parents, pediatricians and the public are looking for clear guidance and accurate information, this ill-considered decision will sow further chaos and confusion and erode confidence in immunizations," said AAP President Andrew D. Racine, MD, PhD, FAAP.² The AAP, along with other health organizations, is currently suing the Trump administration over the reduced childhood immunization schedule.²
Natural experiment design using birthday-based vaccination patterns in young children
The study analyzed national insurance claims data (MarketScan) from 5 influenza seasons between 2016 and 2023, excluding the COVID-19–disrupted 2020–2021 and 2021–2022 seasons.¹ Rather than relying on traditional test-negative observational designs, which are vulnerable to confounding from health-seeking behavior, the investigators exploited a natural experiment rooted in the timing of well-child visits.
Young children with fall birthdays are more likely to receive influenza vaccination than those with summer birthdays, because their annual well-child visits coincide with the autumn window when the seasonal vaccine becomes available.¹ Children born in summer must schedule a separate appointment to get vaccinated, a logistical barrier that effectively reduces uptake without any systematic difference in underlying health risk between the 2 groups. This birthday-based variation allows differences in influenza diagnosis rates between the groups to be credibly attributed to differences in vaccination rates rather than confounders such as socioeconomic status or general care-seeking behavior.¹
Vaccination and influenza diagnosis rates across 5 seasons
In each of the 5 seasons analyzed, fall-birthday children were both more likely to have been vaccinated and less likely to receive a diagnosed influenza infection.¹ Vaccination rates among fall-birthday children exceeded those of summer-birthday children by 8.6 to 12.5 percentage points, depending on the season, while influenza diagnosis rates were 1.0 to 1.4 percentage points lower in the fall-birthday group.¹
The investigators calculated Wald estimates, a standard instrumental variable approach, to convert these differences into a local average treatment effect: the number of diagnosed influenza cases averted per 100 children vaccinated specifically because of the fall timing of their birthday.¹ Across all 5 seasons, that estimate ranged from approximately 9 to 14 fewer cases per 100 children vaccinated, a finding the authors note is consistent with efficacy observed in older randomized controlled trials.¹
To test whether residual confounding from care-seeking behavior could explain the results, the team examined diagnosis rates for non-influenza viral infections, conditions not expected to be affected by influenza vaccination, as falsification outcomes.¹ No meaningful differences were found between the fall- and summer-birthday groups for any of the 4 falsification conditions, including acute nasopharyngitis, viral intestinal infections, viral infections of unspecified site, and acute laryngitis and tracheitis.¹
AAP's diverging stance and the broader policy conflict
The new findings land in the middle of an unusually public split between federal health agencies and the country's largest pediatric professional organization. After the CDC released its revised schedule in January 2026 — reducing the number of universally recommended vaccines from 18 diseases to 11 — the AAP published its own 2026 immunization schedule, which kept all prior routine vaccine recommendations intact, including annual influenza vaccination.³
"As there is a lot of confusion going on with the constant new recommendations coming out of [the] federal government, it is important that we have a stable, trusted, evidence-based immunization schedule to follow, and that's the AAP schedule," said Pia Pannaraj, MD, MPH, a member of the AAP Committee on Infectious Diseases and a professor of pediatrics at the University of California San Diego.³ The downgrade of flu vaccination is particularly notable given that the CDC's schedule revision coincided with a flu season that saw some of the highest pediatric influenza mortality in recent years.
Limitations and next steps
The primary limitation the authors acknowledge is that insurance claims data may not capture all vaccinations or all influenza cases.¹ Because many infections do not prompt a medical visit, the estimates reflect vaccine efficacy against clinically significant disease that led families to seek care, arguably the most policy-relevant outcome. The authors note this limitation is unlikely to introduce bias, since underreporting should affect fall- and summer-birthday children similarly.¹
The quasi-experimental framework also estimates a local average treatment effect, specifically, the impact of vaccination among children who were vaccinated because of their fall birthday who otherwise would not have been, which may not generalize to all children or to populations with different baseline vaccination behaviors.¹ The authors suggest this methodology could serve as a scalable, annual approach to assessing influenza vaccine efficacy using observational data, without requiring large, costly randomized trials.¹
References
Worsham CM, Bray CF, Jena AB. Pediatric influenza vaccination efficacy. JAMA Pediatrics. Published online June 1, 2026. doi:10.1001/jamapediatrics.2026.1546
Racine AD. AAP: CDC plan to remove universal childhood vaccine recommendations 'dangerous and unnecessary.' AAP News. January 5, 2026. Accessed June 8, 2026.
https://publications.aap.org/aapnews/news/34104/AAP-CDC-plan-to-remove-universal-childhood-vaccine Pannaraj P. AAP's 2026 immunization schedule keeps routine recommendations intact after overhaul of federal schedule. AAP News. January 26, 2026. Accessed June 8, 2026.
https://publications.aap.org/aapnews/news/34141/AAP-s-2026-immunization-schedule-keeps-routine





