Natural Experiment Supports Pediatric Influenza Vaccine Effectiveness Amid Shifting Federal Recommendations

Children aged 2 to 5 years with fall birthdays had higher influenza vaccination rates and fewer diagnoses of the disease than those with summer birthdays, with an estimated 9 to 14 prevented influenza cases per 100 vaccinated children, according to research published today in JAMA Pediatrics.¹

Reassessing Pediatric Influenza Vaccine Effectiveness Amid Shifting Recommendations

Earlier this year, the CDC removed its annual childhood influenza vaccination recommendation.² Under the revised guidance, it may only be administered after consultation with a health care provider, which is also the case for both rotavirus and hepatitis A vaccinations. Regarding its reasoning, the agency cited concerns about the lack of contemporaneous randomized data to support an annual influenza vaccination, along with historical reliance on observational data to support vaccine efficacy, which may be confounded.¹

Researchers highlighted an approach to evaluating vaccine effectiveness using observational data that is less susceptible to confounding, describing it as a “natural experiment.” This method leverages instances in which some children are, by chance, vaccinated while others are not.

Because influenza vaccines are typically available in the fall, children with fall birthdays are more likely to receive the vaccine during a scheduled well-child visit, whereas those with summer birthdays may require an additional appointment. The researchers noted that this creates a quasi-random difference in vaccination rates between otherwise similar groups.

Prior research found that children aged 2 to 5 years with fall birthdays are both more likely to be vaccinated and less likely to develop influenza than those with summer birthdays. The researchers used this natural experiment approach to assess influenza vaccine effectiveness across several recent flu seasons, arguing that it may be less vulnerable to confounding than traditional observational studies.

Using national insurance claims data from the MarketScan database, they analyzed rates of influenza vaccination and influenza diagnosis among children aged 2 to 5 with fall vs summer birthdays across 5 typical influenza seasons from 2016 to 2023, excluding the 2020-2021 and 2021-2022 seasons due to disruptions from the COVID-19 pandemic. The analysis assumed that children in both groups had similar baseline risk for influenza, attributing differences in diagnosed infections primarily to variation in vaccination timing.

The researchers also evaluated rates of noninfluenza viral infection diagnoses as a falsification test, as vaccination would not be expected to affect these outcomes; results were aggregated across all seasons.

Natural Experiment Supports Pediatric Influenza Vaccine Effectiveness

Across all 5 influenza seasons, children with fall birthdays were more likely to receive an influenza vaccine and less likely to be diagnosed with the disease than those with summer birthdays. Depending on the season, vaccination rates were 8.6 to 12.5 percentage points higher among children with fall vs summer birthdays, and diagnosis rates were 1.0 to 1.4 percentage points lower.

Meanwhile, the researchers found no differences in rates of noninfluenza viral infections between the groups, suggesting the results were not driven by residual confounding factors such as differences in care-seeking behaviors.

Overall, the analysis estimated that for every 100 children vaccinated because of the timing of their birthday, influenza vaccination prevented approximately 9 to 14 diagnosed flu cases, depending on the season.

Limitations Do Not Undermine Study Conclusions

The researchers acknowledged several study limitations, including that insurance claims may not capture all vaccinations or influenza cases. In addition, because many influenza cases are not formally diagnosed, they hypothesized that those identified in the analysis likely represent clinically significant infections that led parents to seek medical care.

Despite these limitations, the researchers expressed confidence in their findings.

“These findings are consistent with influenza vaccination significantly reducing the risk of diagnosed influenza infection among children aged 2 to 5 years each influenza season,” they wrote. “Quasi-experimental approaches such as this, which take advantage of plausibly exogenous variation in vaccination patterns, can be leveraged to assess the efficacy of influenza vaccination on an annual basis using observational data without the need for large, costly randomized trials.”

References

1. Worsham CM, Bray CF, Jena AB. Pediatric influenza vaccination efficacy. JAMA Pediatr. Published online June 1, 2026. doi:10.1001/jamapediatrics.2026.1546
2. Grossi G. CDC reduces US childhood immunization schedule from 17 to 11 diseases. AJMC^®. January 5, 2026. Accessed June 1, 2026. https://www.ajmc.com/view/cdc-reduces-us-childhood-immunization-schedule-from-17-to-11-diseases