Breastfeeding status up to six months of age was associated with a lower risk of influenza infection during the first year of life among preterm infants, a multicenter cohort study found. The adjusted risk ratio was 0.66, with a 95% confidence interval of 0.48 to 0.89. A risk ratio is a relative comparison between groups, while a confidence interval shows the uncertainty around the study's estimate. Because the research examined an association in a cohort, it cannot determine whether breastfeeding itself caused the difference.
The study also considered infections caused by respiratory syncytial virus, rotavirus and norovirus. For those infections, the results pointed to a trend toward lower risk, but no numerical effect estimate or confidence interval was reported. That leaves influenza as the only infection in the reported results for which the size of the association and its stated uncertainty can be described.
How the study was assembled
The researchers asked whether breastfeeding status up to six months was associated with viral infection incidence during the first year of life. They used data from the Japanese National Hospital Organization Neonatal Intensive Care Unit Registry, covering 2010 to 2014 and 2016 to 2018. The study was a multicenter cohort, so its findings describe an association rather than a tested feeding intervention.
The final analysis included 3,368 preterm infants who completed follow-up at one year. Breastfeeding exposure and infection history were collected through parental questionnaires when the infants reached that age. The comparison therefore depended on information reported by parents at the one-year follow-up point.
The influenza result stood apart
To estimate differences in infection risk, the researchers used modified Poisson regression. The model generated risk ratios and 95% confidence intervals, allowing the reported association to be expressed as a relative risk estimate with a stated range of uncertainty. For influenza, that adjusted estimate was 0.66, with the interval running from 0.48 to 0.89.
For the other viral infections, the report gives a directional finding rather than a quantified result. It says breastfeeding indicated a trend toward lower risk, without giving an effect size or confidence interval for respiratory syncytial virus, rotavirus or norovirus. The available results therefore do not show how large those associations were or how they compared with the influenza estimate.
What the study leaves unresolved
Several details affect how confidently the result can be read. The report says that both breastfeeding exposure and infection history came from parental questionnaires at one year, but it does not describe questionnaire validation or how infections were confirmed. It also reports an adjusted estimate without identifying the adjustment covariates, explaining how missing data were handled or describing sensitivity analyses.
The analyzed group consisted of infants who completed one-year follow-up. The report does not give the initial eligible population, the number lost to follow-up or the sizes of the breastfeeding groups. That leaves important parts of the cohort's composition unclear, including how many infants contributed to each comparison.
The authors call for more evidence
The authors interpret breastfeeding up to six months as potentially associated with lower viral infection risk in preterm infants. They suggest that breastfeeding support or promotion might be relevant to infection-prevention strategies and health outcomes. At the same time, they call for further research to confirm the associations and evaluate long-term health implications. The study itself does not establish long-term health effects.
The version of record was published on 21 August 2026. Written informed consent for participation was obtained from parents or legal guardians at enrollment. The work was supported by the National Hospital Organization Network Study Grant, and the authors declared no competing interests. The analysis datasets cannot be shared publicly or provided on request because ethics approval and participant consent for data sharing were not obtained.
Paper data and sources
Original title: Breastfeeding and viral infections in preterm infants: a multicenter cohort study.
Authors: Yusaku Kamiya, Akihito Takeuchi, Kazue Nakamura et al.
Journal/Repository: Pediatric research
Status: Peer-reviewed
First online: 2026-08-21
DOI: 10.1038/s41390-026-05373-x
Original paper · Full text