The study team recruited infants at 6–12 months and intended to follow them to age 5 with a composite asthma endpoint. The SARS-CoV-2 pandemic halted recruitment, and investigators stopped enrollment early. Study leadership therefore shifted the primary endpoint to atopic sensitization at age 3 years. The manuscript reports the trial proceeded in a non-confirmatory fashion after that decision.
Participants were centrally randomized to receive either a minimally-processed pasteurized full-cream cow's milk or an ultra-heat-treated semi-skimmed comparator, given daily until age 3 years. Caregivers, participants, and study staff were blinded to allocation. The source did not report additional dosing schedules or protocol deviations beyond the pandemic-related recruitment stop.
Investigators observed atopic sensitization in 28 children in the intervention arm (n=99) and 24 children in the comparator arm (n=96). Data indicated a risk ratio (RR) of 1.13 with a 95% confidence interval of 0.67–1.89. That estimate did not support a protective effect of the minimally-processed intervention on atopic sensitization at the prespecified age 3 endpoint. The article did not present subgroup analyses for sensitization beyond the counts and the reported RR.
Across the randomized cohort (n=252), asthma incidence showed a trend toward an inverse association with the intervention (hazard ratio, HR=0.55 [0.27;1.16]). This result did not meet conventional thresholds for statistical confirmation in the full cohort. A post-hoc analysis restricted to 82 children with a family history of asthma produced a stronger inverse signal (post-hoc HR=0.30 [0.09;0.93]). Investigators reported this as an exploratory finding; the source notes it arose after the change in primary endpoint and therefore requires cautious interpretation.
Investigators performed additional post-hoc analyses by age strata. In the older half of enrolled children, the intervention was inversely associated with rhinitis (post-hoc RR=0.82 [0.67–0.998]) and with cough (post-hoc RR=0.75 [0.60–0.94]). These symptom-level associations were reported as exploratory and were not part of the original prespecified outcomes. The source does not provide mechanistic data or microbiological profiles tied to these symptom results in the main text.
The trial team reported no adverse events attributed to the intervention. The source indicates no intervention-related serious safety signals were judged by investigators. The article did not include granular adverse event tables in the body text; readers are referred to supplemental material and the data platform for detailed safety listings.
Funding came from multiple sources including Dutch Longfonds, the Bavarian State-Ministry of Health, the German Center for Lung Research (DZL), and FrieslandCampina. Competing interests and patent holdings are disclosed by several authors; the source lists detailed individual declarations and patent information. The dataset and original R code are available on the protected TRACE platform upon application, and supplemental analysis code is provided with the preprint.
This early-terminated trial was underpowered for its original asthma endpoint and the primary endpoint was formally changed to atopic sensitization at age 3 after pandemic interruptions—both factors limit inferential strength. Several reported analyses are post-hoc, including subgroup signals in children with parental asthma and symptom reductions in the older half of participants; the source warns these are exploratory. The study did not report detailed microbiological or immunologic mechanism data in the main text that could explain observed clinical differences. Investigators concluded these exploratory signals justify, but do not replace, a definitive randomized trial to test whether minimally-processed cow's milk can reduce childhood asthma risk; the source did not specify timelines or regulatory milestones for such a trial.