BackgroundCirculating dietary biomarkers are not direct proxies for intake, as the biomarkers reflect not only food and supplement consumption but also nutrient absorption, metabolism, and tissue distribution. Therefore, along with nutrient intake, several other upstream factors can impact dietary biomarker concentrations, including demographic, medical history, and genetic factors. ObjectivesThe aim of this study was to explore the dietary and nondietary determinants of circulating levels of vitamins A, C, D, and E among children aged 6 mo–4 y. MethodsPlasma retinol, β-carotene, ascorbic acid, 25(OH)D, α-tocopherol, and γ-tocopherol were measured in 2887 samples from 1490 children enrolled in The Environmental Determinants of Diabetes in the Young study. Dietary intake was assessed with 3-d food records. Associations of genetic and environmental factors with biomarker concentrations were examined using multivariable linear regression models with random intercepts. ResultsAll biomarkers except retinol were positively associated with intake of the same nutrient. Inverse associations were identified between recent gastrointestinal infection and β-carotene, ascorbic acid, and α-tocopherol, whereas recent respiratory infection was associated inversely with plasma retinol. Several genetic determinants of biomarker status were identified, validating previously reported findings. For some genetic and environmental exposures, we found evidence of statistical interaction with same-nutrient intake, indicating that the association between intake and biomarker concentration is dependent on the level or status of these other exposures. For example, the association between β-carotene intake and concentration is weaker among children with a recent respiratory infection. ConclusionsOur findings suggest that nondietary exposures including childhood infections can alter micronutrient metabolism. This summary of micronutrient determinants will facilitate improved design of future analyses exploring the role of diet in childhood chronic disease etiology through a better understanding of relevant potential confounders and mediators of the diet–outcome relationships.