PT - JOURNAL ARTICLE AU - Grace J. Lee AU - Catherine S. Birken AU - Patricia C. Parkin AU - Gerald Lebovic AU - Yang Chen AU - Mary R. L’Abbé AU - Jonathon L. Maguire ED - , TI - Consumption of non–cow’s milk beverages and serum vitamin D levels in early childhood AID - 10.1503/cmaj.140555 DP - 2014 Nov 18 TA - Canadian Medical Association Journal PG - 1287--1293 VI - 186 IP - 17 4099 - http://www.cmaj.ca/content/186/17/1287.short 4100 - http://www.cmaj.ca/content/186/17/1287.full SO - CMAJ2014 Nov 18; 186 AB - Background: Vitamin D fortification of non–cow’s milk beverages is voluntary in North America. The effect of consuming non–cow’s milk beverages on serum 25-hydroxyvitamin D levels in children is unclear. We studied the association between non–cow’s milk consumption and 25-hydroxyvitamin D levels in healthy preschool-aged children. We also explored whether cow’s milk consumption modified this association and analyzed the association between daily non–cow’s milk and cow’s milk consumption.Methods: In this cross-sectional study, we recruited children 1–6 years of age attending routinely scheduled well-child visits. Survey responses, and anthropometric and laboratory measurements were collected. The association between non–cow’s milk consumption and 25-hydroxyvitamin D levels was tested using multiple linear regression and logistic regression. Cow’s milk consumption was explored as an effect modifier using an interaction term. The association between daily intake of non–cow’s milk and cow’s milk was explored using multiple linear regression.Results: A total of 2831 children were included. The interaction between non–cow’s milk and cow’s milk consumption was statistically significant (p = 0.03). Drinking non–cow’s milk beverages was associated with a 4.2-nmol/L decrease in 25-hydroxyvitamin D level per 250-mL cup consumed among children who also drank cow’s milk (p = 0.008). Children who drank only non–cow’s milk were at higher risk of having a 25-hydroxyvitamin D level below 50 nmol/L than children who drank only cow’s milk (odds ratio 2.7, 95% confidence interval 1.6 to 4.7).Interpretation: Consumption of non–cow’s milk beverages was associated with decreased serum 25-hydroxyvitamin D levels in early childhood. This association was modified by cow’s milk consumption, which suggests a trade-off between consumption of cow’s milk fortified with higher levels of vitamin D and non–cow’s milk with lower vitamin D content.See related commentary on page 1277 and at www.cmaj.ca/lookup/doi/10.1503/cmaj.141197