CitationKannan S, Visintainer P, Ganguri HB, Conroy R, Gudala M, Wittcopp C. The main objective of this study is to determine predictors of vitamin D deficiency in obese children. Methods: Children aged 5-14 years with body mass index z-score (BMIZ)>95% were enrolled. Data included height, weight, ethnicity, season, dietary intake, serum 25-hydroxy vitamin D test: (25(OH)D), parathyroid hormone, calcium, insulin, glucose, high sensitivity C-reactive protein (CRP), lipids, aspartate aminotransferase (AST), alanine transaminase (ALT) and waist circumference. BMIZ calculation in children was categorized into ≥2.5 (high) and <2.5 (low). Serum 25(OH)D levels were dichotomized as <20 ng/mL (deficient) or ≥20 ng/mL (sufficient). Results: Ninety-one children completed the study. Mean BMIZ was 31.3±6.3 kg/m 2 . Mean 25(OH)D level was 21.2±7.6 ng/mL. A base logistic regression model showed odds of a deficient 25(OH)D was more than 3-times higher in children with a high BMIZ (OR: 3.35, 95% CI: 1.07-10.43). African-American children were more likely than Caucasian children to have low 25(OH)D (OR: 6.68, 95% CI: 1.31-36.13). Children enrolled in seasons other than winter were less likely to have a low 25(OH)D. Adjustment for nutrients and serum cardiovascular disease risk markers did not significantly alter the relationship between low 25(OH)D and BMIZ. Conclusions: The risk for vitamin D deficiency increases in children with severe obesity (BMIZ≥2.5 SD) and does not significantly change by season, ethnicity, nutrient intake, or cardiovascular disease risk parameters.