660 - Vitamin D Status and Hemoglobin A1c in Children and Adolescents

Publication Number: 4646.660

Min Ji Kim, Brookdale Hospital Medical Center, Oakland Gardens, NY, United States; Fernanda Kupferman, one Brooklyn health at Brookdale, Brooklyn, NY, United States; Haiyan Lu, Brookdale Hospital Medical Center, Brooklyn, NY, United States

Background: Vitamin D is implicated in insulin sensitivity and glucose homeostasis. Although studies have examined this relationship in adults, limited research has investigated it in the pediatric population, with mixed findings. With rising pediatric obesity and prediabetes, clarifying whether vitamin D is associated with glycemic status in youth has clinical relevance.

Objective: Assess the association between serum vitamin D status and hemoglobin A1c (HbA1c) in children and adolescents, adjusting for body mass index (BMI), age, and gender; secondarily, examine the relationship between vitamin D and odds of prediabetes.

Design/Methods: This is a retrospective cohort study including children and adolescents aged 5-18 years who attended well-child visits between January 1, 2020, and January 31, 2025. The final sample (n=813) included individuals with available data on vitamin D and HbA1c. We excluded individuals with metabolic disorders other than prediabetes or those taking medications affecting vitamin D or glucose metabolism. Vitamin D was categorized as deficient ( < 20 ng/mL), insufficient (20- < 30), or sufficient (≥30). HbA1c was classified as normal ( < 5.7%) or prediabetic (5.7- < 6.5%). Analyses included descriptive statistics, chi-square tests (trend for ordered groups), and multivariable linear and logistic regression.

Results: Mean age 13.04±3.179 years; 78.1% Black and 11.7% Hispanic. Mean vitamin D 18.1±7.808 ng/mL; 64.6% deficient. Vitamin D deficiency was more common in older adolescents (p < 0.001). Vitamin D status differed by age group (p < 0.001) but not by gender (p=0.22) or race/ethnicity (p=0.431). Across BMI percentiles, there was a non-significant trend (p=0.093). In adjusted linear regression, vitamin D showed a small positive association with HbA1c (B=0.006, p=0.001); BMI was the strongest predictor (B=0.010, p< 0.001). In logistic regression, BMI increased the odds of prediabetes (OR=1.061 per unit, p< 0.001), and age decreased the odds (OR=0.933 per year, p=0.005). Vitamin D had a small positive association (OR=1.024 per ng/mL, p=0.014); gender was not significant. BMI remained the strongest predictor.

Conclusion(s): In this study, vitamin D showed only a small positive association with HbA1c, whereas BMI and age were stronger predictors of glycemic status. These findings do not support a clinically meaningful protective role of higher vitamin D levels on glycemic control in this population, suggesting that pediatric metabolic risk assessment should prioritize weight status. Prospective studies capturing sun exposure, seasonality, lifestyle, or diet are needed to clarify causality.

Table 1. Baseline Characteristics of Study Participants (N = 813)


Table 2. Association Between Vitamin D Categories and BMI, Age, HbA1c Group, Gender, and Race


Table 3. Predictors of HbA1c and Prediabetes
Panel A: Linear model (outcome HbA1c, %). Panel B: Logistic model (outcome prediabetes, HbA1c ≥5.7%). Models adjust for BMI, age, and gender. OR=odds ratio; CI=confidence interval.