Higher Vitamin D Status is Associated With Lower Insulin Resistance in Non-Diabetic U.S. Adults: A Two-Decade Nationwide Analysis
摘要
Beyond its established role in skeletal metabolism, vitamin D has been linked to glucose homeostasis and insulin sensitivity. Vitamin D receptors are expressed in tissues involved in glucose regulation, but large-scale population evidence characterizing the association between serum 25-hydroxyvitamin D [25(OH)D] and insulin resistance remains limited. We hypothesized that lower serum 25(OH)D concentrations would be associated with higher odds of insulin resistance.
MethodsWe conducted a cross-sectional analysis pooling ten NHANES cycles (2001–2018, 2021–2023), comprising 17,991 non-diabetic U.S. adults. Insulin resistance was defined using the Homeostatic Model Assessment of Insulin Resistance (HOMA-IR). Survey-weighted multivariable logistic regression examined the association between serum 25(OH)D and insulin resistance, adjusting for demographic, socioeconomic, clinical, and behavioral covariates. Subgroup and sensitivity analyses tested robustness across alternative outcome definitions.
ResultsOf 17,991 participants, 7,538 (41.9%) had insulin resistance. Each 1 nmol/L increment in 25(OH)D was associated with 0.7% lower odds of insulin resistance (OR = 0.993; 95% CI: 0.991–0.995; p < 0.001), following a linear dose–response pattern (quadratic p = 0.109). Participants with 50 nmol/L had 15% higher adjusted odds, whereas those with 120 nmol/L had 26% lower adjusted odds versus the reference level of 68.8 nmol/L. Findings were consistent across subgroup analyses, alternative HOMA-IR thresholds, TG/HDL-based definitions, and a BMI-excluded model.
ConclusionHigher serum 25(OH)D was independently associated with lower odds of insulin resistance in non-diabetic U.S. adults in a continuous, graded manner. These findings suggest that serum 25(OH)D may serve as a metabolic biomarker associated with insulin resistance, while prospective and interventional studies are needed to clarify temporality and causality.
Graphical Abstract