CONTEXT: Evidence on the link between iron status markers and insulin resistance (IR) is limited. OBJECTIVE: We aimed to explore the relationship between iron status and IR among US adults. METHODS: This study involved 2993 participants from the National Health and Nutrition Examination Survey (NHANES) 2003-2006, 2017-2020. IR is characterized by a homeostatic model assessment (HOMA)-IR value of ≥2.5. Weighted linear and multivariable logistic regression analyses were used to examine the linear relationships between iron status and IR. Furthermore, restricted cubic splines (RCS) were used to identify the nonlinear dose-response associations. Stratified analyses by age, sex, body mass index, and physical activity were also performed. Last, a receiver operating characteristic (ROC) curve was used to evaluate the predictive value of iron status in IR. RESULTS: In weighted linear analyses, serum iron (SI) exhibited a negative correlation with HOMA-IR (β -0.03, 95% CI -0.05, -0.01, P = .01). In weighted multivariate logistic analyses, iron intake and the serum transferrin receptor (sTfR) were positively correlated with IR (OR 1.02, 95% CI 1.00-1.04, P = .04
OR 1.07, 95% CI 1.02-1.13, P = .01). Also, SI and transferrin saturation (TSAT) were negatively correlated with IR (OR 0.96, 95% CI 0.94-0.98, P <
.0001
OR 0.98, 95% CI 0.97-0.99, P <
.001) after adjusting for confounding factors. RCS depicted a nonlinear dose-response relationship between sTfR and TSAT and IR. This correlation remained consistent across various population subgroups. The ROC curve showed that TSAT performed better than iron intake, SI and sTfR in ROC analyses for IR prediction. CONCLUSION: All biomarkers demonstrated significantly lower risk of IR with increasing iron levels, which will contribute to a more comprehensive and in-depth understanding of the relationship between the 2 and provide a solid foundation for future exploration of the mechanisms underlying their relationship.