Hemoglobin-to-Red cell distribution width ratio and asthma risk: cross-population validation and vitamin D mediation analysis

Purpose This study aimed to explore the association between hemoglobin-to-red blood cell distribution width ratio (HRR) and asthma risk using a large U.S. population-based cohort (NHANES), and to validate the core association in an independent Chinese clinical cohort (Ganzhou People's Hospital) to improve cross-population generalizability. The mediating role of vitamin D in the HRR-asthma association was further explored only in the NHANES discovery cohort. Patients and methods Data from NHANES 2011–2018 and 2021–2023 ( n = 19,140) were used as the discovery cohort. An independent clinical cohort from Ganzhou People's Hospital ( n = 519) was used for external validation. Logistic regression models were constructed to assess the HRR-asthma association with stepwise adjustment for confounders. Variance inflation factor (VIF) was used to examine multicollinearity. Discrimination (AUC), calibration (Hosmer–Lemeshow test), and clinical utility (DCA) were evaluated. Mediation analysis was only performed in the NHANES cohort due to unavailability of vitamin D data in the clinical cohort. Results In the NHANES cohort, fully adjusted logistic regression showed a significant inverse association between HRR and asthma (OR = 0.60, 95%CI: 0.45–0.80, P < 0.001), and vitamin D completely mediated this association (indirect effect β =0.011, P = 0.001). In the Ganzhou cohort, the inverse association was consistently validated (fully adjusted OR = 0.21, 95%CI: 0.06–0.75, P = 0.017). The HRR-integrated multivariable model showed moderate discrimination in the NHANES cohort (AUC=0.539, 95%CI: 0.528–0.551) with good calibration ( P = 0.32). In the Ganzhou cohort, the multivariable prediction model exhibited moderate discriminatory ability (AUC=0.718, 95%CI: 0.667–0.770), whereas HRR alone showed low-to-moderate discrimination (AUC=0.375, equivalent to 0.625 for the inverse association). Conclusion HRR is inversely associated with asthma risk across U.S. general and Chinese clinical populations. Vitamin D fully mediates the association in NHANES but could not be validated in the clinical cohort. HRR is a simple, low-cost, cross-population risk-associated component within a comprehensive clinical prediction model, rather than a standalone diagnostic marker.