Waist-to-height ratio for OSA risk: a comparative analysis of NHANES and a clinical data

This study aims to demonstrate the relationship between waist-to-height ratio (WHtR) and obstructive sleep apnea (OSA) risk by analyzing data from the National Health and Nutrition Examination Survey (NHANES) and clinical cases.

This study utilized data from the National Health and Nutrition Examination Survey (NHANES), a nationally representative cross-sectional survey of the US non-institutionalized population, and a retrospective clinical dataset of 200 patients who underwent polysomnography (PSG) at the Sleep Center of Tianjin Chest Hospital, China. Multivariate logistic regression analysis, restricted cubic spline (RCS) analysis, receiver operating characteristic (ROC) curves and subgroup analysis were used to assess the association between WHtR and the OSA risk.

WHtR was positively associated with OSA risk in both datasets. In the adjusted model, each 0.1-unit increase in WHtR was associated with higher odds of OSA (NHANES: OR 2.780, 95% CI: 2.536-3.048; clinical: OR 11.998, 95% CI: 6.010-23.953; both P < 0.001). RCS analysis revealed a significant non-linear relationship (P for non-linearity <0.001), with risk increasing rapidly beyond thresholds of 0.589 (NHANES) and 0.523 (clinical). ROC analysis showed that WHtR achieved an AUC of 0.727 in NHANES and 0.883 in clinical data, outperforming BMI in the clinical setting. DeLong's test confirmed that WHtR's AUC was significantly larger than that of BMI (Z = 7.186, P < 0.001), while no significant difference was detected between WHtR and waist circumference (Z = 0.543, P = 0.587) in the clinical data. In subgroup analyses of clinical data, WHtR maintained superior predictive value across sex and age strata (all P < 0.001 vs. BMI by DeLong's test).

WHtR is a strong indicator of an increased OSA risk. WHtR has superior predictive value for OSA compared to BMI in clinical data, but shows equivalence in population data. WHtR demonstrates promise as a screening tool in a high-prevalence, clinically referred Chinese population.