Association Between Sleep Duration and Obesity Phenotypes in Apparently Healthy Korean Adults: A Cross-Sectional Study Using the Korea National Health and Nutrition Examination Survey 2013–2015

Background and Objectives: Evidence on the association between sleep duration and obesity phenotypes jointly considering adiposity and metabolic health is limited, particularly in adults without known chronic diseases. We aimed to examine this association in apparently healthy Korean adults. Materials and Methods: In this cross-sectional study, we analyzed 7808 apparently healthy adults aged ≥20 years from the Korea National Health and Nutrition Examination Survey 2013–2015, assigned to metabolically healthy normal weight (MHNW), metabolically abnormal normal weight (MANW), metabolically healthy obesity (MHO), or metabolically abnormal obesity (MAO) phenotypes. Self-reported sleep duration was grouped as ≤5, 6, 7, 8, or ≥9 h/day. Survey-weighted multinomial logistic regression was performed with MHNW and ≥9 h as references, adjusting for age and sex (Model 1) and socioeconomic and lifestyle factors (Model 2). Sensitivity analyses used 7 h/day as the reference and a stricter definition of metabolic health; effect modification by sex and age was examined. Results: The proportion of MHNW increased from 61.6% (≤5 h) to 76.0% (≥9 h) (p < 0.001). In Model 2, sleeping ≤5 h was associated with higher odds of MHO (odds ratio [OR] 1.380, 95% confidence interval [CI] 1.098–1.734) and MAO (OR 1.472, 95% CI 1.109–1.953) relative to 7 h, and these associations remained significant after false discovery rate correction, whereas none of the associations relative to ≥9 h, including that with MANW, did. Under the stricter definition, the association with MAO persisted, whereas that with MHO was no longer statistically significant despite a similar effect estimate (vs. 7 h: OR 1.394, 95% CI 0.926–2.098); no significant interactions by sex or age were observed for MHO or MAO. Conclusions: In apparently healthy Korean adults, short sleep duration was associated with higher odds of MHO and MAO, most consistently MAO, but not robustly with MANW. Because of the cross-sectional design, these associations do not imply causality, and prospective studies are needed to determine whether sleep duration is a modifiable target for metabolic health.