Objective. Recent studies reported the presence of unique subsets of body size phenotypes that are more susceptible or more resistant to the development of obesity-associated metabolic disorders, although the underlying mechanism is not yet fully elucidated. We investigated the association between body size phenotypes and sleep duration after adjusting potential confounding factors. Materials and methods. We analyzed data from the Korean National Health and Nutrition Examination Survey V (KNHANES V), a nation-wide, population-based health survey including 9077 Korean adults. The average amount of sleep per night was categorized as: <= 6, 7, 8, and >= 9 h. Body size phenotypes were classified based on body mass index (BMI) and presence of metabolic syndrome; metabolically healthy and normal weight (MHNW), metabolically abnormal but normal weight (MANW), metabolically healthy but obese (MHO), and metabolically abnormal obese (MAO). Results. According to sleep duration, there were significant differences in age, gender, BMI, waist circumference, and blood pressure (all P <0.05). Multivariate analysis showed that obese groups (MHO and MAO) had significantly shorter sleep durations than non-obese groups (MHNW and MANW) (6.78 +/- 0.04 vs. 6.93 +/- 0.03, P <0.001). Sleep duration was significantly different according to body size phenotype, irrespective of confounding factors, such as age, gender, smoking status, alcohol consumption, physical activity, income, and education (MHO; 6.73 +/- 0.05, MAO; 6.82 +/- 0.05, MHNW; 6.94 +/- 0.04, and MANW; 6.91 +/- 0.05; P <0.001). Conclusion. Sleep duration is independently associated with body size phenotype after adjusting for confounding factors in the Korean population. (C) 2015 Elsevier Inc. All rights reserved.