Abstract
BACKGROUND: Short sleep duration has been consistently associated with obesity. However, few studies in adults have assessed whether this association varies by sex and race. METHODS: In the 2013-2016 examination of 1116 participants from the biracial Bogalusa Heart Study, habitual sleep duration was measured in categories (<6 hours, 6-7 hours, 7-8 hours, > 8 hours) using self-report. Anthropometry was performed and adiposity was assessed using body mass index (BMI) and waist circumference. Linear regression models estimated the relationship between sleep duration and adiposity adjusting for age, education, employment, bed partner status, depressive symptoms, smoking status, physical activity, and the use of alcohol and sleeping pills. Effect modification by sex and race was examined using cross-product terms in the models and marginal means were reported. RESULTS: Of 1116 participants (mean age 48.2 years), 58.3% were female, and 31.7% were Black. In adjusted analyses, women sleeping <6 hours had a 3.2 (95% CI 1.4, 5.0) kg/m(2) greater BMI and 6.1 (1.8, 10.4) cm greater waist circumference compared to women sleeping 7-8 hours. In contrast, men had similar BMIs and waist circumferences regardless of sleep duration (p's for interaction = 0.04 & 0.11). There was no effect modification by race. CONCLUSIONS: In this sample, lower habitual sleep duration was associated with greater adiposity among women compared to men. Further research is needed to understand the potential mechanisms of the adverse metabolic effects of short sleep in women.