Japan's largest platform for academic e-journals: J-STAGE is a full text database for reviewed academic papers published by Japanese societies

Background & Aims: Previous cross-sectional studies revealed that short sleep duration has a close relationship with the presence of non-alcoholic fatty liver disease (NAFLD). We aimed to investigate the association between sleep duration and incident NAFLD.Methods: In this historical cohort study of 12,306 participants (5,848 men and 6,458 women), we investigated the effect of sleep duration on incident NAFLD. NAFLD was defined as having fatty liver diagnosed by abdominal ultrasonography in the participants who consumed ethanol less than 30 g/day for men and 20 g/day for women. We divided the participants into four groups according to sleep duration: >7, >6-7, >5-6, and ≤5h. Cox proportional hazards models were performed to investigate the effect of sleep duration on incident NAFLD, adjusting for age, body mass index categories, alanine aminotransferase, triglycerides, high density lipoprotein-cholesterol, fasting plasma glucose, smoking status, alcohol consumption, systolic blood pressure, exercise.Results: During the median 6.8-year follow-up for men and the 7.0-year follow-up duration for women, 2,280 participants (1,581 men and 699 women) developed NAFLD. In Cox proportional hazards models, sleep duration of ≤5 h in both men and women were revealed to be a significant risk for incident NAFLD, compared to men and women with a sleep duration of >7 h (men: hazard ratio 1.39, 95% confidence interval 1.13-1.72, p=0.002; women; 1.46, 1.05-2.04, p=0.023).Conclusion: This is the first study showing that short sleep duration was a risk factor for incident NAFLD.