Habitual sleep duration and sleep duration variation are independently associated with body mass index.

Abstract

Sleep plays a vital role in maintaining homeostasis and promoting health. Previous studies show that shorter sleep duration is associated with elevated body mass index (BMI) and other cardiovascular risk factors. The goal of this study was to investigate the effects of habitual sleep duration and nightly sleep duration variation based on daily device-recorded data on BMI and obesity-related biomarkers. In all, 748 individuals (50.6% females, 85.4% European-Americans, average age: 49.7 years old) participated in a commercial lifestyle coaching program beginning in July 2015. Daily sleep data were recorded by Fitbit Charge HR wristbands. Clinical laboratory blood tests were measured up to three times over a 12-month period. Linear regression models were used for cross-sectional analyses, and generalized estimating equations for longitudinal analyses. All models were adjusted for age, sex, geographic location, season, genetic ancestry inferred from whole genome sequencing data, and BMI (if applicable). Multiple testing issues were corrected by false discovery rate. We calculated habitual sleep duration and nightly sleep duration variation. In general, females slept 15-min longer on average than males. A negative correlation was found between habitual sleep duration and BMI (β=-1.12, standard error=0.25, P<0.001). Moreover, we identified a positive correlation between sleep duration variation and BMI (β=2.97, standard error=0.79, P<0.001) while controlling for sleep duration, indicating that larger sleep duration variation is significantly and independently associated with increased BMI. We explored the impact of habitual sleep duration and sleep duration variation, and identified that shorter habitual sleep duration and larger duration variation were independently associated with increased BMI.