0869 Association Between Accelerometer-Measured Irregular Sleep Duration and Incidence of Obesity in Older Adults

Abstract Introduction Abnormal sleep duration is an established risk factor for obesity. However, the potential adverse effects of sleep's nightly variability, including fluctuations in duration, remain less explored. Irregular sleep duration may increase circadian disruption and subsequently contribute to metabolic alterations, behavioral and mood changes, and obesity development. Methods From 2013-2015, a subsample of the UK Biobank (UKB) participants wore accelerometers for a week. Our longitudinal study included 8,438 participants (mean age: 64 years) with ≥5 days of accelerometer data and repeated weight measurements who had BMI< 30 kg/m2 at baseline. Irregular sleep duration was assessed by the within-person standard deviation (SD) of 7-night accelerometer-measured sleep duration. Weight was objectively measured at UKB center visits or extracted from UK general practice records at two timepoints. The baseline measurement was taken in a period spanning from two years before to one year after the accelerometer study. The follow-up measurement was obtained on average 3.6 years after the baseline measurement (interquartile range: 1.8-5.0) and at least 6 months after the accelerometer study. Poisson regression with robust error variance was used to estimate relative risk (RR) for incident obesity (BMI≥30 kg/m2) according to sleep duration-SD categories. Results A total of 419 participants developed obesity at the follow-up measurement. Compared with participants with a sleep duration-SD < 30 minutes, the RR (95% CI) for incident obesity was 1.22 (0.86, 1.73) for 31-45 minutes, 1.60 (1.14, 2.26) for 46-60 minutes, and 2.01 (1.45, 2.80) for >60 minutes (p-trend=0.0024), after adjusting for age, sex, race, and follow-up period. Additional adjustment for lifestyle (smoking, physical activity, diet, and alcohol use), co-morbidities (dyslipidemia, hypertension, and depression), and other sleep-related factors including average sleep duration resulted in a weaker but statistically significant association (RR comparing >60 versus < 30 min: 1.73; 95% CI: 1.24, 2.41). The association was similar in men and women (P-interaction in the fully-adjusted model=0.84). Conclusion Irregular sleep duration was associated with higher risk of incident obesity in older adults, independent of average sleep duration and other obesogenic behaviors. Support (if any) R01HL155395