Key points are not available for this paper at this time.
Abstract Introduction Epidemiological evidence indicates that short, long and variable sleep duration are associated with cardiometabolic outcomes. However, due to the ubiquitous role of sleep, one would expect to see effects on a broad range of human diseases. Using electronic health records (EHR) and wearable sleep data from the All of Us Research Program, we conducted a cross-sectional analysis of sleep duration and variability with the risk of over 3530 chronic conditions across the human phenome. Methods All of Us participants (≥18 years) with available EHR data and at least 90 days of continuous monitoring using Fitbit devices under a ‘bring your own device’ model were included. Average (mean) and variability (standard deviation) of sleep duration were derived based on device-generated data. Chronic conditions were determined based on mapping of International Classification of Disease 10 codes available in the EHR to phecodes, an ontology representing clinically meaningful concepts. Associations between sleep traits and phecodes were assessed using multiple test corrected logistic regression adjusted for age, sex, and race. Results A total of 8,256 participants with available Fitbit data were included. Participants were 89% White, 72% female, 72% college educated and had median (IQR) age of 51 (36.6-52.0) years, BMI 28.3 (24.4-33.0) kg/m2, and 1,566 (860-2307) days with sleep measurements. The median average nightly sleep duration was 6.34 (5.67-6.91) hours and the median nightly sleep variability was 1.63 (1.33-1.95) hours. In adjusted analyses, participants with greater sleep duration had significantly lower odds of obesity (OR; 95%CI = 0.85; 0.80-0.91, obstructive sleep apnea (0.74; 0.74-0.88), type 2 diabetes (0.68, CI 0.58-0.80), and essential hypertension (0.86; 0.81-0.92). Conversely, participants with greater sleep duration variability had higher odds of major depressive disorder (2.15; 1.84-2.51), anxiety disorder (1.75; 1.51-2.03), gastroesophageal reflux disease (1.61; 1.39-1.86), and abnormal electrocardiogram (1.80; 1.47-2.19). Conclusion Our findings suggest important associations between sleep duration patterns and chronic conditions, emphasizing the importance of sleep health in disease prevention and management. Important limitations include analysis of a selected population with access to wearable devices, which may not represent the general population enrolled in All of Us. Support (if any) All of Us Research Scholars Program
Manetta et al. (Sat,) studied this question.