Digital Biometrics in Predicting Risk for Obstructive Sleep Apnea and Hypertension: Decentralized, Prospective Cohort Study.

BACKGROUND

Sleep is an important component of human health and can be measured longitudinally using digital activity trackers. Further, decentralized digital research has the potential to provide a real-world picture of sleep in large populations.

OBJECTIVE

This study examined whether longitudinal sleep patterns from activity trackers could predict risk of obstructive sleep apnea (OSA) and hypertension, as defined by the Berlin questionnaire and self-report, respectively.

METHODS

We recruited adults aged ≥18 years nationwide to join our sleep-focused smartphone-based study, called the Research Framework for Exploring Sleep Health. Our sample of 391 adults predominantly comprised women (68%, 247/364) with a mean age of 48 (SD 13.62) years. Participants were asked to fill out health-related surveys, including the Berlin questionnaire and the Horne-Ostberg questionnaire for chronotype. Participants were asked to link their own activity tracker to the app to collect longitudinal sleep data.

RESULTS

We analyzed sleep data from 391 participants; the cohort was predominantly White (65%, 231/353) followed by multiracial (17%, 61/353) and Hispanic or Latino (6.5%, 23/353) participants. Collinearity testing showed that OSA risk and self-reported hypertension could be considered independently. Holding BMI at a fixed value, the odds of having high OSA risk increased by 159% for every 1-hour increase in weekday sleep variability (odds ratio [OR] 2.592, 95% CI 1.613-4.400; P<.001), and the odds of high OSA risk increased by 93% for each 1-hour increase in weekend sleep variability (OR 1.928, 95% CI 1.197-3.094; P=.01). The odds of having high OSA risk increased by 22% for each unit (kg/m2) increase in BMI, holding both weekday and weekend sleep at separate fixed values (OR 1.217, 95% CI 1.153-1.293; P<.001). Controlling for age, sex, and BMI, the odds of endorsing hypertension increased by 71% for every 1-hour increase in weekday sleep variability (OR 1.712, 95% CI 1.062-2.917; P=.03). Conversely, for weekend sleep, the odds of endorsing hypertension increased by 43% for a 1-hour increase in weekend sleep variability (OR 1.432, 95% CI 1.062-1.928; P=.04). Increased sleep variability predicted a high risk for both OSA and hypertension in this decentralized cohort, when using data from the Berlin questionnaire.

CONCLUSIONS

Our study demonstrates the utility of decentralized digital health studies in sleep research. It highlights the potential of activity trackers to predict risk for OSA and hypertension without requiring other patient information or assessment. Sleep variability is gaining increasing importance in the context of sleep health. Digital devices have the potential to help individuals assess their risk for certain disorders.