Abstract Introduction Health disparities are recognized in sleep medicine, but longitudinal observations are limited. Digital tools provide scalable methods to assess risk. SLEEP by Cleveland Clinic is a mobile application using evidence-based tools to screen adults for common sleep disorders. We analyzed user-reported data to evaluate risks of OSA, insomnia, and abnormal sleep duration by sex and age. Methods Data from October 2019 through November 2025 were included with 52% of downloads occurring in 2025. Age was categorized into: 18-39, 40-49, 50-59, 60-69, ≥70. High risk OSA (HR-OSA; AHI 15), was assessed by the Cleveland Clinic Sleep Apnea Probability Score (Katzan IL, 2016) and insomnia symptom severity by the Insomnia Severity Index (ISI). Sleep duration was classified as normal (7-9h) or abnormal ( 7h and 9h). Risk changes was evaluated with linear and logistic regression models incorporating sex-specific variances and a time-by-gender interaction (time=month of download). Results The sample was comprised of 11,944 users (mean age 58(16)y, 67.6% female, 86% White). Females with older (59(16) vs 57(16)y, p=0,001) and more likely to be White (89 vs 81%, p 0.001). Median (IQR) HR-OSA probability was lower in females (25(10,42) vs 49 (21,71)%, p 0.001), consistently increasing with age: 18-39y (3 vs. 10), 40-49y (9 vs 23), 50-59y (18 vs 47), 60-69y (31 vs 63), ≥70y (48 vs 79), p 0.001. Median (IQR) ISI was higher in females (16(12,20) vs 14(101, 9), p 0.001), and across all age groups except 60-69y: 18-39y (16 vs 14), 40-49y (15 vs 14), 50-59 (16 vs 14), ≥70y (16 vs 15), suggesting stable, prolonged symptoms in both groups (p 0.001). Abnormal sleep duration was greater in females overall (39.1 vs 32%, p 0.001) and across all age groups except 18-39y and 60-69y: 40-49y (34 vs 27%), 50-59y (36 vs 28%), ≥70y (44 vs 39%), p 0.001. Short and long sleep was observed in 9.1 and 30% of females and 12 and 22% of males, respectively. Conclusion These results highlight sex- and age-specific differences in sleep disorder risk and demonstrate the potential of large-scale digital health data to elucidate population-based sleep health disparities that may increase awareness and improve early detection and treatment. Support (if any)
Ali et al. (Fri,) studied this question.