0140 Operationalizing Sleep for the Military: Advancing an Infrastructure for Fatigue Risk Management with Wearable Technology

Abstract

Abstract Introduction A recent investigation on fatigue within United States (US) Naval Surface Force by the Government Accountability Office recommended that the US Navy incorporate methods to empirically monitor and manage fatigue-related threats. In response, a feasibility effort was undertaken to assess two commercial wearable devices previously evaluated against polysomnography by our group for potential use within a novel fatigue monitoring system onboard US warships. Methods 501 Sailors from three warships (415 men, 85 women, 1 response missing; 29.1±7.4 years, mean±SD) were included in analyses. During multi-day missions at sea, participants were instructed to continuously wear a wrist-worn Fatigue Science Readiband and an Oura Ring. Participants met with the research team daily to sync wearable devices and complete self-ratings of sleep and fatigue on a tablet. While returning to port, participants completed a user experience questionnaire that included 5-point scales for device comfort and interference with daily activities (larger values indicate greater comfort or interference). Results There were no overall differences in total sleep time (TST) and sleep efficiency (SE) between devices (TST: Ring=352.23±87.1, Readiband=353.89±97.79; p=0.45, d=-0.01; SE: Ring=84.59±7.08, Readiband=84.20±10.24; p=0.06, d=0.04). There were 9 days for TST and 4 days for SE (of 66 total days) that resulted in effect sizes >0.50 between devices (for >0.20: TST, 39/66; SE, 29/66). Preliminary analyses between TST and self-reported exhaustion and inability to function suggest predictive relationships with the wearables (Ring: r=-0.20; Readiband: r=-0.21), with no relationship with SE across devices (both r<0.01). Average comfort ratings for Ring (3.74) exceeded Readiband (3.59), but this difference was small (p=0.04, d=0.12). Interference with daily activities was also similar between devices (Ring=2.10, Readiband=2.20; p=0.14, d=0.08). Conclusion Both commercial wearables performed similarly for their TST and SE assessments, and TST was associated with self-rated fatigue within this operational environment. While the ring form factor was selected more for comfort than the wrist form factor, the difference was small. Collectively, findings indicate that both devices show promise for operational use within fatigue monitoring systems. Further testing is being conducted to better understand the strengths and limitations of wearable devices for monitoring sleep in warship environments. Support (If Any) Military Operational Medicine Research Program