Abstract 015: Sleep Health and Cardiometabolic Risk in Health Care Workers: The role of heart rate and heart rate variability

Purpose: Despite strong evidence linking poor sleep with cardiometabolic risk, underlying mechanisms are poorly understood, limiting interventions to improve sleep&cardiometabolic health. Higher heart rate (HR) and lower HR variability (HRV) are associated with worse cardiometabolic health and may be affected by sleep. We determined associations between multidimensional sleep health, HR, and HRV among healthcare workers (HCW) from federally qualified health centers (FQHCs) in Louisiana. Methods: Eighty HCW wore wrist-worn sensors capturing sleep onset, wake, HR, and HRV for 6 weeks. Four dimensions of sleep health were measured: regularity was assessed as the standard deviation (SD) of midpoints of all nighttime sleep periods for a participant; “irregular”=SD >60 minutes. Timing was assessed using participants’ mean sleep midpoint; “poor timing”=midpoint not between 1am&3am. Efficiency was calculated as % of time in bed spent asleep; “inefficient”=mean efficiency across nights <85%. Duration was calculated as time between first sleep onset after 8pm&last wake, minus time spent awake during the night; “insufficient”=mean duration <7 hours. Dichotomized regularity, timing, efficiency&duration were summed; multidimensional “poor sleep”=score ≥2. HR (beats/minute) and HRV (root mean square of successive differences between beats) were measured every 5 minutes. Multivariable linear regression estimated differences in HR and HRV across each sleep metric, adjusting for age and mental health/sleep medications. Results: 74 participants recorded sleep data: 90% women, 41% Black, mean age 45.6 (SD 11.5) years; 36% were taking mental health/sleep medications. Sixty-eight had ≥7 nights of sleep data&complete medication data. Among those, inefficient sleep and insufficient sleep duration were associated with 6.6 (95% CI 0.6, 12.7) and 4.0 (95% CI 0.0, 7.9) higher HR in the 2 hours before wake, respectively. “Poor sleep” was associated with 4.6 (95% CI 0.8, 8.5) higher HR. Irregular sleep and poor timing were not associated with HR. No sleep metrics were associated with HRV in the period spanning the hour before&hour after wake. Conclusion: Louisiana FQHC HCWs with inefficient sleep and insufficient sleep duration had higher mean HR in the 2 hours before wake. Sleep was not associated with HRV, which may be less sensitive to short term change. Further research is needed in larger samples to determine mechanisms linking poor sleep to cardiometabolic risk.