1005 Characterizing Sleep and Glycemia in Emerging Adults with Type 1 Diabetes

Abstract

Introduction: Type 1 Diabetes (T1D) affects 1.25 million Americans, and only 14% of emerging adults (ages 18-30 years) achieve targets for glycemic control (A1C < 7.0%). Sleep deficiency, less than 6.5 hours total sleep time (TST), is associated with poorer glycemic control. Methods: Emerging adults with T1D wore a wrist actigraph and their own or provided continuous glucose monitor (CGM) concurrently 24 hours/day for 6-8 days. Participants completed a 10-minute psychomotor vigilance test (PVT) on a device, 3-minute Trail Making Test on paper, and questionnaires including twice daily Pittsburgh sleep diaries in Research Electronic Data Capture (REDCap). TST, sleep onset latency (SOL), sleep efficiency (SE), wake after sleep onset (WASO), and sleep fragmentation index (SFI) were determined via actigraphy, glycemic control via A1C, and glucose variability via CGM. The purpose of this descriptive study was to explore associations between TST, sleep variability (SD of TST), neurocognitive function (psychomotor vigilance and executive function) and diabetes outcomes (glycemia and distress). Results: The sample included 36 emerging adults (mean age 22.8±3.1; 30.6% male; 91.7% White, 86.1% non-Hispanic; A1C mean 7.1±1.0%, BMI 27.3±4.8 kg/m2). Mean TST was 7.1±1.2 hours, SOL was 19.7±13.5 minutes, SE was 85.5±4.6%, WASO was 34.7±18.2 minutes, and SFI was 17.7±6.2. Shorter TST was associated with more severe sleepiness (r=-0.48,p=0.004) and more diabetes distress (r=-0.37, p=0.03). More sleep variability was associated with more severe sleepiness (r=0.36, p=0.03), longer response times (RT) ≥ 500ms (rho=0.39, p=0.02) measured via PVT, more nocturnal glucose variability (r=0.38, p=0.04), greater mean daily differences in glucose levels (r=0.42, p=0.02). Shorter mean RT was associated with more time in glucose range (rho=-0.37, p=0.04). Conclusion: Improving TST and sleep variability are potential therapeutic targets to improve glucoregulation in this high-risk population. Researchers should consider within-person multi-level modeling to inform our understanding of the true nature of the sleep-glucose association in emerging adults with T1D.