986-P: Identifying Patterns of Glycemic Control from Wearable Device Data

Introduction & Objective: To identify temporal patterns of glucose control measured by continuous glucose monitoring (CGM) devices for type 1 and type 2 diabetes (T1D and T2D) patients in Veterans Affairs Healthcare System. Methods: Observational study using Dexcom CGM device data linked with patients' electronic health records from 2015-2020 to identify temporal and regional patterns of glycemic control. Results: Using a cohort of 785 T1D and 1,466 T2D patients and their 1,551,644 CGM readings over nearly 3-years, we found that glycemic control deteriorated significantly during weekends. Average glucose differed between Sundays (highest) and Wednesdays (lowest) by 2.7mg/dL (Figure). Additionally, time-in-range (between 70 mg/dL and 180 mg/dL) and time-above-range (time above 180 mg/dL) showed similar temporal patterns. We also observed seasonal variations in glycemic trends, with patients having worse glycemic control from October to February and better control from April to August. The average glucose difference between April (lowest) and December (highest) was 4mg/dL and varied regionally, with these patterns accentuated in the Northeast compared to the Southwest. These patterns were generally similar in both T1D and T2D patients. Conclusions: Identifying and understanding these real-world glucose fluctuations may help patients and providers better manage overall glucose control.