Glucose in Motion: Characterizing Temporal Patterns in Real-World Glycemic Control in a Large Observational Cohort

Abstract

OBJECTIVE To identify and characterize daily and weekly patterns in glycemic management in longitudinal real-world continuous glucose monitoring (CGM) data from people with diabetes to enhance clinical interpretation and support more effective personalized diabetes management. RESEARCH DESIGN AND METHODS This observational, retrospective study used two independent, multiyear, real-world CGM cohorts (T1DiabetesGranada and Connected Pen), comprising 86,779 participants with diabetes. Their mean age (SD) was 40.3 (15.8) and 44.0 (15.9), respectively, and there was a total of 21,011,021 measurement days. Weekly variations in CGM metrics and bolus insulin dose were examined using linear mixed-effects models, and daily temporal patterns were assessed. RESULTS Weekly and daily glycemic patterns emerged. Mid-week time in range was consistently higher (e.g., Wednesday: +0.96 percentage points [%-points]; 95% CI 0.93–0.99; P < 0.001), with deterioration during weekends and Mondays (e.g., Sunday: –1.74%-points; 95% CI –1.77, –1.72; P < 0.001), mirrored by shifts in other CGM metrics. Bolus insulin dose increased on weekend days (e.g., Sunday: +0.69 units; 95% CI 0.64–0.74; P < 0.001). Daily analyses revealed distinct diurnal profiles, with the most favorable glycemic control during morning and midday hours on weekdays and consistently poorer control during evening and nighttime periods, particularly on weekends. CONCLUSIONS Daily and weekly glycemic patterns revealed statistically robust and consistent temporal variations in glycemic control across two large real-world cohorts. Although the magnitude of these differences was small, the findings provide novel population-level evidence and underscore the value of longer-term CGM perspectives in routine care.