Accuracy, Utilization, and Effectiveness Comparisons of Different Continuous Glucose Monitoring Systems

Abstract

Background: Accuracy and feature sets of continuous glucose monitoring (CGM) systems may influence device utilization and outcomes. We compared clinical trial accuracy and real-world utilization and effectiveness of two different CGM systems. Materials and Methods: Separately conducted accuracy studies of a fifth-generation and a sixth-generation CGM system involved 50 and 159 adults, respectively. For between-system performance comparisons, propensity score methods were utilized to balance cohort characteristics. Real-world outcomes were assessed in 10,000 anonymized patients who had switched from the fifth-generation to the sixth-generation system and had used connected mobile devices to upload data from both systems, allowing pairwise comparisons of device utilization and glucose concentration distributions. Results: Propensity score-adjusted mean absolute relative differences for the fifth-and sixth-generation systems were 9.0% and 9.9%, and the percentages of values within-20%/20 mg/dL were 93.1% and 92.5%, respectively. The sixth-generation system, but not the fifth-generation system, met accuracy criteria for interoperable CGM systems. Both systems had high real-world utilization rates (93.8% and 95.3% in the fifth-and sixthgeneration systems, respectively). Use of the sixth-generation system was associated with fewer glucose values <55 mg/dL (<3.1 mmol/L) (0.7% vs. 1.1%, P < 0.001) and more values 70-180 mg/dL (3.9-10.0 mmol/L) (57.3% vs. 56.0%, P < 0.001) than the fifth-generation system. Conclusions: CGM performance outcomes can be compared through the propensity score analysis of clinical trial data and pairwise comparisons of real-world data. The systems compared here had nearly equivalent accuracy and utilization rates. Longer term biochemical and psychosocial benefits observed with the fifthgeneration system are also expected with the sixth-generation system.

The Association Between HbA1c and Time in Hypoglycemia During CGM and Self-Monitoring of Blood Glucose in People With Type 1 Diabetes and Multiple Daily Insulin Injections: A Randomized Clinical Trial (GOLD-4)

“…Another limitation is that since this study was performed, CGM technology has improved (e.g., no calibrations are needed with the Dexcom G6). Although direct comparisons do not exist of Dexcom G4 and G6 of accuracy, studies of individual systems indicate an increasing accuracy for Dexcom G6 (15). Data on frequency of calibrations, calibration errors, and transmitter issues were not reported during this study.…”

Glycemic Improvement Using Continuous Glucose Monitoring by Baseline Time in Range: Subgroup Analyses from the DIAMOND Type 1 Diabetes Study

“…This study used an older Dexcom G4 Platinum CGM, but we expect the current generation (Dexcom G6) to obtain the same or greater benefits with respect to increasing TIR and reducing hypoglycemia. 11…”

Real-World Hypoglycemia Avoidance with a Continuous Glucose Monitoring System's Predictive Low Glucose Alert

“…Earlier in silico studies support this hypothesis. 18,19 We examined whether a predictive low glucose alert could provide additional advance warning to rtCGM users before the onset of clinical hypoglycemia by evaluating estimated glucose values (EGVs) from an anonymized convenience sample of 1424 patients before and after their transition from an earlier fifth-generation (G5) system with no predictive alert to a comparably accurate 20,21 sixth-generation (G6) system with predictive ''Urgent Low Soon'' (ULS) alert. The ULS alert is enabled by default and triggered when an EGV £55 mg/dL is predicted in the next 20 minutes; the activation of a ULS alert inhibits low threshold alert activations in the following 30 minutes to limit alarm fatigue.…”