Glycemic Outcomes and Feature Set Engagement Among Real-Time Continuous Glucose Monitoring Users With Type 1 or Non-Insulin-Treated Type 2 Diabetes: Retrospective Analysis of Real-World Data

Background: The benefits of real-time continuous glucose monitoring (RT-CGM) are well established for patients with type 1 diabetes (T1D) and patients with insulin-treatedtype 2 diabetes (T2D). However, the usage and effectiveness of RT-CGM in the context of non-insulin-treated T2D has not been well studied. Objective: We aimed to assess glycemic metrics and rates of RT-CGM feature utilization in users with T1D and non-insulin-treated T2D. Methods: We retrospectively analyzed data from 33,685 US-based users of an RT-CGM system (Dexcom G6; Dexcom, Inc) who self-identified as having eitherT1D (n=26,706) orT2D and not using insulin (n=6979). Data included glucose concentrations, alarm settings, feature usage, and event logs. Results: The T1D cohort had lower proportions of glucose values in the 70 mg/dl to 180 mg/dl range than the T2D cohort (52.1% vs 70.8%, respectively), with morevalues indicating hypoglycemia or hyperglycemia and higher glycemic variability. Discretionary alarms were enabled by a large majority in both cohorts. The data sharing feature was used by 38.7% (10,327/26,706) of those with T1D and 10.4% (727/6979) of those with T2D, and the mean number of followers was higher in the T1D cohort. Large proportions of patients with T1D or T2D enabled and customized their glucose alerts. Retrospective analysis features were used by the majority in both cohorts (T1D: 15,783/26,706, 59.1%; T2D: 3751/6979, 53.8%). Conclusions:Similar to patients with T1D, patients with non-insulin-treated T2D used RT-CGM system features, suggesting beneficial, routine engagement with data by patients and others involved in their care. Motivated patients with diabetes could benefit from RT-CGM coverage.