Association between continuous glucose monitoring metrics and clinical outcomes in adults with type 1 diabetes in a real-world setting

Highlights•Consensus statements have set targets for CGM metrics, supported by RCT data.•We analyzed real-world CGM and clinical data to assess how these targets perform.•For every 10% increase in TIR, HbA1c decreased by 0.34%, smaller than in RCTs.•Relationships between HbA1c, glycemic variability, and hypoglycemia were complex.•These relationships may differ in clinical practice compared with controlled trials.AbstractAimsContinuous glucose monitoring (CGM) metrics can assist diabetes management. Consensus statements recommend > 70 % time in range (TIR) and ≤ 36 % glucose coefficient of variation (CV). However, how these targets perform in clinical practice is unknown. This retrospective, longitudinal cohort study analyzed relationships between TIR, CV, glycated hemoglobin (HbA1c), and hypoglycemia in a real-world setting.MethodsData of 542 adults with type 1 diabetes who used CGM (January 2014–July 2020) were analyzed. Associations between TIR and HbA1c at the same and subsequent visits, incidence rate ratios (IRRs) for hypoglycemia at different CVs, and number of hypoglycemic events at cross-sections of HbA1c and CV were estimated by regression.ResultsTIR was inversely related to HbA1c; for every 10 % increase in TIR, HbA1c was significantly reduced by 0.34 % (4 mmol/mol) and 0.20 % (2 mmol/mol) at the same and subsequent visits, respectively. Level 2 hypoglycemia was significantly reduced at CV < 30 %, 30–33 %, 33.1–36 %, and 36.1–40 %: adjusted IRRs vs CV ≥ 40.1 % of 0.14, 0.28, 0.32, and 0.50, respectively. Hypoglycemic events were reduced at lower CV across HbA1c levels and at higher HbA1c across CV levels.ConclusionThis study quantifies HbA1c improvements with increased TIR and hypoglycemia reductions with improved CV in clinical practice.