Evidence-Based Targets for Continuous Glucose Monitoring Metrics in Adults with Type 1 Diabetes

About This Grant

Abstract: Robust evidence supports the use of continuous glucose monitoring (CGM) to assess glycemic control and inform clinical decisions in adults with type 1 diabetes (T1D). The voluminous raw CGM data (glucose reading every few minutes) is summarized into 8 standard CGM metrics (e.g., time in range [TIR] defined as percent time with glucose readings between 70–180 mg/dl). However, recommended targets for each CGM metric (e.g., >70% TIR) are based on expert opinion and have not been evaluated empirically. To establish evidence-based goals of care in adults with T1D, our research will identify optimal CGM metric targets that minimize diabetes complications using causal machine learning models. The proposed work will leverage data from a contemporary (2016–2025) cohort of ~4000 adults with T1D from Kaiser Permanente Northern California (KPNC). Consistent with current T1D care guidelines, all analyses will be stratified into an older/high-risk group for whom clinical decisions typically prioritize short term safety (i.e., avoidance of hypoglycemia and diabetic ketoacidosis) and the remaining younger/healthier group for whom clinical decisions prioritize avoidance of acute metabolic events as well as long term microvascular and cardiovascular complications. In Aim 1, for each of the 8 standard CGM metrics included in the ambulatory glucose profile, we will identify optimal targets that are associated with the lowest risk of each of 4 complication categories (hypoglycemia, diabetic ketoacidosis, microvascular, and cardiovascular events) separately. To identify opportunities for quality improvement, and inform future guidelines and translational research, in Aim 2 we will assess the proportion of patients achieving the novel optimized CGM metric targets (calculated in Aim 1) as well as CGM metric targets recommended in the 2026 American Diabetes Association Standards of Care in Diabetes. Identifying CGM metric targets based on real-world evidence rather than expert opinion will lead to improved patient-centered care and quality of life and for adults with T1D.