The Glycemia Risk Index (GRI) is a recently introduced continuous glucose monitoring (CGM)-derived metric intended to summarize glycemic risk. Prior evaluations of GRI have been limited in number and duration among adults with type 1 diabetes using Advanced Hybrid Closed-Loop (AHCL) systems. This single-center, observational, retrospective, real-world study aimed to further characterize the clinical utility of GRI and to investigate its relationships with established CGM-derived metrics, with particular focus on Time in Tight Range (TITR), over a 24-month follow-up period.
This investigation was conducted as a single-center, observational, retrospective analysis of adults with type 1 diabetes managed in routine clinical practice. The study design and reporting classify the work as a real-world assessment rather than a randomized or interventional trial. No additional demographic details, inclusion or exclusion criteria, or baseline characteristics beyond device allocation were reported in the source abstract.
Participants used commercially available AHCL platforms. Two device cohorts were specified: MiniMed 780G (n = 45) and Tandem t:slim X2 IQ technology (n = 20). GRI and other CGM-derived metrics were assessed from data collected during routine use of these AHCL systems across the 24-month observation period. The source abstract does not provide further technical details about data extraction, CGM models, or software versions.
The Glycemia Risk Index (GRI) is described as a CGM-derived metric intended to reflect glycemic risk; the abstract notes it is a recently introduced measure but does not provide the formula or computational method in the summary. Time in Tight Range (TITR) is referenced as a comparator CGM-derived metric; the abstract does not define the numeric bounds for TITR. Both GRI and TITR were central to analyses exploring associations with long-term glycemic outcomes.
Over the 24-month follow-up, GRI demonstrated progressive improvement, a trend the authors interpret as consistent with sustained glycemic control in the cohort of adults using AHCL systems. The abstract reports that improvements in GRI continued throughout the observation period, indicating a longitudinal pattern rather than a transient change. Specific numerical values, measures of central tendency, variability, or statistical test results were not provided in the abstract and therefore are not reported here.
Baseline GRI showed a positive correlation with glycated hemoglobin (HbA1c), indicating higher initial GRI values were associated with higher HbA1c. Baseline GRI was also inversely correlated with Time in Tight Range (TITR) at both 12 and 24 months, implying that a worse baseline GRI predicted lower TITR at follow-up time points. The abstract does not provide correlation coefficients, confidence intervals, or p-values; those statistical details were not reported in the source summary.
The authors conclude that GRI may serve as a useful, readily interpretable metric for predicting long-term glycemic outcomes in adults with type 1 diabetes using AHCL systems. The results support a complementary role for GRI alongside TITR in clinical assessment: while TITR measures the proportion of time glucose remains within a target range, GRI may add risk-focused information that correlates with conventional laboratory markers such as HbA1c and with subsequent TITR.
The study was conducted in a real-world clinical setting and included users of two commonly used AHCL platforms, which may enhance generalizability within similar outpatient adult populations. However, the abstract does not report subgroup analyses by device type or other stratified findings.
As summarized in the abstract, this work is a single-center, retrospective, observational study. The abstract does not provide granular methodological details such as sample selection criteria, handling of missing CGM data, statistical methods, or effect sizes. Those details were not reported in the source summary and therefore cannot be elaborated upon here.
Conflict of interest: the authors declared no competing interests in the source document.
Identifiers and publication: the report is indexed with PMID 42560625 and DOI 10.1007/s12020-026-04720-6. Keywords listed in the source include Automated insulin delivery, Continuous glucose monitoring, Glycemia Risk Index, Time in Tight Range, and Type 1 diabetes.