Time in Range and Incident Diabetic Retinopathy: Clinical and Economic Evidence From Real-World Type 1 Diabetes Care
Authors' conclusion
These findings can guide real-world CGM studies and support diabetes simulation models to predict future treatment outcomes from CGM metrics.
Study context
- Population
- Patients with T1D using continuous glucose monitoring (CGM) devices (mean duration 13.7 years), with long-term follow-up for incident diabetic retinopathy (DR)
- Participants
- 161 patients
- Exposure
- Per 10% increase in time in range (TIR)
- Casesn = 71
- Controlsn = 90reference
- Outcome
- Progression to DR by longitudinal changes in TIR (70–180 mg/dL)
Methodological detailsreal-world CGM and electronic health record data, multilevel mixed-effects parametric survival models, Markov models
- Real-world CGM and electronic health record data
- Multilevel mixed-effects parametric survival models
- Markov models
Main finding
As reported in the abstract
Retinopathy progression per 10% increase in TIR
HR per 10% increase in TIR
0.88
95% CI 0.78–0.98
Observational data: associations, not effects.
Further findings
Blindness prevented (TIR 70% vs 40%) · over time
- Estimated cases of blindness prevented per 1000 individuals
- 39
Economic simulation
- Lower costs per person
- $2581
- Quality-adjusted life years (QALYs) gained
- 0.13
Original publication
Time in Range and Incident Diabetic Retinopathy: Clinical and Economic Evidence From Real-World Type 1 Diabetes Care. J Clin Endocrinol Metab. Published 4 October 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →