Advancing Continuous Glucose Monitoring for Inpatient Clinical Decision Support: Individual Algorithmic Mean Absolute Relative Difference
Authors' conclusion
Patient-specific algorithmic refinements of CGM data demonstrate the potential to adequately address the unique demands of inpatient diabetes care by reducing intrapersonal MARD, paving the way for the adoption of CGM systems into hospital environments.
Study context
- Population
- Inpatients with diabetes using continuous glucose monitoring (CGM)
- Participants
- 226 vs 24
- Comparison
- Patient-specific optimization algorithm vs unadjusted CGM data
- Initialn = 226
- Secondn = 24
- Outcome
- CGM accuracy vs point-of-care glucose: MARD, Clarke Error Grid analysis and modified U.S. FDA agreement rule
Methodological detailsdevelopment cohort and second validation cohort
- Development cohort and second validation cohort
Performance
As reported in the abstract
CGM accuracy
Initial mean absolute relative difference (MARD)
10.30%
Data points in zones A and B of the Clarke Error Grid
99.02%
MARD improvement with patient-specific optimization algorithm
4.33%
Further findings
Intrapersonal MARD with patient-specific algorithm in second inpatient cohort
- Reduction in intrapersonal MARD
- 5.58%
Original publication
Advancing Continuous Glucose Monitoring for Inpatient Clinical Decision Support: Individual Algorithmic Mean Absolute Relative Difference. Diabetes Care. Published 3 November 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →