Assessing the Accuracy of Continuous Glucose Monitoring Metrics: The Role of Missing Data and Imputation Strategies
Authors' conclusion
The findings suggest that while missing data may have varying effects depending on the metric and imputation method, removing periods without data is a general acceptable approach.
Study context
- Population
- Individuals with T1D and T2D with 14-day continuous glucose monitoring (CGM) profiles933 14-day CGM profiles
- Participants
- 468 individuals
- Comparison
- Six imputation strategies vs reference CGM metrics
- Outcome
- Accuracy of CGM-derived metrics with missing data
Methodological detailsmodelled missing data patterns (missing completely at random, segmental, block-wise gaps), missing proportions 5% to 50%, six imputation strategies compared
- Modelled missing data patterns (missing completely at random, segmental, block-wise gaps)
- Missing proportions 5% to 50%
- Six imputation strategies compared
Main finding
As reported in the abstract
CGM data required to achieve high accuracy (R2 > 0.95)
at least 70%
Original publication
Assessing the Accuracy of Continuous Glucose Monitoring Metrics: The Role of Missing Data and Imputation Strategies. Diabetes Technol Ther. Published 14 May 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →