Imputation of Missing Continuous Glucose Monitor Data
Authors' conclusion
Missing CGM data are to be expected. For periods of wear with up to 20% missing data, hot-deck imputation and complete case analysis may be acceptable if data are missing completely at random. Explored imputation techniques are robust, but each has their own limitations, which should be considered if these techniques are implemented.
Study context
- Population
- 105 CGM profiles with nine days of complete glucose measurements from the GO MOMs study; missingness introduced at 2%, 5%, 10% and 20%
- Participants
- 105 CGM profiles
- Comparison
- Imputation techniques vs complete case analysis for missing CGM data
- Primary endpoint
- Mean relative bias (MRB) between true and imputed CGM data for CGM summary metrics
Methodological detailssecondary use of GO MOMs study data, simulated missingness (zero-inflated negative binomial hurdle model)
- Secondary use of GO MOMs study data
- Simulated missingness (zero-inflated negative binomial hurdle model)
Primary result
As reported in the abstract
MRB of CGM summary metrics
- Mean relative bias (MRB) minimal across most metrics and imputation methods at 2% missing data; increases with higher missing data frequency, trends depending on metric and method
- Hot-deck imputation and complete case analysis show consistently low MRB
Original publication
Imputation of Missing Continuous Glucose Monitor Data. J Diabetes Sci Technol. Published 31 December 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →