Differences Between Glycated Hemoglobin and Glucose Management Indicator in Real-Time and Intermittent Scanning Continuous Glucose Monitoring in Adults With Type 1 Diabetes
Authors' conclusion
The GMI is overestimated in G6 users, particularly those with well-controlled diabetes, but the GMI and HbA1c discordance improved with a revised equation from the observed data. The FL profile showed greater discordance for lower HbA1c levels or higher HbA1c levels.
Study context
- Population
- Adults with T1D using continuous glucose monitoring (CGM) sensors over 90 days; 682 G6 and 515 FL glycemic profiles coinciding with HbA1c682 + 515 profiles
- Participants
- 275 adults
- Exposure
- Glucose Management Indicator (GMI) vs HbA1c with Dexcom G6 and FreeStyle Libre 1
- G6n = 117
- FLn = 158
- Outcome
- Difference between GMI and HbA1c
- Follow-up
- 90 days
Main finding
As reported in the abstract
Mean difference
0.58 %
P < 0.001
G6 profiles; GMI higher than HbA1c irrespective of mean glucose; ± 0.49%
Mean difference
0
P = 0.966
FL profiles, overall
Observational data: associations, not effects.
Further findings
Mean HbA1c
GMI–HbA1c discordance by HbA1c level, G6
- Highest discordance, HbA1c 5.0%-5.9%
- 0.9%
- GMI significantly higher than HbA1c when HbA1c below 8.0% (all P < .001)
Proportion of discordance >0.5%, G6, original vs revised GMI equation
- Improved with revised GMI equation
- 60.1% to 30.9%
GMI–HbA1c discordance by HbA1c level, FL
- GMI lower than HbA1c of 9.0% to 9.9% (all P < .001)
- 0.9%
- GMI higher in HbA1c of 5.0% to 5.9% (all P < .001)
- 0.5%
Original publication
Composed by AI from the published abstract and checked against it. How Briefs are made →