In one sentence
The older way of turning a mean glucose into an HbA1c-like figure, and the same line run the other way to turn a laboratory HbA1c into an average glucose; shown beside the GMI, and the figure VT records each day as your sensor's HbA1c estimate.
Where it comes from
The A1c-Derived Average Glucose study followed 507 people, 268 with type 1 diabetes, 159 with type 2 and 80 without diabetes, in ten centers for three months, with sensor wear and seven-point finger-prick profiles giving about 2,700 glucose values each, and fitted a line between the laboratory HbA1c at the end and the average glucose across the period: average glucose (mg/dL) = 28.7 × A1c (%) − 46.7, with an R² of 0.84, and the line did not differ by age, sex, type of diabetes, ethnicity or smoking (Nathan 2008). Laboratories still print the "estimated average glucose" beside an HbA1c result using this line.
Why GMI replaced it as a sensor figure
The ADAG line was built partly on finger-prick profiles and on the sensors of 2008; the GMI line was fitted to modern sensors and renamed so that nobody mistakes an estimate for a laboratory result (Bergenstal 2018). The two lines are close but not identical, so a sensor app that still shows eA1c will differ slightly from VT's GMI for the same month.
How Vitality Tracker uses it
From the mean of your last 90 days, the estimated A1c is shown beside the GMI on the Live Dashboard, labeled as the formula Libre and LibreView use; at the averages most members see, it reads lower than the GMI for the same months. It is also the figure VT records each day as your sensor's HbA1c estimate, once the 90 days hold about two weeks' worth of readings, and the one the Vitality Score's HbA1c slot uses when it is more recent than your laboratory result.
How far to trust the number
Calculated from a population line with an R² of 0.84, which means the line explains most but not all of the person-to-person difference; the same cautions as for the GMI apply.
Sources
- Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ; A1c-Derived Average Glucose Study Group. Translating the A1C assay into estimated average glucose values. Diabetes Care 2008;31:1473–8. doi:10.2337/dc08-0545
- Bergenstal RM, et al. Diabetes Care 2018;41:2275–80. doi:10.2337/dc18-1581.