Compare
GMI vs A1c
Your CGM report shows a GMI. Your lab sends an A1c. They look alike, they are not the same measurement, and they often disagree. Here is how they relate.
The short answer
- A1c is a blood test. It measures how much glucose is attached to hemoglobin, which reflects about 3 months (NIDDK). It is the number used to diagnose diabetes (ADA Standards 2026, sec. 2).
- GMI (Glucose Management Indicator) is calculated from the mean glucose of your CGM data. It estimates what an A1c would be from sensor readings alone (Bergenstal 2018).
- Neither is "the true one". They answer different questions, and a gap between them is common (Bergenstal 2018).
Try it with your mean glucose
In mmol/L? Multiply by about 18, or use the GMI calculator, which takes both units.
GMI vs A1c chart
Same mean glucose, two formulas. GMI comes from Bergenstal 2018 (Bergenstal 2018). The ADAG column reverses the formula labs use to turn an A1c into estimated average glucose (Nathan 2008). The two were fitted to different study groups, so they drift apart as glucose rises.
| Mean mg/dL | mmol/L | GMI % | GMI mmol/mol | ADAG A1c % |
|---|---|---|---|---|
| 100 | 5.6 | 5.7 | 39 | 5.1 |
| 110 | 6.1 | 5.9 | 41 | 5.5 |
| 120 | 6.7 | 6.2 | 44 | 5.8 |
| 130 | 7.2 | 6.4 | 47 | 6.2 |
| 140 | 7.8 | 6.7 | 49 | 6.5 |
| 150 | 8.3 | 6.9 | 52 | 6.9 |
| 160 | 8.9 | 7.1 | 54 | 7.2 |
| 170 | 9.4 | 7.4 | 57 | 7.6 |
| 180 | 10.0 | 7.6 | 60 | 7.9 |
| 190 | 10.5 | 7.9 | 62 | 8.2 |
| 200 | 11.1 | 8.1 | 65 | 8.6 |
| 210 | 11.7 | 8.3 | 68 | 8.9 |
| 220 | 12.2 | 8.6 | 70 | 9.3 |
| 230 | 12.8 | 8.8 | 73 | 9.6 |
| 240 | 13.3 | 9.1 | 75 | 10.0 |
| 250 | 13.9 | 9.3 | 78 | 10.3 |
| 260 | 14.4 | 9.5 | 81 | 10.7 |
| 270 | 15.0 | 9.8 | 83 | 11.0 |
| 280 | 15.5 | 10.0 | 86 | 11.4 |
| 290 | 16.1 | 10.2 | 88 | 11.7 |
| 300 | 16.7 | 10.5 | 91 | 12.1 |
GMI % = 3.31 + 0.02392 × mean glucose (mg/dL). ADAG A1c % = (mean glucose + 46.7) ÷ 28.7. Around 150 mg/dL the two agree; at 250 mg/dL GMI reads about 9.3% while ADAG gives about 10.3%.
Why your GMI and lab A1c differ
- They measure different things. A1c depends on red blood cells and how much glucose attaches to them. GMI depends only on sensor glucose (Bergenstal 2018).
- Red blood cell changes move A1c, not GMI. Blood loss, transfusion, iron-deficiency anemia, kidney failure, liver disease and some hemoglobin variants can make an A1c falsely high or low (NIDDK). See when A1c can mislead.
- Different time windows. A1c weighs the last 30 days most, but reflects about 3 months (NIDDK). GMI covers only the days in your CGM report; its authors advise at least 14 days of data (Bergenstal 2018).
- Test variation. Repeat A1c tests on the same sample can differ a little (NIDDK).
Which one to use
For diagnosis, the A1c is the standard test; GMI is not a diagnostic test (ADA Standards 2026, sec. 2). For day-to-day management, GMI shows what your recent CGM data adds up to. If the two differ by a lot, that gap is worth showing your care team, because it can point to something that affects the A1c (NIDDK).
So after a change, GMI can reflect it as soon as your CGM report covers the new weeks, while A1c takes roughly 3 months to fully catch up (Tahara 1995). See how fast A1c can change.
Sources
- Bergenstal RM, Beck RW, Close KL, et al. Glucose Management Indicator (GMI): a new term for estimating A1C from continuous glucose monitoring. Diabetes Care. 2018;41(11):2275-2280. doi.org/10.2337/dc18-1581
- Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ; A1c-Derived Average Glucose (ADAG) Study Group. Translating the A1C assay into estimated average glucose values. Diabetes Care. 2008;31(8):1473-1478. doi.org/10.2337/dc08-0545
- National Institute of Diabetes and Digestive and Kidney Diseases (NIH). The A1C Test & Diabetes. www.niddk.nih.gov/health-information/diagnostic-tests/a1c-test
- American Diabetes Association Professional Practice Committee. 2. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes 2026. Diabetes Care. 2026;49(Suppl 1). doi.org/10.2337/dc26-S002
- Tahara Y, Shima K. Kinetics of HbA1c, glycated albumin, and fructosamine and analysis of their weight functions against preceding plasma glucose level. Diabetes Care. 1995;18(4):440-447. doi.org/10.2337/diacare.18.4.440
Written and maintained by the controla1clevels.com editorial team. We do not have a medical reviewer; every claim links to its primary source above. Last reviewed: 2026-10-26. Spotted an error? See About.