Glucose in the blood attaches to hemoglobin, the part of red blood cells that carries oxygen, and the higher the glucose, the more of it attaches. The test measures that share and reports it as a percentage. NIDDK puts a normal A1C below 5.7%.1
The ranges
- Normal: below 5.7%.
- Prediabetes: 5.7 to 6.4%. Within this range, the higher the A1C, the greater the risk of type 2 diabetes.
- Diabetes: 6.5% or above.
No fasting is needed, so the blood can be drawn at any time of day. A result in the diabetes or prediabetes range without symptoms should be confirmed by a repeat test on a different day. For diagnosis, NIDDK says the sample should come from a vein and go to a lab using an NGSP-certified method; quick tests done in a doctor's office should not be used. The A1C test is not used to diagnose type 1 diabetes, gestational diabetes or diabetes related to cystic fibrosis.1
From A1C to average glucose
A 2008 study, known as ADAG, worked out the conversion. Its 507 participants, 268 with type 1 diabetes, 159 with type 2 and 80 without diabetes, wore continuous glucose monitors for at least 2 days four times and checked their own blood seven times a day on other days. That gave about 2,700 glucose values each over 3 months, and the average fit the formula: average glucose in mg/dL equals 28.7 times the A1C minus 46.7. The formula held across age, sex, diabetes type, race and ethnicity, and smoking status.2 By it, an A1C of 5.7% matches an average of about 117 mg/dL, 6.5% about 140, 7% about 154 and 8% about 183.
Some labs report this estimated average glucose, or eAG, with the A1C. It will not match readings on a home meter, because it is a long-term average rather than the glucose at a single moment. Large changes over the past month do show up in the A1C, but sudden, temporary rises and falls do not.1
Goals and how often to test
NIDDK says studies have shown that some people with diabetes can lower the risk of complications by keeping A1C below 7%. That goal is not safe for everyone: if it leads to low blood sugar, an A1C between 7 and 8%, or higher, may be appropriate. This applies to people with limited life expectancy, long-standing diabetes, severe low blood sugar or advanced complications. Experts recommend an A1C test at least twice a year for people with diabetes, and more often when goals are not being met.1
The test is not exact to the decimal. An A1C reported as 6.8% on one test could come back anywhere from 6.4 to 7.2% on a repeat of the same blood sample, though NIDDK notes the range has narrowed under stricter quality standards.1
When the result can be wrong
- Red blood cell changes. Recent blood loss, sickle cell disease, erythropoietin treatment, hemodialysis or a transfusion can change A1C.
- Low iron. Iron-deficiency anemia can give a falsely high result.
- Kidney failure or liver disease. Both are listed as causes of false results.
- Hemoglobin variants. In people of African, Mediterranean or Southeast Asian descent, a different type of hemoglobin can interfere with some A1C tests. A mismatch between A1C and glucose tests is the usual clue, and a different A1C method may be needed.
NIDDK, the A1C test and diabetes.1
A1C on GLP-1 medicines
A1C is the main result in the trials of GLP-1 medicines for type 2 diabetes. In SURPASS-2, A1C fell by 2.0 to 2.3 points on tirzepatide against 1.9 on semaglutide 1 mg.3 Because the test reflects about 3 months, an A1C taken a few weeks after starting still includes the time before the medicine. Our page on Ozempic vs Mounjaro compares the two drugs in people with diabetes.
Sources
- National Institute of Diabetes and Digestive and Kidney Diseases. The A1C test and diabetes. Last reviewed April 2018, read November 20, 2026. niddk.nih.gov
- Nathan DM, Kuenen J, Borg R, et al. Translating the A1C assay into estimated average glucose values. Diabetes Care 2008;31:1473-1478. doi:10.2337/dc08-0545
- MOUNJARO (tirzepatide) injection prescribing information, NDA 215866/S-052, revised April 2026. U.S. Food and Drug Administration; sections 1, 2.1, 2.2, 3, 5.3, 5.5, 5.7, 6.1, 7.2, 14.2 and 16.2. accessdata.fda.gov
Related analyses
Ozempic vs Mounjaro: what the head-to-head trial and the two labels show
Mounjaro dosing schedule: 2.5 mg to 15 mg, week by week, with your own dates
Ozempic dosing schedule: 0.25 mg to 2 mg, week by week, with your own dates
Ozempic side effects: how common each one is by dose, and when they ease
Log each A1C result next to your doses and weight, and see how they move together.
Open Dose Insider