Written by: Dr. Albana Greca Sejdini, MD, MMedSc
Medically reviewed by: Dr. Ruden Cakoni, MD, Endocrinologist
Last medically reviewed: July 2026
HbA1c and estimated average glucose describe the same general period of glucose exposure in different units. HbA1c is measured from glycated hemoglobin in a blood sample. eAG is not measured directly; it is calculated from HbA1c using a regression formula.
Convert A1C to estimated average glucose, or convert a true average glucose value back to an estimated A1C. Do not enter one fasting or after-meal reading as though it were an average.
Educational limitation: The formulas describe the average relationship found in the ADAG study. Individual A1C may be higher or lower than the calculated value because of red-blood-cell biology, hemoglobin variants, kidney disease, pregnancy, glucose variability, measurement differences, and other factors.
Need only a glucose-unit conversion? Use the mmol/L and mg/dL blood sugar converter.
HbA1c—also called A1C, glycated hemoglobin, or hemoglobin A1C—measures the percentage of hemoglobin with glucose attached to it.
Hemoglobin is the oxygen-carrying protein inside red blood cells. Glucose attaches to hemoglobin through a natural nonenzymatic process called glycation. The higher glucose has been over time, the greater the proportion of glycated hemoglobin.
A laboratory reports A1C as:
A1C does not require fasting. For diagnosis, it should be performed using an appropriate standardized laboratory method. Learn how it compares with other tests in the blood tests for diabetes guide.
Red blood cells circulate for roughly 120 days, but A1C is not a simple average of every day across four months. It is a weighted average because circulating blood contains red cells of different ages.
Glucose exposure during the most recent month contributes more strongly than exposure three or four months earlier. This is why a major treatment or glucose change can begin affecting A1C before 120 days have passed.
A1C does not directly show the current glucose level, hypoglycemia frequency, post-meal peaks, overnight patterns, or glucose variability. Two people can therefore have the same A1C but very different glucose patterns and safety risks.
The A1C-Derived Average Glucose, or ADAG, study compared laboratory A1C with frequent glucose measurements collected through continuous glucose monitoring and structured finger-stick testing. The resulting regression equation was:
eAG (mg/dL) = 28.7 × A1C (%) − 46.7
eAG (mmol/L) ≈ 1.59 × A1C (%) − 2.59
eAG = (28.7 × 7) − 46.7 = approximately 154 mg/dL
154 ÷ 18.0182 = approximately 8.6 mmol/L
The reverse equation is:
Estimated A1C (%) = [eAG (mg/dL) + 46.7] ÷ 28.7
The relationship is strong at the population level but does not produce a perfect match for every person.
| A1C | IFCC | eAG mg/dL | eAG mmol/L |
|---|---|---|---|
| 5% | 31 mmol/mol | 97 | 5.4 |
| 6% | 42 mmol/mol | 126 | 7.0 |
| 7% | 53 mmol/mol | 154 | 8.6 |
| 8% | 64 mmol/mol | 183 | 10.2 |
| 9% | 75 mmol/mol | 212 | 11.8 |
| 10% | 86 mmol/mol | 240 | 13.4 |
| 11% | 97 mmol/mol | 269 | 14.9 |
| 12% | 108 mmol/mol | 298 | 16.5 |
The chart gives estimated averages. It does not mean glucose remained near that value throughout the day.
| Diagnostic category | A1C | IFCC equivalent |
|---|---|---|
| Normal diagnostic range | Below 5.7% | Below 39 mmol/mol |
| Prediabetes | 5.7%–6.4% | 39–47 mmol/mol |
| Diabetes range | 6.5% or higher | 48 mmol/mol or higher |
Without unequivocal hyperglycemia, an A1C in the diabetes range normally requires confirmation with a repeat A1C or another abnormal diagnostic test.
These diagnostic categories should not be applied as treatment labels to a person already known to have diabetes. An A1C of 6.2% while using diabetes medicine does not automatically mean the condition has become “prediabetes.”
For many nonpregnant adults with diabetes, a common treatment goal is below 7%. This is not the correct target for everyone.
A different goal may be appropriate according to age, pregnancy, diabetes duration, severe hypoglycemia, impaired awareness, heart or kidney disease, complications, medicine burden, and personal circumstances. A lower A1C is not automatically better when it is achieved through frequent low glucose.
Learn how goals are individualized in HbA1c Explained: Good Levels and Average Blood Sugar.
Do not assume that the A1C is always wrong or that the meter is always wrong. Compare dates, data completeness, laboratory method, red-cell health, and glucose patterns. See A1C vs. Blood Sugar.
| Situation | Possible effect |
|---|---|
| Recent blood loss, hemolysis, or shortened red-cell survival | May make A1C falsely low. |
| Iron-deficiency anemia | May make A1C falsely high in some circumstances. |
| Blood transfusion | Can make A1C difficult to interpret in either direction. |
| Erythropoietin treatment or dialysis | May alter red-cell turnover and A1C accuracy. |
| Hemoglobin variants | May interfere with some assay methods or change interpretation. |
| Pregnancy | Changes red-cell turnover; pregnancy uses specific glucose tests and targets. |
| Advanced kidney or liver disease | May affect red-cell survival and create discordance. |
If A1C and repeated glucose data do not match, the clinician may repeat the test, review the laboratory method, use plasma glucose criteria, or consider another marker such as fructosamine or glycated albumin in selected situations.
GMI is not a laboratory A1C. CGM also provides time in range, time below range, time above range, and variability—information A1C and eAG cannot show.
Measurement approximately every three months helps determine whether goals have been reached and maintained.
There is no single A1C level that automatically requires hospitalization. A1C reflects chronic glucose exposure and cannot show whether a person is currently dehydrated, acidotic, producing ketones, severely hypoglycemic, or mentally altered.
A very high A1C suggests prolonged hyperglycemia and needs prompt treatment review. Emergency or hospital care depends mainly on current glucose, ketones, vomiting, hydration, breathing, mental status, and the possibility of DKA or HHS.
No. Fasting glucose is one measurement after fasting. eAG estimates average glucose across day and night over a longer period.
No. Entering one reading into the reverse calculator produces a mathematically possible number but not a medically meaningful A1C estimate.
No. Glucose may have moved above and below 154 mg/dL. The estimate does not reveal variability or hypoglycemia.
Yes. Recent weeks have greater influence, so a major sustained glucose change can alter A1C before three months have passed.
The laboratory eAG is calculated from A1C, while the CGM average is calculated from sensor readings during a specific period. Biological and measurement differences can produce discordance.
Medical disclaimer: This calculator and article provide general education. They do not diagnose diabetes, set an A1C goal, prescribe treatment, or determine whether emergency care is required. Do not change insulin or medicine based only on a calculated eAG or estimated A1C.