An A1C result is a percentage, but most of us think about glucose in mg/dL or mmol/L, the units on a meter or sensor. The conversion is a short formula, and this page puts the formula, a full chart, and the caveats in one place.
Quick answer: Estimated average glucose (eAG) in mg/dL = 28.7 × A1C − 46.7. An A1C of 5.7 percent corresponds to about 117 mg/dL, 6.5 percent to about 140 mg/dL, and 7 percent to about 154 mg/dL. To get mmol/L, divide mg/dL by 18. The result estimates your average glucose over roughly the past three months. It is not a prediction of any single reading, and it does not replace the advice of the clinician who knows your history.
The A1C to average glucose chart
This table applies the formula directly and rounds to the nearest whole mg/dL. Use the A1C and average glucose calculator to convert any value, in either direction.
| A1C (%) | Estimated average glucose (mg/dL) | mmol/L (mg/dL ÷ 18) |
|---|---|---|
| 4.5 | 82 | 4.6 |
| 5.0 | 97 | 5.4 |
| 5.5 | 111 | 6.2 |
| 5.7 | 117 | 6.5 |
| 6.0 | 126 | 7.0 |
| 6.4 | 137 | 7.6 |
| 6.5 | 140 | 7.8 |
| 7.0 | 154 | 8.6 |
| 7.5 | 169 | 9.4 |
| 8.0 | 183 | 10.2 |
| 9.0 | 212 | 11.8 |
| 10.0 | 240 | 13.4 |
| 11.0 | 269 | 14.9 |
| 12.0 | 298 | 16.5 |
Notice that each 1-point change in A1C is worth about 28.7 mg/dL of average glucose, and each half-point is worth about 14. That is why a drop from 7.0 to 6.5 percent represents an average glucose that is roughly 14 mg/dL lower, which is a real but modest shift on a meter.
Where the formula comes from
The formula is from the A1C-Derived Average Glucose (ADAG) study, published in 2008 by Nathan and colleagues. The researchers analyzed 507 people: 268 with type 1 diabetes, 159 with type 2 diabetes, and 80 without diabetes. Each participant contributed about 2,400 continuous glucose monitor readings and about 300 fingerstick values, and the team fitted a line between A1C and average glucose. The relationship was strong enough to justify the conversion, and laboratories began reporting eAG alongside A1C. The study’s authors also noted that eAG is accurate for most people with type 1 and type 2 diabetes, not for everyone, which matters for the caveats below.
Worked examples
- A1C to glucose. A lab reports 6.1 percent. 28.7 × 6.1 = 175.1, and 175.1 − 46.7 = 128.4. The estimated average glucose is about 128 mg/dL, or about 7.1 mmol/L (128 ÷ 18).
- Glucose to A1C. A sensor shows a 14-day average of 150 mg/dL. (150 + 46.7) ÷ 28.7 = 6.85. The A1C-equivalent is about 6.9 percent.
- Sensor average to GMI. A CGM’s glucose management indicator uses its own formula: GMI = 3.31 + 0.02392 × mean glucose in mg/dL. For a mean of 120 mg/dL that is 6.18, reported as 6.2 percent. For a mean of 100 mg/dL it is 5.70 percent, and for 154 mg/dL it is 6.99 percent.
The GMI formula comes from a separate analysis of CGM and lab A1C data (Bergenstal and colleagues, 2018). It is not the same equation as eAG, so the two will not always agree exactly even for the same person.
What the A1C categories are
The American Diabetes Association defines normal A1C as below 5.7 percent, prediabetes as 5.7 to 6.4 percent, and diabetes as 6.5 percent or higher on confirmed testing. Here is what those thresholds mean in average glucose terms.
| ADA A1C category | A1C | Approximate eAG (mg/dL) |
|---|---|---|
| Normal | Below 5.7% | Below 117 |
| Prediabetes | 5.7% to 6.4% | 117 to 137 |
| Diabetes (confirmed testing) | 6.5% or higher | 140 or higher |
These bands show why an A1C of 6.4 percent, still in the prediabetes range, equals an average glucose of about 137 mg/dL, and 6.5 percent is about 140. The categories are cut-offs for conversation and follow-up, not a clinical diagnosis by themselves. For the full explanation, see the complete guide to A1C and glucose and the comparison of A1C, fasting glucose, and OGTT. The NIDDK A1C test page is a good federal reference.
Why your meter or sensor average may not match
People are often surprised when the eAG on a lab report does not line up with the numbers on their meter. There are several reasons, and none of them means one test is simply wrong.
- Different time windows. A1C is weighted toward the most recent weeks, while a sensor average covers the days it was worn.
- Different things measured. A1C measures the share of hemoglobin that has glucose attached. That depends on glucose, but also on how long red blood cells live.
- Biology. Anemia, certain hemoglobin variants, kidney disease, pregnancy, and recent blood loss or transfusion can shift A1C in ways that do not track average glucose.
- Sampling. A meter only captures the moments you test. A sensor captures far more, but it has its own error.
Two people with the same A1C can have different average glucose levels. We cover this in why A1C can be wrong and in continuous glucose monitor vs A1C. A GMI that differs from a lab A1C by a few tenths is common and does not mean either number is mistaken.
Quick unit conversions you will actually use
If you read research papers, international websites, or a sensor app set to different units, these conversions come up constantly. They use the mg/dL ÷ 18 rule.
| Glucose (mg/dL) | Glucose (mmol/L) | Where you may see it |
|---|---|---|
| 54 | 3.0 | A common “level 2” low threshold in CGM reports |
| 70 | 3.9 | The usual lower edge of the CGM target range |
| 100 | 5.6 | A familiar fasting-glucose reference point |
| 126 | 7.0 | A fasting glucose threshold used in diagnostic criteria |
| 140 | 7.8 | Often used as an upper reference after meals |
| 180 | 10.0 | The usual upper edge of the CGM target range |
These are reference points, not personal targets. The point of the table is to make unit differences less confusing when you compare your numbers with something you read.
How often A1C is measured and why timing matters
Because A1C reflects roughly the past three months, the test is generally repeated at intervals rather than weekly. A change in food, activity, or medicine shows up gradually, and a retest too soon will mostly show the old average. If you are making a change and want faster feedback, a meter or sensor is the better short-term tool, and A1C is the better long-term check. Ask your clinician how often your own A1C should be measured, since the answer depends on whether you have a diagnosis, whether your treatment has changed, and how close you are to your target.
Common mistakes when using the chart
- Treating eAG as a target. Targets are individual. The chart converts a number; it does not tell you what number to aim for.
- Comparing it with one reading. An average of 126 mg/dL contains readings well above and well below 126. A single fasting value cannot be compared directly.
- Ignoring conditions that affect A1C. If you have anemia, a hemoglobin variant, or kidney disease, ask whether A1C is reliable for you.
- Mixing units. Make sure you know whether a number is mg/dL or mmol/L before you compare.
- Chasing small changes. A shift of one or two tenths of a percent can come from test variation. Look at the trend across tests.
How to use A1C alongside daily numbers
A1C gives a three-month summary. A meter or sensor gives the daily shape. Together they answer different questions: A1C asks how your glucose has averaged lately, while readings ask what is happening after a particular meal or overnight. If you wear a sensor, our guides on how to read your CGM data and normal CGM glucose range without diabetes explain which numbers matter. Food choices that influence both are covered in diet for prediabetes and diet for type 2 diabetes, and carb counting is explained in how to read a nutrition label for carbs.
Low blood sugar warning: if you take insulin, a sulfonylurea, or another glucose-lowering medicine, changing what you eat or adding a product can push glucose too low. Symptoms include shakiness, sweating, confusion, and a fast heartbeat. Talk to your prescriber before changing your eating or your medicine, and follow your own low-blood-sugar plan.
Questions to bring to your appointment
- Could anything in my health history make my A1C unreliable?
- Does my A1C match my meter or sensor readings?
- What A1C target is right for me, and why?
- When should we retest, and should we use another test as well?
- Would a sensor or fructosamine test add useful information in my case?
Frequently asked questions
Is eAG the same as my average glucose from my meter or sensor?
No. eAG is a statistical estimate calculated from A1C. Your meter or sensor average comes from the readings you actually took. They are often close but can differ by a meaningful amount, because A1C reflects the share of hemoglobin with glucose attached, weighted toward recent weeks, and because meters and sensors only sample certain moments.
Why does my lab report show an eAG that is different from what I expected?
The formula is a line fitted to data from many people. Any one person can sit above or below it. Factors that change red blood cell lifespan or hemoglobin type can also move A1C without moving average glucose in the same way.
Can I convert average glucose back to A1C?
Yes, by rearranging the formula: A1C = (average glucose in mg/dL + 46.7) ÷ 28.7. An average of 150 mg/dL works out to about 6.9 percent. The calculator on this site does it in either direction.
How is GMI different from A1C and eAG?
GMI, the glucose management indicator, starts from a CGM’s mean glucose and reports an A1C-style percentage. eAG starts from a lab A1C and reports a glucose. Each is an estimate of the other number, so GMI and lab A1C can disagree by a few tenths.
Does a lower A1C always mean a safer A1C?
Not for everyone. Targets are individual, and pushing A1C lower with certain medicines raises the risk of low blood sugar. Ask your prescriber what target fits your age, health, and medicines.
What is the difference between mg/dL and mmol/L?
They are two units for the same measurement. United States labs usually report mg/dL, while many other countries use mmol/L. Divide mg/dL by 18 to get mmol/L, or multiply mmol/L by 18 to go the other way.
Does one A1C result diagnose diabetes?
A1C is one test. Diagnosis generally relies on confirming an abnormal result, usually with a repeat test or a different test, as explained in the comparison of A1C, fasting glucose, and OGTT. Your clinician interprets the result in context.
About the author: Charles Kirkland is the founder of Metabolic Clarity Labs and author of The Journey Back. He is not a physician. This is general education, not a medically reviewed treatment plan.
Sources checked October 31, 2026. Discuss personal decisions with a qualified clinician. Do not start, stop, or change prescribed treatment based on this article.
Leave a Reply