On This Page – Quick Medical Summary
Where your estimated average glucose number comes from
If your lab report shows a second number beside your A1C — something like eAG 154 mg/dL — you are in the right place. Estimated average glucose is your A1C rewritten in the units your glucose meter uses. It was not measured separately and no extra blood was drawn for it: it is your A1C percentage put through a formula.
That one fact shapes everything else on this page, so it is worth saying plainly before the arithmetic: eAG is calculated, not measured.
Where you are now decides where to go next. If you want the conversion itself, the next section has it; if your eAG and your meter disagree, skip to the third; if you want to know how close the estimate is, the fourth is the honest answer. If you are reading someone else’s result, the fifth section is written for you.
If your lab report already shows an eAG
Many labs now calculate it for you, though not all do. If yours does not, nothing is wrong with your result — the conversion is simply not part of every report format, and you can work it out from what your A1C result actually is in one step.
If you are reading someone else’s result
Caregivers often meet this number first, on a parent’s or partner’s printout, with no explanation attached. You do not need to reconcile it with anything before the next appointment. Reading it correctly is enough, and the fifth section sets out exactly what it can and cannot settle.
ℹ️ Medical Disclaimer: This article explains how a laboratory value is calculated and what it can reasonably be used for; it does not diagnose diabetes or prediabetes, does not interpret your individual result, and does not recommend, start, stop or adjust any medication, insulin dose, diet or monitoring device. Decisions about diagnosis, treatment, medication and testing frequency belong with a board-certified physician — a primary care clinician or an endocrinologist — who can see your full record, and insurance coverage for testing varies by plan and is not addressed here.
What your A1C works out to in mg/dL
The conversion comes from a single published equation. An A1C of 7% works out to an estimated average glucose of about 154 mg/dL; an A1C of 6% works out to about 126 mg/dL. Those are population averages with a real range around them, which the next two sections explain.
🔬 How It Works: The A1C test measures the share of your hemoglobin that has glucose attached to it, which builds up over the life of a red blood cell; to turn that percentage into a glucose figure, researchers measured both in the same people and fitted a line between them. The published equation is estimated average glucose (mg/dL) = 28.7 × A1C − 46.7, and it came from the A1c-Derived Average Glucose study, which followed 507 people across ten international centers and collected roughly 2,700 glucose values from each of them over three months. The fit was strong but not exact — at an A1C of 7%, the published 95% confidence interval around that 154 mg/dL runs from 123 to 185 mg/dL.
The conversion, worked through on one number
Take an A1C of 7.0%. Multiply by 28.7 to get 200.9, then subtract 46.7. The answer is 154.2, which published tables print as 154 mg/dL.
Four common anchor points, as the standards bodies publish them:
- A1C 6.0% — about 126 mg/dL
- A1C 7.0% — about 154 mg/dL
- A1C 8.0% — about 183 mg/dL
- A1C 9.0% — about 212 mg/dL
Use the published figures rather than re-rounding your own arithmetic. At 6.0% the equation returns 125.5, and published tables print 126 — a page that rounds the other way quietly disagrees with the standards bodies.
The same number in mmol/L
Outside the United States the same estimate is reported in millimoles per liter. An A1C of 7.0% is published as 8.6 mmol/L, and 8.0% as 10.2 mmol/L. Our blood sugar unit converter handles mg/dL and mmol/L in both directions once you have the mg/dL figure.
One oddity worth knowing, because it will otherwise look like an error: the source paper published a second equation for mmol/L directly, and at some A1C values it lands a tenth of a unit away from the printed table. That gap is roughly one-tenth of a millimole, while the estimate’s own margin at an A1C of 7% spans about 60 mg/dL. Worrying about the rounding is worrying about the wrong end of the problem.
Why your lab may not print it at all
Reporting eAG is a formatting choice, not a clinical one. Some laboratories include it, some do not, and its absence says nothing about your result or your care. If you want the full set of conversions rather than four anchors, the pillar article carries the complete table alongside what each A1C level means.
Why your eAG will not match your meter average
This is the single most common reason people arrive at this page, and the answer is reassuring: a gap is expected. Federal patient guidance states it directly — the eAG number will not match your daily glucose readings, because it is a long-term average rather than your blood glucose level at a single moment. That guidance page was last reviewed in April 2018 and this part of it has not been superseded.
So the two numbers are not competing. They are answering different questions.
A three-month average against a handful of moments
Your A1C reflects roughly three months of glucose exposure, with the most recent weeks weighing more heavily. Your meter reflects the specific seconds you chose to test. Averaging a set of chosen moments does not reconstruct a continuous three-month average, and it was never going to.
When you test changes what your average looks like
Most people test before meals and at bedtime, because that is when testing is useful for daily decisions, and those are systematically the lower points of the day. An average built mostly from pre-meal readings will tend to sit below a true round-the-clock average — one ordinary, non-alarming reason a meter average can come in under an eAG. The mechanics are worked through in why your A1C and your meter can disagree, and the same logic applies to a single fasting result against a three-month average.
Steadier is not the same as more precise
The A1C is the steadier of the two measurements, because it is less affected by short-term change than a one-off glucose test. That is a statement about the A1C itself, not about the eAG, which is a calculation performed on the A1C afterwards. Holding those two apart is the whole trick to reading this number well, and the next section is where it matters most.
✅ Patient Action: Do not try to reconcile the two numbers yourself before your appointment. A gap between an eAG and a meter or sensor average is expected and, on its own, is not evidence that either number is wrong. But a large or persistent gap is worth showing someone: bring your meter or your sensor download to your next visit and ask your primary care clinician or diabetes care team, “My meter average and the eAG on my report are about this far apart — is that the gap you would expect from how and when I test, or is it worth looking into?”
How close the estimate actually is
Before any figure: your A1C is a measured value, and your eAG is a calculation performed on it afterwards. Everything below is about how much give there is in the arithmetic. None of it is a reason to discount the A1C your clinician is actually acting on.
📊 Clinical Data Point: The relationship between A1C and average glucose in the source study had a coefficient of determination of R² = 0.84 (P < 0.0001). At an A1C of 7%, the published estimated average glucose is 154 mg/dL, with a 95% confidence interval of 123–185 mg/dL. Sources: A1c-Derived Average Glucose study, Diabetes Care 2008;31(8):1473–1478; interval as published in Diabetes Care 2009;32(1):e10.
What a correlation of 0.84 does and does not promise
An R² of 0.84 is a strong relationship. Across a large group, A1C and average glucose track each other closely enough that the translation is genuinely useful.
What a strong group relationship does not promise is precision for one person. The same statistic that makes the average reliable leaves real room around any single converted value — and the study authors reported that room rather than hiding it.
The range around a single converted number
At an A1C of 7%, the published estimate is 154 mg/dL and the 95% interval runs from 123 to 185 mg/dL. That interval widens as the A1C rises.
Put concretely: 154 is the middle of a band roughly 60 mg/dL wide, not a reading. That is why the next section argues against using it as a target. The margin around the A1C measurement itself is a separate question, worked through in the margin around the A1C itself and in what the lab has to prove about your A1C.
What the guideline body calls its own table
The American Diabetes Association created and promotes eAG. In the 2026 Standards of Care in Diabetes, Section 6, the association describes its own A1C-to-mean-glucose table as providing rough equivalents based on the A1c-Derived Average Glucose data. That wording is unchanged from the 2024 and 2025 editions.
It is an unusually candid line, and it is the most useful sentence on this page. The organization with every reason to present the conversion as exact describes it as rough. Treat it the way its authors do.
What your eAG can and cannot tell you
Once you stop reading eAG as a measurement, its real uses become clear — and so do the questions it cannot settle. The table below is the short version.
| Question you actually have | Does your eAG answer it? | What does answer it | Why |
|---|---|---|---|
| Roughly what glucose level has my A1C been reflecting? | Yes | eAG | This is the exact job the conversion was built for |
| Is my long-term control moving up or down? | Yes, over time | Successive A1C or eAG values | A trend across visits is more informative than any single converted number |
| What is my blood sugar right now? | No | A meter or sensor reading | eAG describes months, not moments |
| Is my meter average correct? | No | Your own recorded readings | An estimate cannot audit a measurement |
| Am I having highs and lows that cancel out? | No | Continuous glucose monitoring metrics | An average conceals variability by design |
| Should I change anything I am doing? | No | Your clinician, with your full record | No single converted value supports a treatment decision |
Sources: NIDDK A1C guidance (last reviewed April 2018); MedlinePlus A1C test entry (review date May 2024); ADA Standards of Care in Diabetes—2026, Section 6.
Questions this number answers well
The conversion exists to make a percentage feel like something. For a treatment-comparer watching an A1C move from 7.4% to 7.1% across visits, seeing that shift in mg/dL is genuinely easier to hold onto, and the direction of travel is real even when each individual point is fuzzy.
Trends survive the uncertainty better than single values do, because the same margin applies at both ends of the comparison.
Questions it cannot answer, and what does
If your question is what a normal reading looks like, that is a different measurement entirely and we answer it separately in what a normal blood sugar reading looks like. If your question is whether highs and lows are cancelling out inside your average, no conversion can reach it — that is what how time in range is reported is for.
If you use a sensor, you may also see a glucose management indicator on your app. It is not the same quantity as an eAG: it is calculated from sensor data by a different method, over a different window, and the two can differ for the same person in the same month. Neither is wrong when they disagree; they are different calculations, and they belong in the same conversation with your care team rather than in competition.
🩺 Physician Note: A point that causes real confusion: the A1C is the steadier measurement, and yet MedlinePlus notes that your actual recorded readings are usually more reliable than the estimated average glucose calculated from an A1C. Both are true because they describe different things — the A1C is steady as a measurement, while the eAG is a translation of it, and the looseness lives in the translation. If you already have weeks of real readings, you are holding measurements rather than needing an estimate of them.
When the estimate does not fit you
The conversion assumes an ordinary relationship between glucose in your blood and glucose stuck to your hemoglobin. Several conditions change that relationship, and where they do, the eAG stops describing you well — not because the arithmetic failed, but because its input did.
Conditions that change how long red blood cells live
Because the A1C builds up over a red blood cell’s lifespan, anything that shortens or lengthens that lifespan shifts the result. Federal guidance names recent blood loss, sickle cell disease, erythropoietin treatment, hemodialysis and transfusion among the conditions that can change A1C levels, and notes that a falsely high result can occur in people who are very low in iron.
The 2026 Standards of Care add hemolytic and other anemias, glucose-6-phosphate dehydrogenase deficiency and pregnancy to the list of states that can interfere with A1C accuracy. If you are being worked up for low hemoglobin or anemia, that is worth raising in the same conversation.
Inherited hemoglobin variants
Some inherited hemoglobin variants interfere with some A1C methods. They are more common in people whose ancestors came from Africa, South and Southeast Asia, or the Mediterranean, and most carriers have no symptoms and do not know they have one. NIDDK covers this in detail for people of African, Mediterranean or Southeast Asian descent.
The framing matters. The test is not unsuitable for anyone’s ancestry. A particular variant can interfere with a particular assay, most assays used in the United States handle the common single-copy variants accurately, and where one does interfere the answer is a different method.
Why two people with the same average can have different A1Cs
This is the honest edge of the evidence. People appear to differ in how readily glucose attaches to hemoglobin, so the same average glucose can produce somewhat different A1C values — and therefore different eAG figures — in different people.
How much this matters, and whether it can be predicted for an individual, is not settled. NIDDK lists it among the questions still being researched, including whether knowing a person’s red blood cell lifespan would help interpret their result. Anyone telling you this is a solved problem is ahead of the evidence.
⚠️ Clinical Warning: In some situations the A1C cannot be measured at all — the 2026 Standards of Care state that people with sickle cell disease or another homozygous hemoglobin variant lack the ordinary adult hemoglobin the test depends on, and where A1C cannot be interpreted or measured, guidelines point to other approaches including self-monitoring, continuous glucose monitoring, or shorter-window laboratory markers reflecting roughly two to four weeks rather than three months.
If any condition on this page applies to you, ask your primary care clinician: “Is there a reason my A1C would not reflect my actual average, and should we be using a different measure?” Note also that the source study’s own conclusion describes eAG as applicable to most patients with type 1 and type 2 diabetes, so if you have a result in the 5.7 to 6.4 range and no diagnosis, that is the population the translation was validated in.
What not to do with this number
Two conclusions are easy to reach from everything above, and both are wrong. This section exists to close them.
Do not decide your A1C does not matter
If you have come away thinking the conversion is rough, so the A1C behind it must be unreliable — that is the wrong lesson, and it is the one this page can most easily cause.
The looseness described here lives in the translation from a percentage into mg/dL. Your A1C is the measured value, it is the number your clinician acts on, and an elevated one does not stop mattering because its converted twin comes with a wide interval. If an A1C result has worried you, that worry is about the A1C, and it belongs at an appointment.
Do not change anything on your own
An eAG is not a target and it is not a scorecard. Nothing on this page supports adjusting food, activity, monitoring or any prescribed treatment, and a converted number is the weakest possible basis for doing so.
⚠️ Clinical Warning: First, we do not turn A1C thresholds into blood sugar targets — converting a diagnostic or goal A1C into an mg/dL figure and presenting it as a number to hit invents a cut-off no guideline body has issued, and the interval around the conversion is wider than the gaps between the thresholds people would compare it against.
Second, we do not recommend or link to any home A1C kit or glucose device here, because our own cited sources rule it out: federal guidance says point-of-care A1C results should not be used for diagnosis, MedlinePlus says your real recorded readings beat an estimate calculated from an A1C, and recommending a device to check an estimate would be selling against our own argument. On diagnosis itself, a result of 6.5% or above is the criterion and any test used to diagnose diabetes requires confirmation with a second measurement unless clear symptoms are present — wording varies across published sources, and that is the criterion as written.
Common questions about A1C and average blood sugar
1. What is estimated average glucose?
Estimated average glucose, or eAG, is your A1C percentage converted into the units a glucose meter uses. It is calculated from your A1C rather than measured separately, so no additional blood is drawn for it. Many laboratories now print it beside the A1C result, though not all do, and its absence says nothing about the quality of your result.
2. How do you convert A1C to average blood sugar?
Multiply your A1C by 28.7 and subtract 46.7 to get estimated average glucose in mg/dL. An A1C of 7.0% gives 154.2, which published tables print as 154 mg/dL. Use the published figures rather than your own rounding, since standards bodies print 126 mg/dL at an A1C of 6.0% where the raw equation returns 125.5.
3. Why is my eAG different from my meter average?
Because they measure different things: your eAG reflects roughly three months of glucose exposure, while your meter reflects the particular moments you tested. Federal guidance states that the eAG will not match daily readings for exactly this reason. Most people also test before meals, which are systematically lower points in the day, pulling a meter average down.
4. Is eAG the same as my average blood sugar?
No — it is an estimate of your average rather than a measurement of it. The conversion comes from a study of 507 people and describes how A1C and average glucose relate across a group. For any one person the published estimate carries a real range: at an A1C of 7%, a 95% interval of 123 to 185 mg/dL around a midpoint of 154.
5. How accurate is the eAG number?
The underlying relationship is strong, with an R² of 0.84, but a strong group relationship still leaves room around a single value. The American Diabetes Association describes its own conversion table as providing rough equivalents. Treat your eAG as a well-founded approximation rather than a reading, and discuss any result that concerns you with your clinician.
6. Is eAG the same as GMI on my CGM?
No — a glucose management indicator is calculated from continuous sensor data by a different method and over a different window than the eAG derived from your A1C. The two can differ for the same person in the same month without either being wrong. Bring both to your care team rather than trying to decide between them yourself.
7. Can I use eAG as a blood sugar target?
No, and this is the most consequential answer on this page. The interval around a converted value is wider than the differences between the targets people would compare it against, so an eAG used as a number to hit is misleading. Targets belong to your clinician, who sets them against your full record.
8. Why isn’t eAG on my lab report?
Reporting eAG is a formatting decision made by the laboratory, not a clinical one. Some report it, some do not. Nothing about its absence reflects on your result, your care or your A1C, and you can work out the estimated average glucose yourself from your A1C using the published equation in this article.
9. Can my eAG be wrong?
It can fail to describe you well. Conditions that change how long red blood cells live — recent blood loss, certain anemias, kidney failure, transfusion — shift the A1C the conversion depends on, and some inherited hemoglobin variants interfere with some testing methods. Ask your clinician whether any of these apply to you.
10. What is eAG in mmol/L?
The same estimate reported in millimoles per liter, which is standard outside the United States. Published tables give 8.6 mmol/L at an A1C of 7.0% and 10.2 mmol/L at 8.0%. Our blood sugar converter handles mg/dL and mmol/L in both directions once you have the mg/dL figure from the equation in this article.
11. Should I ask my doctor about my eAG?
Bring it if the gap between it and your own readings is large or persistent, or if you are unsure what your A1C is being compared against. A useful question is whether your clinician wants to look at your A1C, your day-to-day readings, or both when deciding whether anything needs to change.
What to take away from your eAG
Your estimated average glucose is your A1C restated in familiar units. It is calculated rather than measured, it comes with a real range around it, and the organization that publishes the conversion calls its own table a set of rough equivalents.
That does not make it useless. It makes it a translation — good for holding a percentage in your head, good for watching a direction of travel across visits, and poor as a target to check your meter against. A gap between it and your own readings is expected, not a fault.
If something here has left you uncertain about your own result, the next step is a conversation rather than a calculation. Take your A1C, your eAG and your own readings to your next appointment, and read the full A1C result alongside them before you go.
About this content
How this article was put together: researched from recognised health sources, drafted with the help of AI tools, and edited by hand, with sources linked throughout.
Sameer Patel is the founder and editor of My Medicine Advisor. He is not a doctor or medical professional — before starting this site he worked in banking,…
Medical disclaimer
The content on MyMedicineAdvisor is provided for general informational and educational purposes only and is not a substitute for professional medical advice, diagnosis, or treatment. Health information on this website should not be used to diagnose, treat, cure, or prevent any condition without guidance from a qualified healthcare professional. Always seek the advice of your doctor, physician, or another licensed healthcare provider with any questions you may have regarding a medical condition, symptoms, medications, or treatment decisions.
