A1C Calculator
Convert an A1C result to estimated average glucose in mg/dL or mmol/L, or an average back to A1C, with the ADAG equation, and A1C between % and mmol/mol.
Result
Enter a1c, % to see the result.
https://calcgallery.com/health-risk/a1c-calculator/
How to use the A1C Calculator
- Choose what you are starting from, an A1C result or an average glucose, and the unit it is in.
- Enter the number exactly as your lab report, meter or sensor report shows it.
- Read the estimated average glucose, or the A1C that matches your average, in both units.
- Take any question about what the number means for you to the clinician who ordered the test.
Read this first
This is a conversion, not a diagnosis, and not medical advice. It places a number on a published line. It cannot tell you whether your A1C is where it should be, whether anything needs to change, or what to do next. Diabetes and prediabetes are diagnosed, and A1C results are interpreted, by a clinician. Discuss your results with them.
What A1C and eAG are
A1C, also written HbA1c, is the share of your haemoglobin that has glucose attached. Glucose attaches to haemoglobin at a rate that rises with the glucose around it, and red blood cells circulate for months, so A1C reflects your glucose over roughly the previous two to three months rather than on the day of the test.
Estimated average glucose, eAG, restates an A1C in the units a glucose meter uses, mg/dL or mmol/L. It is the average glucose the ADAG study measured, on average, in people with that A1C. It is an estimate for a population, not a measurement of you.
How it is calculated
The ADAG study (Nathan and colleagues, Diabetes Care, 2008) measured glucose in 507 adults, with type 1 diabetes, type 2 diabetes or no diabetes, using continuous monitoring and fingersticks for about three months, then compared each person’s average with their A1C. A straight line fitted the data best:
eAG (mg/dL) = 28.7 × A1C − 46.7eAG (mmol/L) = 1.5944 × A1C − 2.594A1C is the NGSP percentage. The mmol/L line is printed under the paper’s table; it is the mg/dL line divided by 18, and the page uses each line for its own unit so the answers match the paper’s table. Going the other way, from an average back to an A1C, the page solves the same line for A1C:
A1C = (eAG in mg/dL + 46.7) ÷ 28.7A1C = (eAG in mmol/L + 2.594) ÷ 1.5944The paper states the line in one direction only; the reverse is its algebra, so a number converted there and back returns to itself.
A1C in mmol/mol. The NGSP publishes the master equations linking its percentage with the IFCC reference unit:
IFCC (mmol/mol) = 10.93 × NGSP (%) − 23.50NGSP (%) = 0.09148 × IFCC (mmol/mol) + 2.152The two were fitted separately and are not exact inverses, so each direction uses its own equation, as laboratories do. The difference is a few hundredths and never shows at the precision results are reported to.
Each number has its own field
An average glucose in mmol/L is about eighteen times smaller than the same average in mg/dL, and an A1C in mmol/mol is several times larger than the same A1C in %. A number typed against the wrong unit can still produce a plausible-looking answer. So the page asks what you are starting from first, and each unit has its own field and its own limits: a value that belongs to another unit is refused rather than converted.
What the line does not account for
The ADAG line describes the people in the study, and the paper sets out who they were:
- Adults in stable control only. Children and pregnant women were excluded, so the line is not established for them.
- No conditions that affect red blood cells. People with haemoglobin variants, anaemia or other red cell disorders were excluded, since these change A1C independently of glucose.
- Scatter around the line. In the study, about 90% of measured averages fell within 15% of the value the line predicts. Your own average can sit anywhere in that spread, or outside it.
The line was fitted over A1C values from about 4% to above 8.5%, and the paper tabulates it from 5% to 12%. The page accepts A1C from 4% to 14%; the further from the middle of that range, the less the study’s data say about it.
Before you compare numbers
Match the units. Check whether your report prints A1C in %, mmol/mol or both, and whether your meter reads mg/dL or mmol/L.
Compare like with like. eAG is a population estimate, a meter average is a set of spot readings, and a sensor’s glucose management indicator is a different estimate again. They are not expected to agree exactly.
One result is one result. A1C is read over time and alongside other tests. Its meaning for you is a question for your clinician.
With your numbers
- IFCC A1C = 10.93 × 7 − 23.5 = 53 mmol/mol
- eAG = 1.5944 × 7 − 2.594 = 8.57 mmol/L
- eAG = 28.7 × 7 − 46.7 = 154.2 mg/dL
- An A1C of 7% sits at an estimated average glucose of 154 mg/dL on the ADAG line
Quick reference
A population average from one study, not your own glucose. What an A1C means for you is for your clinician to interpret.
| Example | Result |
|---|---|
| A1C 5.7% | Estimated average glucose, mg/dL: 117 |
| A1C 6.5% | Estimated average glucose, mg/dL: 140 |
| A1C 7% | Estimated average glucose, mg/dL: 154 |
| A1C 8% | Estimated average glucose, mg/dL: 183 |
| A1C 10% | Estimated average glucose, mg/dL: 240 |
| 48 mmol/mol | Estimated average glucose, mg/dL: 141 |
| 53 mmol/mol | Estimated average glucose, mg/dL: 154 |
| Average 154 mg/dL | A1C, %: 7.0 |
| Average 183 mg/dL | A1C, %: 8.0 |
| Average 8.6 mmol/L | A1C, %: 7.0 |
Where the numbers come from
The fixed figures this calculator relies on, and the published source each one was checked against. How the calculators are built and checked
5 sources
Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ, for the A1c-Derived Average Glucose (ADAG) Study Group, "Translating the A1C assay into estimated average glucose values", Diabetes Care 31(8):1473-1478, 2008, doi:10.2337/dc08-0545, read at https://pmc.ncbi.nlm.nih.gov/articles/PMC2742903/: "Linear regression analysis between the A1C and AG values provided the tightest correlations (AGmg/dl = 28.7 × A1C − 46.7, R2 = 0.84, P < 0.0001)"; Table 2 footnote, "Linear regression eAG (mg/dl) = 28.7 × A1C − 46.7".
Used for
- Slope of the ADAG regression of average glucose on A1C, in mg/dL.
Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ, for the A1c-Derived Average Glucose (ADAG) Study Group, "Translating the A1C assay into estimated average glucose values", Diabetes Care 31(8):1473-1478, 2008, doi:10.2337/dc08-0545, read at https://pmc.ncbi.nlm.nih.gov/articles/PMC2742903/: "AGmg/dl = 28.7 × A1C − 46.7"; Table 2 footnote.
Used for
- Intercept subtracted in the ADAG regression of average glucose on A1C, in mg/dL.
Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ, for the A1c-Derived Average Glucose (ADAG) Study Group, "Translating the A1C assay into estimated average glucose values", Diabetes Care 31(8):1473-1478, 2008, doi:10.2337/dc08-0545, read at https://pmc.ncbi.nlm.nih.gov/articles/PMC2742903/: Table 2 footnote, "Linear regression eAG (mmol/l) = 1.5944 × A1C − 2.594"; the text gives "AGmmol/l = 1.59 × A1C − 2.59".
Used for
- Slope of the ADAG regression of average glucose on A1C, in mmol/L, as the paper prints it under Table 2 (the text rounds it to 1.59).
Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ, for the A1c-Derived Average Glucose (ADAG) Study Group, "Translating the A1C assay into estimated average glucose values", Diabetes Care 31(8):1473-1478, 2008, doi:10.2337/dc08-0545, read at https://pmc.ncbi.nlm.nih.gov/articles/PMC2742903/: Table 2 footnote, "eAG (mmol/l) = 1.5944 × A1C − 2.594".
Used for
- Intercept subtracted in the ADAG regression of average glucose on A1C, in mmol/L, as printed under Table 2.
NGSP, "IFCC Standardization: IFCC and NGSP", master equations, https://ngsp.org/ifccngsp.asp (undated page; equations from Hoelzel W et al., Clin Chem 50:166-174, 2004, and Geistanger A et al., Clin Chem 54(8):1379, 2008). Table 2: "NGSP = (0.09148*IFCC) + 2.152" and "IFCC = (10.93*NGSP) - 23.50", IFCC in mmol/mol; Table 1 pairs NGSP 7.0% with IFCC 53 mmol/mol.
Used for
- Slope of the IFCC-NGSP master equation giving NGSP (DCCT) HbA1c in % from IFCC HbA1c in mmol/mol.
- Intercept of the IFCC-NGSP master equation giving NGSP HbA1c in % from IFCC mmol/mol.
- Slope of the NGSP-to-IFCC equation giving IFCC HbA1c in mmol/mol from NGSP %. Published alongside, not derived from, the equation the other way.
- Intercept subtracted in the NGSP-to-IFCC equation, in mmol/mol.
Common questions about the A1C Calculator
- Does this tell me whether I have diabetes or prediabetes?
- No. This page converts a number from one scale to another and does nothing else. Diabetes and prediabetes are diagnosed by a clinician, who reads an A1C result alongside other tests, how it was measured, your history and anything that can make A1C read high or low. The page deliberately shows no ranges, targets or labels, so that a conversion is never mistaken for a diagnosis.
- Why does my meter or sensor average differ from the eAG for my A1C?
- Because eAG is not your average: it is the average the ADAG study found, across 507 adults, for people with your A1C. Individuals sit above or below that line, and the study itself found wide scatter around it. A fingerstick average also depends on when you test, since readings taken mostly before meals or mostly after them miss part of the day. A continuous sensor sees the whole day but over a shorter window than the two to three months an A1C reflects. None of this means either number is wrong.
- Why does the A1C this page gives from my average differ from my lab A1C?
- The same reasons, in reverse. The page puts your average on the study's line, and your own relationship between glucose and A1C can sit off that line. Conditions that affect red blood cells or haemoglobin, such as anaemia or a haemoglobin variant, can move A1C without the glucose changing, which is why the ADAG study left out people who had them. When a lab A1C and a glucose average disagree, the lab result is the one a clinician works from, and the disagreement itself is worth mentioning to them.
- What is the difference between A1C in % and in mmol/mol?
- They are two units for the same test. The percentage is the NGSP unit, aligned to the DCCT trial, and is what US labs report. The IFCC unit, millimoles of glycated haemoglobin per mole of haemoglobin, comes from the international reference method; the IFCC recommended it in 2007, and many countries outside the US report it, often beside the percentage. The NGSP publishes the equations that link the two, and those are what this page uses.
- Why 28.7 and 46.7?
- They are the slope and intercept of the straight line the ADAG study fitted between A1C and the average glucose it measured, with continuous monitoring and fingersticks over about three months per person, published in Diabetes Care in 2008. Other formulas exist and give somewhat different numbers, so calculators that use them will disagree with this one.