Blood Sugar to A1C Calculator

The one formula that connects them

A1C and average blood sugar are two views of the same thing. A1C measures the share of your haemoglobin that has glucose stuck to it, which reflects roughly the last two to three months of blood sugar. Average glucose is what you see when you add up your meter readings. The relationship between them was worked out in a large international study and is now the standard used by the American Diabetes Association.

The formulas are:

• Average glucose to A1C: A1C (%) = (average glucose in mg/dL + 46.7) ÷ 28.7
• A1C to average glucose: average glucose (mg/dL) = 28.7 × A1C − 46.7

That is the whole calculation. The rest is unit conversion: divide mg/dL by 18.0182 to get mmol/L, or multiply mmol/L by 18.0182 to get mg/dL.

This is a reference estimate, not medical advice, and it is not a substitute for a laboratory A1C. Everything runs in your browser, the numbers you type are never uploaded or stored.

Worked example: an average of 154 mg/dL

Say two weeks of readings average out to 154 mg/dL.

A1C = (154 + 46.7) ÷ 28.7 = 200.7 ÷ 28.7 = 6.99%, which rounds to 7.0%. That is not a coincidence: 154 mg/dL is exactly the estimated average glucose that goes with a 7% A1C, which is why the ADA uses it as a common target.

In mmol/L, 154 ÷ 18.0182 = 8.5 mmol/L.

Now check the reverse. Take a lab A1C of 6.5%: average glucose = 28.7 × 6.5 − 46.7 = 186.55 − 46.7 = 139.9 mg/dL, or 7.8 mmol/L. And an A1C of 5.7% works out to 116.9 mg/dL, or 6.5 mmol/L. Those two numbers, 139.9 and 116.9, sit right on the boundaries that labs use for the diabetes and prediabetes ranges.

A1C to average glucose, in both units

This table is the whole conversion in one place. Use the mg/dL column if you are in the United States, and the mmol/L column if your meter reports in the units used in most other countries.

Notice the spacing. Each 1% of A1C is worth about 28.7 mg/dL, so the steps are even. Moving from 6% to 7% is a 28.7 mg/dL change in your daily average, a big shift in how you feel and in long-term risk.

What the bands mean

The reference bands below are the standard ones used on a laboratory A1C. They are useful context, but they are not a diagnosis, and they do not strictly apply to a number you calculated from home readings.

Two cautions. First, the diagnosis is made from a lab test, usually confirmed on a second occasion, not from a calculator. Second, the 6.5% threshold is a population cut-off, not a magic line; risk rises steadily as the number rises rather than jumping at exactly 6.5.

Targets are personal. A commonly quoted goal for adults with diabetes is below 7%, and below 6.5% for some, but age, how long you have had diabetes, kidney function and your risk of hypoglycaemia all move the right number up or down. That call belongs to your clinician.

Why your estimate and your lab result will not match

The formula is accurate for populations. Individual people vary, and a handful of conditions move A1C away from what your average glucose would predict.

• Anaemia and iron deficiency. Low iron can push A1C up or down depending on the type, without any change in blood sugar.
• Haemoglobin variants. Sickle cell trait and other variants can make some A1C tests read inaccurately. Laboratories usually screen for this.
• Pregnancy. Red blood cells turn over faster, which lowers A1C, so pregnancy uses its own glucose thresholds instead.
• Recent transfusion or blood loss. A transfusion replaces old red cells with new ones and can distort the result for weeks.
• Kidney disease and some medications. Both can shift the reading.
• Lab-to-lab variation. The same sample can differ by about 0.2 to 0.3 percentage points between instruments.

There is also a testing problem on your side of the equation. A meter average only reflects the moments you tested. If you check fasting every morning and never after a meal, your average will look better than your A1C, because the post-meal spikes are missing from the data. For background on how A1C is measured and interpreted, the American Diabetes Association's A1C page and the National Institute of Diabetes and Digestive and Kidney Diseases are both worth reading, and the CDC's diabetes pages cover the testing basics in plain language.

How to build an average that is worth converting

Garbage in, garbage out. A conversion is only as good as the average you feed it, and most home averages are biased.

• Cover the whole day. Include a fasting reading, a reading before a meal, and at least one about two hours after a meal. Post-meal numbers are the ones people skip and they are the ones that move the estimate most.
• Two weeks minimum. Ten readings is a snapshot. Twenty-eight or more across different times of day is an average.
• Average by day, then average the days. If you test five times on Monday and once on Tuesday, a flat average of all readings over-weights Monday. Work out each day's mean first, then average those.
• Leave out the outliers you know are wrong. A reading taken with unwashed hands after peeling an orange can be wildly high. If you know why a number is off, exclude it rather than letting it drag the average.
• Use a CGM average if you have one. Continuous monitors capture overnight and post-meal glucose that fingersticks miss, and they report an average directly.

One more note on what a difference looks like in practice. If your two-week average is 180 mg/dL, the formula gives an A1C near 7.9%. Get that average down to 154 mg/dL and the estimate drops to about 7.0%. Same person, same formula, one percentage point of A1C, and 26 mg/dL of daily average between them.

Finally, and this is the part that matters most: this tool is a reference estimate, not medical advice. It cannot tell you whether you have diabetes, whether a medication is working, or what your target should be. Do not change any medication based on a number from a web page. Take the estimate to your next appointment and let your clinician compare it with a lab A1C.