To convert glucose to A1C, start from your average glucose, not from any single reading, and use the reverse of the American Diabetes Association formula: A1C percent equals your average glucose in mg/dL plus 46.7, divided by 28.7. An average glucose of 154 mg/dL therefore converts to about 7.0 percent, an average of 126 mg/dL converts to about 6.0 percent, and an average of 100 mg/dL converts to about 5.1 percent. The same formula runs the other way too: estimated average glucose equals 28.7 times A1C minus 46.7, so an A1C of 7.0 percent corresponds to an estimated average glucose of about 154 mg/dL. This guide explains the formula step by step, works the examples in full, gives the conversion table from 80 to 300 mg/dL, explains why a meter or CGM average can differ from a laboratory A1C, and explains why a single fasting glucose value cannot be converted into an A1C at all. Nothing here sets a target for you, diagnoses anything, or replaces a clinician: a converted A1C is an estimate to help you understand a number, and what that number means for you is a conversation with your clinician.

The reverse ADA formula, step by step

Converting average glucose to A1C starts with a formula most people meet in the opposite direction. The American Diabetes Association publishes estimated average glucose, often shortened to eAG, from A1C as: eAG in mg/dL equals 28.7 times A1C percent minus 46.7. That formula came out of the A1C-Derived Average Glucose study, which compared laboratory A1C results with frequent glucose measurements in adults and fitted a straight line through them. To convert glucose to A1C you simply reverse that line. Add 46.7 to both sides and divide by 28.7, and you get: A1C percent equals average glucose in mg/dL plus 46.7, divided by 28.7. The two formulas are the same relationship read in opposite directions, which is why converting back and forth lands you close to where you started, apart from rounding. An A1C of 7.0 percent gives an eAG of 28.7 times 7.0 minus 46.7, which is 154.2 mg/dL, and feeding 154 mg/dL back through the reverse formula gives 6.99 percent, which rounds to 7.0 percent. The small wobble of a tenth of a percent is rounding, not a contradiction.

Three features of the formula are worth noticing before you use it. First, every 28.7 mg/dL of average glucose is worth exactly one percentage point of A1C, so every 2.87 mg/dL is worth about a tenth of a point. That sensitivity tells you how precise a converted result can honestly be: an average that is off by 10 mg/dL moves the converted A1C by about 0.35 points, which is why a converted value should always be quoted to one decimal place and treated as approximate. Second, the formula assumes the input is a genuine average over a long window, roughly the previous two to three months, because that is the window an A1C reflects. An average over a single good week is not the same input, and the formula cannot correct for that. Third, the formula is a population average line, fitted across many people. Any one person can sit a little above or below that line, which is one more reason a converted A1C is an estimate. The A1C calculator runs this same calculation in both directions, so you can check a conversion there without doing the division by hand, but the arithmetic below shows every step so you never have to trust a black box.

The IFCC figure that appears beside A1C in many laboratories is the same result in different units, millimoles per mole rather than percent. The conversion is: IFCC in mmol/mol equals 10.929 times A1C percent minus 2.15. An A1C of 7.0 percent is therefore 10.929 times 4.85, which is 53 mmol/mol. An A1C of 6.0 percent is 42 mmol/mol, and an A1C of 8.0 percent is 64 mmol/mol. If your laboratory reports only the IFCC number, convert it to percent first by reversing that formula, or read the IFCC column of the table in the next-but-one section, which lists both side by side. Do not mix the two: 53 mmol/mol and 5.3 percent are not the same thing, and confusing them is one of the easiest errors in this whole topic. Throughout this article, A1C means the familiar NGSP percent figure unless the IFCC unit is stated explicitly.

Worked examples: from average glucose to A1C

The worked examples below are the ones people actually ask about, and each one is shown with its full arithmetic so you can check it on a pocket calculator. Whether you describe it as average glucose to A1C or average blood sugar to A1C, the calculation is the same, and it is the same blood sugar to A1C conversion in either wording. They are also the anchor points of the conversion table: if you remember 100, 154 and 212, you can place almost any average roughly, because 100 mg/dL is about 5.1 percent, 154 mg/dL is about 7.0 percent, and 212 mg/dL is about 9.0 percent.

Work the first example slowly. An average glucose of 100 mg/dL plus 46.7 is 146.7. Divide 146.7 by 28.7 and you get 5.11, which rounds to 5.1 percent. For 126 mg/dL: 126 plus 46.7 is 172.7, divided by 28.7 is 6.02, which rounds to 6.0 percent. For 140 mg/dL: 140 plus 46.7 is 186.7, divided by 28.7 is 6.51, which rounds to 6.5 percent. For 154 mg/dL: 154 plus 46.7 is 200.7, divided by 28.7 is 6.99, which rounds to 7.0 percent. For 183 mg/dL: 183 plus 46.7 is 229.7, divided by 28.7 is 8.00, exactly 8.0 percent to two decimal places. For 212 mg/dL: 212 plus 46.7 is 258.7, divided by 28.7 is 9.01, which rounds to 9.0 percent. Notice how the rounding behaves: the unrounded values 6.99 and 9.01 are not errors, they are the formula telling you that the neat anchor points are rounded summaries of a continuous line. A result of 6.99 percent and a result of 7.0 percent are the same result to the precision this estimate can support.

Before any of those examples can be used, you need a genuine average, and averaging readings is itself a step worth doing carefully. Suppose a week of readings, taken at mixed times of day, reads 98, 132, 145, 118, 160, 121 and 139 mg/dL. Add them: 98 plus 132 is 230, plus 145 is 375, plus 118 is 493, plus 160 is 653, plus 121 is 774, plus 139 is 913. Divide by 7 readings and the average is 130.4 mg/dL. Apply the formula: 130.4 plus 46.7 is 177.1, divided by 28.7 is 6.17, which rounds to 6.2 percent. That is a correct conversion of that week’s average, but it is not an A1C, because one week is not two to three months, and seven readings at mixed times are not a continuous record. Treat it as what it is: a converted weekly average, useful for following a trend, not a substitute for a laboratory A1C. The longer and more evenly spread the readings behind the average, the closer the conversion comes to what a laboratory A1C would show, and the section on meter and CGM averages explains exactly where the remaining gap comes from.

The full conversion table: 80 to 300 mg/dL

The table below gives the conversion for every average glucose from 80 to 300 mg/dL in steps of 10 mg/dL, with the same average in mmol/L and the IFCC value in mmol/mol alongside. Use it to look up an average without dividing, to check a calculator result, or to see at a glance how far apart two averages sit in A1C terms. Every figure in the table was computed from the same reverse formula, A1C equals average glucose plus 46.7 divided by 28.7, with mmol/L from dividing by 18.0156 and IFCC from 10.929 times A1C minus 2.15, and then rounded: A1C to one decimal place, mmol/L to one decimal place, IFCC to the nearest whole number.

Read the table as a slope, not just as a lookup. Moving up one row, 10 mg/dL, moves A1C by about 0.35 points, so adjacent rows differ by 0.3 or 0.4 points after rounding. That is why an average of 145 mg/dL, which falls between the 140 and 150 rows, converts to 6.7 percent: 145 plus 46.7 is 191.7, divided by 28.7 is 6.68, which rounds to 6.7, sitting neatly between the 6.5 percent of the 140 row and the 6.9 percent of the 150 row. If your average falls between rows, interpolate this way or run the formula directly rather than rounding to the nearest row, because rounding the input to the nearest 10 mg/dL first can shift the result by up to 0.2 points. The mmol/L column follows the same slope in metric units: 10 mg/dL is about 0.6 mmol/L, so the column climbs by roughly 0.5 to 0.6 at each step. The IFCC column climbs by about 4 mmol/mol per row, which is the same 0.35-point step expressed in IFCC units, since one A1C percentage point is about 10.9 mmol/mol.

Two cautions belong with the table. First, the low end and the high end are both extrapolations in the same sense: the formula is a straight line fitted to the range of values in the original study, and values far outside the everyday range should be treated with extra caution and discussed with a clinician rather than read off a table. An average of 80 mg/dL converts to 4.4 percent and an average of 300 mg/dL converts to 12.1 percent by the same arithmetic, and both figures are arithmetically correct applications of the formula, but neither, on its own, tells you anything about a person’s health, and neither should be used to judge one. Second, the table converts averages, and only averages. If you look up a single reading in this table, you will get a number that looks like an A1C and is not one. The next sections explain why that distinction matters more than any other point in this article.

Why a meter or CGM average differs from a lab A1C

A meter average and a laboratory A1C can disagree by half a point or more without either one being wrong, because they measure different things over different windows. A laboratory A1C measures the proportion of haemoglobin in red blood cells that has glucose attached to it. Red blood cells live for roughly 120 days, so the result reflects glucose over roughly the previous two to three months, weighted more heavily toward recent weeks, since older cells have already been replaced. A finger-prick meter average, by contrast, reflects only the moments you tested. If you test mostly before breakfast and before meals, your average is built from fasting and pre-meal values and contains few after-meal peaks and no overnight values at all, so it will tend to sit below a laboratory A1C that silently includes every hour of every day. If you test mostly after meals, the opposite bias appears. A continuous glucose monitor, or CGM, closes most of that gap because it records day and night, but it does not close it completely: sensors have their own error, wear periods have gaps, and sensor fluid lags behind blood glucose when levels are changing quickly.

Meter error itself is a second, separate source of difference, and it is larger than many people expect. Under the ISO 15197 standard for blood glucose meters, at least 95 percent of results at or above 100 mg/dL must fall within plus or minus 15 percent of a laboratory reference, and results below 100 mg/dL within plus or minus 15 mg/dL. A true glucose of 150 mg/dL can therefore read anywhere from about 128 to 173 mg/dL on a meter that meets the standard, and an average of many such readings will be closer to the truth than any single reading, but not exact. Strip storage, an unwashed hand, a low battery, or an expired strip can add further error on top of that, which is why technique matters as much as arithmetic. None of this makes a meter useless: averaged over many readings, a meter gives a genuinely useful trend. It does mean that converting a meter average to A1C and expecting it to match a laboratory result to the tenth of a percent is expecting more precision than the inputs carry.

A third source of difference is biological rather than technical, and it is the one that most often surprises people. Anything that changes red blood cell lifespan or haemoglobin changes the relationship between average glucose and A1C, even when glucose itself has not changed. Anaemia, recent blood loss or transfusion, some haemoglobin variants, and advanced kidney disease are among the conditions known to shift A1C relative to glucose, in either direction depending on the condition. That is why clinicians do not treat a converted A1C and a laboratory A1C as interchangeable, and why a persistent gap between your converted value and your laboratory value is worth raising with your clinician rather than resolving by picking whichever number you prefer. The gap itself is information: it may point to a testing pattern that misses part of the day, or it may point to something about red blood cells that only a laboratory work-up can sort out. Either way, the right response to a gap is a conversation, not a recalculation. This article gives you the conversion so you can have that conversation with numbers in hand, not so you can replace it.

Units, time windows, and the single-reading trap

Unit conversion is the smallest step in this topic and the one most worth getting exactly right, because a unit error is a factor-of-eighteen error. In the United States glucose is reported in milligrams per decilitre, mg/dL. In most of the rest of the world it is reported in millimoles per litre, mmol/L. To go from mg/dL to mmol/L, divide by 18.0156. To go from mmol/L to mg/dL, multiply by 18.0156. An average of 126 mg/dL is therefore 126 divided by 18.0156, which is 7.0 mmol/L. An average of 154 mg/dL is 8.5 mmol/L. An average of 100 mg/dL is 5.6 mmol/L. In the other direction, 7.0 mmol/L times 18.0156 is 126.1 mg/dL, and 8.5 mmol/L times 18.0156 is 153.1 mg/dL, which is 154 mg/dL to the precision of a one-decimal-place metric figure. Many people use 18 as a shortcut, and for everyday purposes it is close: 126 divided by 18 is exactly 7.0, and 154 divided by 18 is 8.6 rather than 8.5. The shortcut is fine for a rough sense, but the table in this article and the calculator use 18.0156, so if you check a table value with the shortcut and land a tenth out, that is the reason. Whichever factor you use, convert the units before you apply the A1C formula, because the formula expects mg/dL: feeding a mmol/L figure into a formula built for mg/dL gives a meaningless answer.

The time window deserves the same care as the units. An A1C reflects roughly two to three months because that is roughly how long the red blood cells it measures have been circulating, with recent weeks weighing more heavily than months ago, since the oldest cells are continuously replaced. A converted value inherits that window: it is only a fair stand-in for an A1C if the average behind it covers a similar span, reasonably evenly. A month of diligent readings is better than a week, and three months of readings at varied times of day is better still. There is no magic number of readings that makes a short window into a long one, and a single intensive weekend of testing, however careful, cannot stand in for months of glucose exposure. Equally, an old average does not stay relevant: an average from six months ago describes six months ago, and converting it today gives you an estimate of what an A1C would have been then, not what it is now.

That brings us to the single-reading trap, which is the most common misuse of this conversion and the one this article most wants to prevent. A fasting glucose reading is a single measurement taken after an overnight fast. It describes one moment, under one condition, at the quietest point of the day. An A1C describes an average over months, including every meal, every overnight stretch, and every rise and fall in between. No formula can recover a months-long average from one moment, because the information simply is not in that one number: two people with the same fasting value can have very different averages once meals, activity, and overnight patterns are included, and one person’s fasting value on two different mornings can differ while their average has barely moved. So a fasting glucose to A1C conversion, as a concept, does not work, and any tool or table that appears to offer one is either converting a fasting value as if it were an average, which it is not, or hiding an assumed relationship between fasting and average glucose that may not hold for you. If all you have is a fasting reading, the honest use of it is as a single data point to discuss with your clinician, alongside other readings taken at other times of day. If you want a converted A1C, collect an average first, over weeks to months and across times of day, and then convert that average. The formula is short, the table is complete, and neither can supply the missing months.

Sources: American Diabetes Association estimated average glucose (eAG) formula, eAG (mg/dL) = 28.7 x A1C% – 46.7, from the A1C-Derived Average Glucose (ADAG) study, reversed here as A1C% = (average glucose + 46.7) / 28.7; IFCC conversion IFCC (mmol/mol) = 10.929 x (A1C% – 2.15); unit conversion mg/dL to mmol/L dividing by 18.0156; ISO 15197 meter accuracy standard (95% within 15% at or above 100 mg/dL, within 15 mg/dL below 100 mg/dL). A1C Calculator hub (CalcFolk). This article is general information only, not medical advice, and does not diagnose, treat, or set targets for any person. Adults and general readers only; no paediatric targets are given.