HbA1c mmol/mol to % Converter

HbA1c in two units: the short answer

To convert HbA1c from IFCC units (mmol/mol) to NGSP/DCCT units (%), use the master equation: % = 0.09148 × mmol/mol + 2.152. To go from % to mmol/mol: mmol/mol = (% − 2.152) × 10.929. An HbA1c of 48 mmol/mol is therefore 6.5%, and 7.0% is 53 mmol/mol. The converter above does this in both directions and shows where your result falls against the diagnostic thresholds.

Unlike most lab conversions on this site, HbA1c cannot be converted by multiplying by a single factor, because the two units come from different reference methods with different zero points. That is why rules of thumb such as "multiply by 11" give the wrong answer at some values. This guide explains why there are two units, how the master equation works, how to convert key thresholds, and what to keep in mind when comparing results from different countries or over time.

What HbA1c measures

HbA1c (glycated hemoglobin, or simply A1C) is the fraction of hemoglobin in red blood cells that has glucose irreversibly attached to the N-terminal valine of its beta chain. Because red cells live for around 120 days, HbA1c reflects average blood glucose over the previous two to three months, weighted towards the most recent weeks. It is used to diagnose diabetes and prediabetes and to monitor long-term glucose control.

Why there are two units

NGSP/DCCT units (%)

In the 1980s and 1990s, the landmark Diabetes Control and Complications Trial (DCCT) and the UK Prospective Diabetes Study (UKPDS) linked HbA1c values to the risk of diabetes complications. Their results were expressed as a percentage of total hemoglobin, using methods that measured HbA1c together with some other glycated components. In 1996, the National Glycohemoglobin Standardization Program (NGSP) was established to make all US laboratory methods give results traceable to the DCCT method, so that patients' results could be interpreted against the DCCT and UKPDS evidence.

IFCC units (mmol/mol)

The International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) then developed a more specific reference method that measures only the precisely defined glycated molecule. Because it excludes other glycated components, it gives lower values: an IFCC result expressed as a percentage would be about 1.5–2 percentage points lower than the NGSP result. To avoid confusion with the familiar percentages, the IFCC result is reported as mmol of HbA1c per mol of total hemoglobin – a different scale altogether.

Since 2009–2011, laboratories in the UK, most of Europe, Australia, New Zealand and many other countries have reported HbA1c in mmol/mol, often alongside the percentage during transition periods. The United States continues to report in %.

The master equation

Large comparison studies established a linear relationship between the two reference systems:

• NGSP (%) = 0.09148 × IFCC (mmol/mol) + 2.152
• IFCC (mmol/mol) = 10.929 × (NGSP % − 2.152)

The intercept of 2.152 is why a single multiplication factor cannot work. For example, multiplying 7% by 11 gives 77, but the correct IFCC value for 7% is 53 mmol/mol.

Mental shortcuts

For approximate mental conversion near the clinically important range:

• % to mmol/mol: subtract 2.15 from the percentage and multiply by 11. For 8%: (8 − 2.15) × 11 = 64 mmol/mol (exact: 63.9).
• mmol/mol to %: divide by 11 and add 2.15. For 58 mmol/mol: 58 ÷ 11 + 2.15 = 7.4% (exact: 7.46).

Another handy anchor: every 11 mmol/mol is about 1 percentage point. So 53 mmol/mol is 7.0%, 64 is 8.0%, 75 is 9.0% and 86 is 10.0%.

How to convert HbA1c step by step

1. Identify the unit. Values between about 4 and 15 are almost always %; values between about 20 and 140 are mmol/mol.
2. From mmol/mol to %, multiply by 0.09148 and add 2.152.
3. From % to mmol/mol, subtract 2.152 and multiply by 10.929.
4. Round % to one decimal place and mmol/mol to whole numbers, as laboratories do.

Worked examples

• 31 mmol/mol → 0.09148 × 31 + 2.152 = 5.0%.
• 39 mmol/mol → 0.09148 × 39 + 2.152 = 5.7% – the ADA lower limit for prediabetes.
• 42 mmol/mol → 6.0% – the UK threshold for "non-diabetic hyperglycaemia".
• 48 mmol/mol → 0.09148 × 48 + 2.152 = 6.5% – the diagnostic threshold for diabetes.
• 53 mmol/mol → 7.0% – a common treatment target.
• 64 mmol/mol → 8.0%.
• 75 mmol/mol → 9.0%.
• 7.5% → (7.5 − 2.152) × 10.929 = 58 mmol/mol.
• 11.0% → (11.0 − 2.152) × 10.929 = 97 mmol/mol.

Key thresholds in both units

HbA1c and average glucose

HbA1c can also be translated into an estimated average glucose (eAG) using the ADAG formula: eAG (mg/dL) = 28.7 × A1C (%) − 46.7. This formula uses NGSP percentages, so an IFCC result must be converted to % first. For example, 64 mmol/mol = 8.0% = an eAG of about 183 mg/dL (10.2 mmol/L). Our A1C calculator does this calculation and shows a full chart.

For a full A1C chart with average glucose, mmol/mol and the normal, prediabetes and diabetes ranges side by side, see our A1C calculator and chart.

Why converting correctly matters

People often encounter both units:

• when moving between countries or reading international guidelines and online material
• when comparing older results in % with newer results in mmol/mol from the same laboratory
• when CGM software reports a Glucose Management Indicator (GMI) in % while their laboratory reports mmol/mol
• in research papers, which may use either unit

A conversion error can lead to serious misunderstanding. For instance, someone who reads a UK target of "48" as a percentage might think it is unattainable; someone who reads "7" as mmol/mol would think it dangerously low. Using the master equation avoids this.

Precision and rounding

Laboratories typically report % to one decimal place and mmol/mol as whole numbers. Because 1 mmol/mol is about 0.09%, one decimal place in % (0.1%) corresponds to about 1.1 mmol/mol. Converting a rounded value and back can therefore produce a difference of one unit – for example 6.5% → 47.5 mmol/mol, which may be shown as 47 or 48. Diagnostic decisions near thresholds should rely on the laboratory's reported value in its own unit, and borderline values should be repeated.

Biological and analytical variation also matters. A change of about 5 mmol/mol (about 0.5%) between two tests is generally considered clinically meaningful; smaller changes may reflect normal variation.

Using HbA1c to diagnose diabetes

HbA1c has been accepted for diagnosing type 2 diabetes since 2009–2011 (ADA, WHO and later NICE), with advantages over glucose testing: no fasting is needed, it is less affected by short-term illness and stress, and it reflects longer-term glucose. Key principles:

• Confirmation is required. In someone without symptoms, a single result of 48 mmol/mol (6.5%) or more should be confirmed with a repeat test, unless another test (such as fasting glucose) is also clearly in the diabetic range.
• A normal HbA1c does not exclude diabetes. HbA1c is less sensitive than the oral glucose tolerance test, and some people with diabetes by glucose criteria have an HbA1c below 48 mmol/mol.
• It is not suitable in some groups, including children and young people, people with suspected type 1 diabetes, those with symptoms for less than two months, pregnant women, people taking medicines that rapidly raise glucose (such as corticosteroids or antipsychotics), people with acute pancreatic damage, and people with conditions affecting red cells.

HbA1c in pregnancy

For women with pre-existing diabetes, HbA1c before and in early pregnancy is a strong predictor of birth defects and other complications. NICE advises women with diabetes planning pregnancy to aim for an HbA1c below 48 mmol/mol (6.5%) if achievable without problematic hypoglycaemia, and strongly advises against pregnancy when HbA1c is above 86 mmol/mol (10%) because of the high risks. The ADA suggests a target below 6% (42 mmol/mol) during pregnancy if it can be achieved safely. In later pregnancy, HbA1c falls because of increased red cell turnover, so glucose monitoring – increasingly with CGM – is the main tool.

Point-of-care HbA1c tests

Some clinics and pharmacies use small point-of-care analysers that give an HbA1c result in a few minutes from a finger-prick sample. Several are certified by the NGSP and perform well, but their accuracy varies between devices, and many guidelines recommend that diagnosis be made with a laboratory test. Point-of-care devices may display either % or mmol/mol depending on their settings; check the unit before comparing with previous laboratory results.

Alternatives when HbA1c is unreliable

When conditions affecting red cells make HbA1c misleading, other measures help:

• Fructosamine – measures glycated serum proteins, mainly albumin, reflecting average glucose over about two to three weeks. It is affected by low albumin and protein loss.
• Glycated albumin – more specific than fructosamine; used in some countries, especially in kidney disease.
• Continuous glucose monitoring – provides average glucose, time in range and the Glucose Management Indicator (GMI), which is reported in % and can be converted to mmol/mol with the master equation.

Comparing results over time: a worked example

A woman with type 2 diabetes had HbA1c results of 8.4% in 2009 and 7.9% in 2010 from a laboratory that then switched to IFCC units, reporting 57 mmol/mol in 2012 and 61 mmol/mol in 2014. Converted to one unit, her results were 68, 63, 57 and 61 mmol/mol – or 8.4%, 7.9%, 7.4% and 7.7%. Her control improved until 2012 and then worsened slightly, a pattern that is hard to see while the units are mixed.

What affects HbA1c results

HbA1c can read higher or lower than true average glucose when red blood cell lifespan or hemoglobin structure is altered:

• Falsely low HbA1c: haemolysis, recent blood loss or transfusion, treatment of iron or B12 deficiency, erythropoietin therapy, pregnancy (second and third trimesters), some HIV medicines, advanced chronic kidney disease, and liver cirrhosis.
• Falsely high HbA1c: untreated iron-deficiency anaemia and B12 or folate deficiency (older red cells), splenectomy, and some forms of chronic kidney disease.
• Hemoglobin variants such as HbS (sickle cell trait), HbC, HbE and HbD, and raised fetal hemoglobin (HbF), can interfere with some methods; the NGSP website lists the methods affected.

In these situations, glucose-based tests, CGM data, fructosamine or glycated albumin may be more reliable.

HbA1c and the risk of complications

The reason HbA1c matters is its link to complications. In the DCCT, people with type 1 diabetes whose average HbA1c was about 7% (53 mmol/mol), compared with about 9% (75 mmol/mol), had roughly 50–75% fewer eye, kidney and nerve complications. Long-term follow-up (the EDIC study) showed that early good control continued to protect against complications – and reduced cardiovascular events – many years later, an effect called "metabolic memory" or the "legacy effect". The UKPDS showed similar benefits in type 2 diabetes. Because these landmark trials reported HbA1c in %, the familiar percentage thresholds remain closely tied to the evidence, and IFCC values are always interpreted through the master equation.

The relationship is continuous: there is no single HbA1c below which risk is zero. However, very aggressive lowering in some groups – particularly older people with long-standing type 2 diabetes and existing heart disease, as in the ACCORD trial, which targeted an HbA1c below 6.0% (42 mmol/mol) – increased mortality, largely because of hypoglycaemia and other harms. This is why targets are individualised.

Converting HbA1c in other systems

A few countries have used other national schemes. Japan historically used the JDS (Japan Diabetes Society) scale, which gave values about 0.4 percentage points lower than NGSP; since 2012, Japan has reported NGSP-equivalent values. Sweden used the Mono-S scale, about 1 percentage point lower than NGSP, before adopting IFCC units. If you encounter older results from these systems, check which scheme was used before converting.

Practical tips for reading your report

• Check which unit is printed next to the result, not just the number. Many reports show both.
• If your laboratory shows only mmol/mol and your care plan is written in %, convert using the master equation rather than a fixed factor.
• When comparing with previous results, use the same unit for all of them and look at the direction of change over at least two or three tests.
• Remember that a change of around 5 mmol/mol (0.5%) is usually needed before it is considered meaningful.

When to see a doctor

See a doctor if your HbA1c is 39 mmol/mol (5.7%) or above and you have not been assessed for prediabetes, or 48 mmol/mol (6.5%) or above. If you have diabetes, discuss results above your individual target, rising results, or results that do not match your home glucose readings. Seek urgent care for symptoms of very high blood glucose, such as extreme thirst, frequent urination, weight loss, vomiting or drowsiness.