Metabolic health

An HbA1c of 5.7 to 6.4: what pre-diabetes means

By Dr. D. R. Gamalathge, MBBS · SLMC No. 36999 · Published 23 October 2026 · Reviewed 23 October 2026 · 6 min read

In short. An HbA1c between 5.7% and 6.4% (39 to 47 mmol/mol) is the pre-diabetes range. Below 5.7% (39 mmol/mol) is normal; 6.5% or more is the diabetes range. Pre-diabetes is not diabetes, and it is not a fixed state. The top of the range carries roughly double the risk of the bottom, and it is the stage at which lifestyle change has the strongest trial evidence behind it.

A report comes back with one line in bold. HbA1c 6.1%. Beside it, in smaller print: pre-diabetic range.

Nothing else on the page says what to do with that. No risk, no time frame, no plan. So it gets filed, and the next time anyone looks at it is the year the number crosses 6.5% and the word in the margin changes.

What an HbA1c of 5.7 to 6.4 means

HbA1c is the share of haemoglobin carrying glucose. Red cells live about 120 days and the recent weeks count most, so it reflects average glucose over the past two to three months, and it does not need a fast.

The American Diabetes Association places pre-diabetes at an HbA1c of 5.7% to 6.4% (39 to 47 mmol/mol), or a fasting glucose of 100 to 125 mg/dL (5.6 to 6.9 mmol/L). Diabetes starts at 6.5% or at 126 mg/dL.

UK guidance draws the line higher. NICE treats 6.0% to 6.4% (42 to 47 mmol/mol) as high risk. Same blood, two labels: a value of 5.8% is pre-diabetes on one report and unflagged on another, which is part of why the phrase on a printout means so little on its own.

The top of the range is not the bottom

A systematic review of prospective studies put five-year incidence of diabetes at 9% to 25% for an HbA1c of 5.5% to 6.0%, and 25% to 50% for 6.0% to 6.5%. The same label covers both. The risk does not.

So 5.8% and 6.3% are different conversations. The first is a reason to look at the rest of the panel. The second is a reason to act this year.

Locally, the range is crowded. In a 2025 Western Province survey (Katulanda et al.), 35.6% of adults were pre-diabetic and a further 27.8% already had diabetes, and most people in the first group feel entirely well.

What moves it

The Diabetes Prevention Program randomised 3,234 adults with raised glucose. Over an average of 2.8 years, a structured lifestyle programme, aiming at 7% weight loss and 150 minutes of activity a week, cut new diabetes by 58%. Metformin cut it by 31%. Roughly seven people had to follow the lifestyle programme for three years to prevent one case.

The Indian Diabetes Prevention Programme is the closer comparison for a South Asian body, and its numbers are smaller: 28.5% for lifestyle change and 26.4% for metformin over three years, with no extra benefit from combining them. Smaller, but real, in a population that develops diabetes at lower body weight and younger age.

Both trials chose participants by a glucose tolerance test, not by HbA1c. A later analysis of the Diabetes Prevention Program measured HbA1c instead and found lifestyle change still cut diabetes by 49%.

Neither trial was a supplement, a detox or a test kit. Both were food and walking, kept up for years, with the glucose measured again to see whether it was working.

What HbA1c can get wrong

HbA1c depends on the red cells carrying it, so anything that changes them changes the number.

Iron deficiency tends to push HbA1c up, which can place someone in the pre-diabetes range on a reading their glucose does not support. Thalassaemia trait and other haemoglobin variants can make the result unreliable in either direction, and in Sri Lanka around 2% to 3% of people carry beta-thalassaemia trait, many without knowing it.

That is why an HbA1c is read beside the full blood count (the red cell indices, MCV and MCH) and a fasting glucose drawn on the same morning. When the three agree, the HbA1c is likely to be reliable. When they disagree, the disagreement is the finding.

And HbA1c is late by design. It moves when glucose moves, and glucose is held normal for years by rising insulin. Fasting insulin and HOMA-IR show that earlier stage, before the HbA1c has started to climb.

Fasting and timing for an HbA1c test

HbA1c needs no fast. The fasting glucose read beside it does: eight to twelve hours, water only.

One value places you in a band. The second, three months later, shows which way it is going, because by then most of the red cells the first value was read from have been replaced. A fall from 6.2% to 5.9% after three months of change says more than either number alone.

Frequently asked questions

Is pre-diabetes a diagnosis of diabetes?

No. It describes a raised risk of developing type 2 diabetes, defined by an HbA1c of 5.7% to 6.4% or a fasting glucose of 100 to 125 mg/dL. Diabetes is diagnosed from 6.5% or 126 mg/dL, usually confirmed on a second test.

Can pre-diabetes go back to normal?

It can. In the Diabetes Prevention Program, a lifestyle programme aimed at 7% weight loss and 150 minutes of activity a week reduced new diabetes by 58% over about three years, and in the Indian Diabetes Prevention Programme lifestyle change reduced it by 28.5%. Some participants in both trials returned to normal glucose. Whether that has happened is confirmed by measuring again, not assumed.

Why does one report call 5.8% pre-diabetes and another call it normal?

Because two definitions are in use. The American Diabetes Association starts pre-diabetes at 5.7%, while UK guidance treats 6.0% to 6.4% as high risk. The blood is the same; the label depends on which threshold the report follows.

Can anaemia affect HbA1c?

Yes. Iron deficiency tends to raise HbA1c, and thalassaemia trait or other haemoglobin variants can make it unreliable. Reading it beside a full blood count and a fasting glucose taken on the same morning shows whether the three agree.

---

A printout that says pre-diabetic range and nothing else has measured the right thing and stopped one step short. The number needs a band, a reason it might be wrong, and a second reading three months on. HbA1c, fasting glucose and the full blood count are read together in both programmes, and the HbA1c is repeated at Month 3.

This article is general information, not individual medical advice. It does not replace assessment by your treating physician. If you have symptoms — chest pain, breathlessness or any urgent concern — seek medical attention immediately (Suwa Seriya 1990).