What Your HbA1c Actually Tells You — and What It Doesn't

No single number in metabolic health gets quoted more often, or understood more shakily, than HbA1c. People recite it like a school grade — "mine's 39, what's yours?" — track it on apps, and occasionally despair over a two-point change that means nothing at all. It deserves better, because it's a genuinely clever measurement. It's also one that misleads in specific, knowable ways, and knowing them is the difference between reading the number and being ruled by it.

What it actually measures

HbA1c stands for glycated haemoglobin. Haemoglobin is the protein in red blood cells that carries oxygen; glucose in the blood sticks to it slowly and more or less irreversibly. The more glucose has been circulating, the more of your haemoglobin ends up coated. Measure the proportion that's glycated and you get an indirect read on average blood glucose.

Crucially, red blood cells live around three months. So HbA1c reflects roughly the previous two to three months of average glucose — weighted towards the more recent weeks, because newer cells dominate the population. This is its great strength: unlike a finger-prick that captures a single instant, HbA1c can't be gamed by skipping breakfast before the test. It's an average, and averages are hard to flatter.

A quick word on units. UK results are reported in mmol/mol; older results and most American sources use a percentage. The two scales describe the same thing: 48 mmol/mol equals 6.5%, the diabetes threshold; 42 mmol/mol equals 6.0%. If you're comparing your number to something you read online, check which scale you're looking at before drawing conclusions.

Where HbA1c misleads

Here's the part that rarely makes it onto the app. HbA1c is only as honest as the red blood cells underneath it — and any condition that changes how long those cells live, or how many there are, distorts the result.

If red cells are being destroyed or lost faster than usual — some anaemias, recent significant bleeding, conditions causing haemolysis — they have less time to accumulate glucose, and HbA1c can read falsely low. If red cells linger longer than usual, as in iron-deficiency anaemia or after the spleen is removed, they accumulate more glucose over their extended lifespan, and HbA1c can read falsely high. Pregnancy, certain kidney and liver conditions, and recent blood transfusions all complicate it further.

Then there are haemoglobinopathies — inherited variants such as sickle cell trait or thalassaemia — which can interfere with the measurement itself, depending on the laboratory method. This matters disproportionately for populations in whom these variants are common, and is a reason clinicians sometimes turn to other tests.

The honest summary: in a person with otherwise normal blood, HbA1c is excellent. In a person whose red cells are abnormal in number or lifespan, it can quietly lie — and the number on the page gives no hint that it's doing so. This is precisely why interpretation belongs with a clinician who knows your full blood picture, not with a tracking app.

HbA1c, fasting glucose, and CGM — different tools, different jobs

HbA1c is not the only way to look at glucose, and the alternatives answer different questions.

A fasting glucose is a single snapshot after not eating overnight. It's sensitive to the moment — useful, cheap, but easily nudged by a poor night's sleep or acute illness, and blind to what happens after meals.

Continuous glucose monitoring (CGM) — a sensor worn on the skin — shows the shape of glucose across days: the spikes after meals, the overnight pattern, the variability that an average hides entirely. Two people with an identical HbA1c can have very different lived experiences of glucose, one steady and one a rollercoaster, and only CGM reveals which. It's increasingly used in type 1 and insulin-treated type 2 diabetes, and is creeping into the wellness market, where its readings in people without diabetes are easy to over-interpret.

None of these is "best." HbA1c gives the long-run average; fasting glucose gives a snapshot; CGM gives the pattern. They complement rather than replace one another.

Why targets are personal, not universal

It's tempting to treat the diabetes threshold as a single finish line, but HbA1c targets in people who have diabetes are deliberately individualised. A younger person early in the condition may aim tighter, because decades of slightly raised glucose accumulate risk. An older person, or someone for whom medications carry a meaningful risk of dangerous low glucose (hypoglycaemia), may have a more relaxed target on purpose — because chasing a lower number can do more harm than the number itself.

In other words, the "right" HbA1c depends on the whole person: their age, their other conditions, their treatment, and how much a given target costs them to reach. A figure that's ideal for one person is wrong for another. That nuance is exactly the sort of thing a number-on-a-screen erases, and a clinician restores.

Practical takeaways

• HbA1c estimates average blood glucose over roughly the past two to three months — its strength is that a single meal can't flatter it.
• UK results use mmol/mol; 48 mmol/mol (6.5%) is the diabetes threshold, 42–47 is the prediabetes range — check the scale before comparing numbers.
• Anaemias, haemoglobinopathies, pregnancy, recent transfusion and some kidney or liver conditions can make HbA1c read falsely high or low, with no warning on the result.
• Fasting glucose (a snapshot) and CGM (the pattern) answer different questions; HbA1c targets are individualised, not one-size-fits-all.

What this doesn't mean

A single HbA1c does not define your health, and a small change between tests is usually noise rather than signal — the assay and biology both have wobble. Nor does a "good" number guarantee everything is fine, or a "high" one always mean diabetes; the reading has to be interpreted in context, sometimes alongside other tests.

When to seek medical advice

If your HbA1c falls in the prediabetes or diabetes range, discuss it with your GP rather than acting on the number alone. If you have a known blood condition, are pregnant, or have had a recent transfusion, mention it to whoever interprets the result, because it may change how the number should be read. Any decisions about glucose-lowering treatment belong with your own clinical team.

A closing thought

HbA1c is a small marvel — a way of reading three months of metabolism from a single tube of blood. Like any clever instrument, it rewards being understood and punishes being worshipped. Know what it measures, know where it lies, and it becomes what it should be: one useful sentence in a longer story, not the whole verdict.

Further reading and sources

• NICE NG28 — Type 2 diabetes in adults: management
• Diabetes UK — information on HbA1c and blood glucose monitoring
• World Health Organization — guidance on the use of HbA1c in the diagnosis of diabetes
• NHS — information on the HbA1c (blood sugar) test
• American Diabetes Association — Standards of Care in Diabetes (glycaemic assessment)