Updated 13 August 2026 · 7 min read
A1C is the number most clinicians use to judge how glucose control has been going. It is also the number most commonly misunderstood, because people treat it as a measurement of today when it is really a summary of the last few months.
Glucose in the bloodstream attaches to haemoglobin, the protein inside red blood cells that carries oxygen. This happens continuously and without any enzyme involved — the more glucose present, the more of it ends up bound. Red blood cells live around three months before being replaced. So if you measure what proportion of haemoglobin is glycated, you get a figure that reflects average glucose exposure over roughly that lifespan.
That is the whole idea. A1C is reported as a percentage: the share of haemoglobin carrying attached glucose.
| A1C | Category |
|---|---|
| Below 5.7% | Normal |
| 5.7% – 6.4% | Prediabetes |
| 6.5% or above | Diabetes range |
These are the cut-offs used by the American Diabetes Association. A diagnosis normally requires confirmation — either a repeat test or a second, different test — rather than resting on one result.
Treatment targets are a separate matter and are individualised. A target that suits a healthy 45-year-old is not automatically right for an 80-year-old with heart disease, where the risks of aiming too low can outweigh the benefit.
Because red cells turn over gradually, A1C responds slowly. Recent weeks weigh more heavily than older ones, but the number still cannot fully reflect a change made ten days ago.
This is also why A1C is usually measured a few times a year rather than monthly. Testing it more often than the biology moves adds cost without adding information.
Because A1C depends on red blood cells, anything that changes how long those cells live will shift the result — regardless of what glucose is actually doing.
Conditions that can push A1C falsely low generally involve red cells being replaced faster than usual, giving glucose less time to attach:
Conditions that can push A1C falsely high tend to involve cells living longer:
Separately, certain haemoglobin variants — including the sickle cell and haemoglobin C traits, which are more common in people of African, Mediterranean and South Asian descent — can interfere with some laboratory assays. Modern methods handle this better than older ones, but it remains a known source of error, and it is a reasonable thing to ask about if a result seems inconsistent with your daily readings.
Advanced kidney disease also complicates interpretation for several overlapping reasons.
If your meter consistently shows one story and A1C tells another, that mismatch is worth raising rather than explaining away. The possibilities include a distorting condition from the list above, an assay issue, a meter calibration problem, or a testing pattern that systematically misses the highs — for example, only ever testing fasting and never after meals.
An A1C that looks acceptable can also sit on top of large swings in both directions that average out to a reasonable-looking figure. This is one of the reasons continuous glucose monitoring has become more common: it shows the variability that a single averaged number hides.
A1C is a useful summary and a poor snapshot. It reflects roughly three months, it moves slowly by design, and it can be distorted by anything that changes red blood cell turnover. Read alongside daily readings it is informative. Read alone, it can mislead in both directions.
← All articles