A1C of 5.8% is an average glucose of 120 mg/dL

Estimated average glucose at A1C 5.8%

120 mg/dL

In mmol/L: 6.7 mmol/L
A1C entered: 5.8%
ADA range: the range the ADA defines as prediabetes

Putting 5.8% through the ADA equation gives an estimated average glucose of 120 mg/dL — the same figure expressed as 6.7 mmol/L. Laboratories reporting in IFCC units would print that A1C as roughly 40 mmol/mol.

Inside the ADA prediabetes range. It helps to know what the assay can resolve: NGSP certification permits about half a percentage point of variation, so movement smaller than that is not really movement.

A1C 5.8% is the single most-searched per-value A1C figure, at several thousand queries a month, and it falls just inside the ADA prediabetes range. The conversion to an average glucose is straightforward arithmetic; what it does not capture is variability, which is why continuous glucose data and a lab A1C sometimes tell slightly different stories.

Step up one tenth and the estimate moves by around 3 mg/dL: 5.9% gives about 123 mg/dL. Since a certified assay may itself vary by roughly 0.5 percentage points, a single tenth of movement between two tests is not really movement.

What this number cannot show is glucose variability. Continuous monitoring data reveals a pattern through the day that no single averaged figure can express.

Because the measurement covers a rolling window of months, retesting much sooner than three months largely re-measures the same period and adds little information.

The precision of the reported figure invites more confidence than it deserves. A result printed to one decimal place still carries the method tolerance around it, so treat the last digit as approximate.

Writing the value and its date somewhere you will find it again is worth more than reading another page about it, because the comparison is the part that carries information.

Convert a different A1C value

A1C to estimated average glucose, 5.3% to 6.3%
A1C	eAG (mg/dL)	eAG (mmol/L)
5.3%	105 mg/dL	5.8 mmol/L
5.4%	108 mg/dL	6 mmol/L
5.5%	111 mg/dL	6.2 mmol/L
5.6%	114 mg/dL	6.3 mmol/L
5.7%	117 mg/dL	6.5 mmol/L
5.8%	120 mg/dL	6.7 mmol/L
5.9%	123 mg/dL	6.8 mmol/L
6.0%	126 mg/dL	7 mmol/L
6.1%	128 mg/dL	7.1 mmol/L
6.2%	131 mg/dL	7.3 mmol/L
6.3%	134 mg/dL	7.4 mmol/L

Computed with eAG (mg/dL) = 28.7 x A1C - 46.7 (Nathan 2008). mmol/L is derived from mg/dL using the glucose factor 18.0182.

Estimated average glucose (ADAG / Nathan) — eAG (mg/dL) = 28.7 x A1C - 46.7. Verified 2026-07-30 against Nathan et al., Diabetes Care 2008 (ADAG study) (A population regression: in the ADAG study, individuals at the same A1C had meaningfully different mean glucose. The A1C assay itself carries about +/- 0.5 percentage points under NGSP certification.)

Official sources

Nathan DM et al., Translating the A1C Assay Into Estimated Average Glucose Values, Diabetes Care 2008 — The ADAG study that produced the eAG equation used on this page.
American Diabetes Association — Standards of Care in Diabetes — Where the A1C ranges quoted here are defined.
NGSP — A1C assay standardisation and certification criteria — Why two lab results half a point apart can mean nothing changed.

Frequently asked questions

An A1C of 5.8% equals what blood sugar?

120 mg/dL is the estimate, which is 6.7 mmol/L. It follows directly from the ADAG regression applied to 5.8%, and represents a population average rather than your personal figure.

Does 5.8% count as prediabetes or diabetes?

The ADA describes 5.8% as the range the ADA defines as prediabetes. That is a definition of a numeric range rather than a diagnosis: the ADA asks clinicians to confirm results with a repeat test and to read them alongside your history.

Is 5.8% a target?

This page converts a number rather than recommending one. Glycemic targets are individualised by a clinician according to your diagnosis, your other results and your risk of hypoglycaemia.

Would a different lab report the same 5.8%?

Not necessarily. Analytical methods vary between laboratories, and a point-of-care fingerstick can read differently from a venous draw, so comparisons are cleanest within one lab.

Get a quarterly nudge to re-check your numbers

Four emails a year: a reminder to re-take your numbers, and what changed in the guidance behind these tools. No spam, unsubscribe any time.

Double opt-in: you will get one confirmation email and nothing else until you click it.