Haemoglobin A1c (HbA1c)

What is HbA1c?

HbA1c, or glycated haemoglobin, is the fraction of haemoglobin in your red blood cells that has glucose irreversibly attached to it, expressed as millimoles of glycated haemoglobin per mole of total haemoglobin (mmol/mol) or as a percentage (%).

Glucose in the bloodstream attaches to haemoglobin in a slow, non-enzymatic reaction. The higher the average glucose concentration, the greater the proportion of haemoglobin that becomes glycated. Because the reaction is irreversible and the average red blood cell lives about 120 days, the measured fraction is a running average of glycaemia over roughly the preceding 8–12 weeks — weighted towards the most recent month, which contributes around half the result.

That averaging is the whole point of the test. A fasting glucose is a photograph taken at one instant, easily distorted by a poor night's sleep, a late meal or the stress of the blood draw. HbA1c is closer to a time-lapse, and it needs no fasting, which is why it can be added to any panel at any hour.

Two unit systems are in use. The IFCC system reports mmol/mol and is the reference method internationally; the NGSP system reports a percentage and derives from the DCCT trial. Australian laboratories report both, with mmol/mol as the primary figure and % in parentheses. American laboratories report the percentage only.

Why HbA1c is tested

HbA1c is tested to diagnose type 2 diabetes, to identify people at high risk before diabetes develops, and to monitor glycaemic control over time in people already diagnosed.

As a screening test its strength is stability. It requires no fasting, it is not perturbed by a single meal, and its within-person biological variation is small, which makes a rising trend across annual tests genuinely meaningful rather than noise.

There is a second reason HbA1c belongs on a men's hormone panel specifically. Insulin resistance suppresses SHBG production in the liver and increases aromatisation of testosterone to oestradiol in adipose tissue, so worsening glycaemia and falling testosterone frequently travel together. Interpreting a low testosterone result without the metabolic context beside it misses the more actionable finding in a large proportion of men aged 35–55.

In Australia, HbA1c is also used with the AUSDRISK risk assessment tool: a high AUSDRISK score is the usual prompt to check HbA1c even in the absence of symptoms.

What a high HbA1c means

A high HbA1c means average blood glucose over the previous two to three months has been elevated — at 48 mmol/mol (6.5%) or above it meets the Australian diagnostic threshold for diabetes, and between 42 and 47 mmol/mol (6.0–6.4%) it identifies high risk of developing diabetes.

What to check next: an elevated HbA1c used for diagnosis must be confirmed, either by a repeat HbA1c or by a fasting plasma glucose, on a separate occasion in an asymptomatic person. Sensible companion markers are fasting glucose, fasting insulin (allowing HOMA-IR to be calculated), a lipid profile including triglycerides and HDL, liver enzymes with attention to fatty-liver patterns, and a full blood count plus ferritin to make sure a red cell abnormality is not distorting the reading.

• Insulin resistance and type 2 diabetes — the most common explanation, particularly with central adiposity, hypertension or dyslipidaemia.
• Type 1 diabetes or latent autoimmune diabetes in adults, more likely with rapid onset, weight loss and younger age.
• Steroid-induced hyperglycaemia from glucocorticoid therapy.
• Conditions that lengthen red cell survival — iron deficiency anaemia, B12 deficiency, post-splenectomy — which raise HbA1c without a corresponding rise in average glucose.
• Chronic kidney disease and some haemoglobin variants, which can bias the assay in either direction depending on method.

What a low HbA1c means

A low HbA1c — below about 4.0% (20 mmol/mol) — is uncommon and, in someone not taking glucose-lowering medication, more often indicates shortened red blood cell survival than genuinely low average glucose.

Because HbA1c depends on red cells surviving long enough to accumulate glucose, anything that shortens their lifespan lowers the measured value: haemolytic anaemia, recent significant blood loss or blood donation, hypersplenism, and some haemoglobinopathies. Recent transfusion, pregnancy and erythropoietin therapy have the same effect.

In people using glucose-lowering therapy, a low HbA1c can reflect frequent hypoglycaemia averaging down an otherwise variable glucose profile — a pattern that continuous glucose monitoring reveals and HbA1c conceals. In either case a low result is investigated rather than celebrated, usually with a full blood count, reticulocytes, ferritin and, where relevant, haemoglobin electrophoresis.

HbA1c reference ranges and conversion

HbA1c reference ranges are defined by fixed diagnostic thresholds rather than by age or sex, with mmol/mol (IFCC) as the primary Australian unit and % (NGSP) reported alongside it.

The conversion between the two systems is exact and worth knowing, because unit confusion is the single most common error in online HbA1c content. The IFCC master equation is: IFCC (mmol/mol) = (NGSP % − 2.15) × 10.929. Working it through: 5.7% gives (5.7 − 2.15) × 10.929 = 38.8, which rounds to 39 mmol/mol. 6.0% gives 42.1, which rounds to 42 mmol/mol. 6.5% gives 47.6, which rounds to 48 mmol/mol. Sources that state 5.7% = 42 mmol/mol have conflated the American prediabetes cut-off with the Australian high-risk cut-off; they are different numbers describing different thresholds.

What affects the result — confounders to rule out

HbA1c is distorted by anything that changes red blood cell lifespan or interferes with the assay, and in those situations it should be replaced by fasting glucose or an oral glucose tolerance test rather than reinterpreted.

• Iron deficiency anaemia and B12 deficiency — both prolong red cell survival and can raise HbA1c by several mmol/mol without any change in average glucose.
• Haemolytic anaemia, recent blood loss, blood donation or transfusion — all shorten effective red cell exposure and lower HbA1c.
• Haemoglobin variants (HbS, HbC, HbE, thalassaemia) — can bias results in either direction depending on the assay method; laboratories flag this where detected.
• Chronic kidney disease and dialysis — shortened red cell survival plus carbamylation interference make HbA1c unreliable.
• Pregnancy — red cell turnover increases and different diagnostic criteria apply; HbA1c is not used to diagnose gestational diabetes.
• Recent acute hyperglycaemia — HbA1c lags real change by weeks, so it will underestimate a glucose rise that began a fortnight ago.
• Point-of-care versus laboratory testing — only a laboratory-measured, NGSP/IFCC-standardised HbA1c is accepted for diagnosis in Australia.

Which FORM panels include HbA1c

HbA1c is included from the Pro panel upward and can be added to any custom panel, always reported alongside the glucose and insulin markers needed to interpret it.

An HbA1c in isolation tells you where average glucose has been, not why. FORM panels pair it with fasting glucose and, at higher tiers, fasting insulin — so that HOMA-IR can be calculated and insulin resistance detected years before HbA1c itself moves. Lipids, liver enzymes and hs-CRP round out the metabolic picture, and a doctor reviews the combination with you rather than sending a number and leaving you to interpret it.