Frequently asked questions about A1C

A1C stands for hemoglobin A1c, also called HbA1c or glycated hemoglobin. It measures the percentage of hemoglobin — the oxygen-carrying protein in red blood cells — that has glucose permanently bound to it. Because red blood cells circulate for roughly 2–3 months before being replaced, the percentage reflects your average blood sugar exposure over that window, weighted most heavily toward the most recent month.

Under ADA criteria, a normal A1C is below 5.7%. A result from 5.7% to 6.4% is classified as prediabetes. 6.5% or higher, confirmed on a second occasion or with a second test method, is classified as diabetes. The UK and much of Europe report the same thresholds in IFCC units: below 39, 39–47, and 48 mmol/mol and above, respectively.

No. Fasting plasma glucose measures blood sugar at one specific moment after roughly 8 hours without food. A1C measures a rolling multi-month average regardless of when you last ate. Clinicians often use both together, since they answer different questions and can confirm each other when a diagnosis is borderline.

They describe the same underlying average glucose, just in different units. A1C is a percentage; eAG (estimated average glucose) restates that percentage in mg/dL or mmol/L — the same units your glucose meter already displays — using the ADAG study's conversion formula.

Some research has found small A1C differences across certain age groups and ethnic backgrounds even at matched average glucose, though the reasons aren't fully settled and the effect is generally modest. Clinicians typically factor this into how they interpret a borderline result rather than using a different threshold outright.

Adults without diabetes are generally screened every 1–3 years starting around age 35, or earlier with additional risk factors. With prediabetes, annual retesting is common. People managing diabetes and meeting their goals are typically tested twice yearly; every 3 months is standard after a treatment change or when goals aren't yet met, since that roughly matches red blood cell turnover time.

No. Because A1C reflects a multi-month average rather than a single moment, eating or drinking beforehand has no meaningful effect on the result, unlike fasting glucose or an oral glucose tolerance test.

It's a simple blood draw, either from a vein or, for some point-of-care devices, a finger-stick. The sample is analyzed using a certified method — commonly high-performance liquid chromatography, immunoassay, or boronate affinity chromatography — that isolates and quantifies the glycated fraction of hemoglobin.

Many at-home kits use methods certified through the same NGSP program that certifies lab equipment, and can be reasonably accurate when used correctly. That said, technique errors (an insufficient blood sample, for example) are more common outside a lab setting, so an unexpected or borderline home result is worth confirming with a clinician-ordered lab test.

A1C can be unreliable — reading falsely high or low relative to true average glucose — in people with certain hemoglobin variants, hemolytic or iron-deficiency anemia, chronic kidney disease, recent significant blood loss, or a recent transfusion, because these all affect how long red blood cells circulate. In these situations, clinicians often lean more heavily on direct glucose monitoring.

A meter average is built from however many finger-sticks or CGM readings you happen to take, which under-samples overnight hours and rapid post-meal spikes. A1C reflects a continuous, unbroken average across every hour of the day. The two numbers measuring the same underlying glucose exposure can legitimately differ even when both are correct.

A calculator like this one applies the published ADAG regression formula precisely, so the math itself is exact. What varies is how closely any one individual's true physiology matches the population the formula was built from — published validation puts individual variation at roughly ±15–20 mg/dL. Use it to build intuition and track trends, not as a replacement for a lab result.

No. This tool converts between validated units, but a diagnosis requires a laboratory-drawn A1C (or another validated test) interpreted by a qualified clinician, generally confirmed on a second occasion or with a second test method.

General guidance suggests below 7.0% for many non-pregnant adults with diabetes, with a tighter target sometimes appropriate for younger people early in their diagnosis, and a more relaxed target — sometimes above 8.0% — for older adults or those with a history of severe low blood sugar. There's no single number that fits everyone; targets are set individually with a care team.

No. Pregnancy changes red blood cell turnover and blood volume in ways that distort the usual A1C-to-glucose relationship, so gestational diabetes is diagnosed with a glucose challenge test or oral glucose tolerance test instead. Women with pre-existing diabetes who become pregnant typically follow an individualized target set by their care team.

NGSP reports A1C as a percentage and is standard in the US. IFCC reports the same measurement in millimoles of glycated hemoglobin per mole of total hemoglobin (mmol/mol) using a more analytically specific reference method, and is standard in the UK and much of Europe since around 2011. They describe the same blood chemistry through two different, linearly related numeric scales — see our formula page for the exact conversion.

Divide a mg/dL value by 18.0182 to get mmol/L, or multiply a mmol/L value by 18.0182 to get mg/dL. Our homepage calculator toggles between the two automatically.

An A1C of 7.0% corresponds to approximately 53 mmol/mol in IFCC units, and an estimated average glucose of about 154 mg/dL (8.6 mmol/L). See our full conversion chart for every value from 4.0% to 15.0%.