A1C ranges sort results into four practical zones: normal below 5.7 percent, prediabetes from 5.7 to 6.4 percent, diabetes at 6.5 percent or higher, and a treatment goal usually below 7 percent for people already diagnosed. Each threshold is set by hemoglobin glycation, not by symptoms, which is why a person can sit at 6.3 percent for years feeling fine. The number reflects average glucose over roughly the prior 90 to 120 days, weighted toward the most recent month.
How Hemoglobin Glycation Produces the Number
Hemoglobin is the oxygen-carrying protein inside red blood cells. When glucose circulates in plasma, a fraction of it attaches irreversibly to the amino groups on hemoglobin, forming glycated hemoglobin, or HbA1c. The reaction is slow and concentration-dependent: higher average glucose drives more attachment, and the bond does not reverse when glucose falls. Because red cells live about 120 days before the spleen removes them, the percentage of glycated hemoglobin at any blood draw is a rolling weighted average.
The weighting is the key mechanism most explainers skip. Red cells produced last week carry glycation from last week's glucose. Cells made three months ago carry glycation from that era. Since newer cells dominate the population, roughly 50 percent of the A1C value reflects the most recent 30 days. The prior 30 to 60 days supply about 25 percent, and days 60 to 120 supply the remaining 25 percent.
In plain terms: A1C is a three-month report card, but the final month carries half the grade.
That weighting explains a common clinical surprise. A patient who improves diet in the four weeks before a test can drop A1C by 0.5 to 1.0 percentage point even if the earlier months were poor. Conversely, a short binge in the final weeks raises the value less than intuition suggests.
The Four A1C Ranges and Their Thresholds
The cut points below come from the American Diabetes Association's diagnostic standards, which have held these values stable since 2010. The same numbers appear in the 2017 ACC/AHA guideline that framed A1C as a screening and risk tool for cardiovascular disease.
| Range | A1C value | Estimated average glucose | What it signals |
|---|---|---|---|
| Normal | Below 5.7% | Under about 117 mg/dL | No diabetes; routine rescreen every 3 years |
| Prediabetes | 5.7% to 6.4% | About 117 to 137 mg/dL | Elevated risk; lifestyle or medication window |
| Diabetes | 6.5% and above | About 140 mg/dL or higher | Diagnostic if confirmed on a second test |
| Treatment goal | Usually below 7.0% | About 154 mg/dL | Individualized target for diagnosed patients |
The estimated average glucose column is not arbitrary. The A1C-Derived Average Glucose study, published in 2008, measured continuous glucose in 507 subjects and derived a formula now printed on lab reports: eAG in mg/dL equals 28.7 times A1C minus 46.7. An A1C of 7.0 percent therefore corresponds to an average glucose near 154 mg/dL. An A1C of 5.5 percent sits near 111 mg/dL.
The gap between the diagnostic threshold at 6.5 percent and the treatment goal below 7.0 percent is deliberate. Diagnosis demands a value far enough from normal to avoid false positives. Treatment tolerates a slightly higher ceiling because the risk curve flattens and because tight control carries hypoglycemia costs.
A1C Versus Fasting Glucose and Oral Glucose Tolerance
A1C is one of three accepted diagnostic tests. The other two measure glucose directly at a moment in time. They disagree more often than patients expect, and reconciling them is part of good care.
| Test | Diagnostic cutoff | What it captures | Main limitation |
|---|---|---|---|
| A1C | 6.5% | 90- to 120-day average | Skewed by red cell lifespan and hemoglobin traits |
| Fasting plasma glucose | 126 mg/dL | Single overnight baseline | Requires 8-hour fast; varies day to day |
| Oral glucose tolerance | 200 mg/dL at 2 hours | Response to a 75-gram glucose load | Cumbersome; poor reproducibility |
A person can hit 6.6 percent on A1C while fasting glucose reads 118 mg/dL, below the 126 cutoff. This discordance occurs in roughly 10 to 20 percent of screened adults depending on population. The 2017 ACC/AHA guidance treats A1C as the preferred screen for convenience, since no fasting is required, but recommends confirming any diagnostic result with a second test.
The reconciliation sentence that matters: an A1C of 6.5 percent, a fasting glucose of 126 mg/dL, and a two-hour glucose of 200 mg/dL are three routes to the same diagnosis, not three different diseases.
What Shifts A1C Without Changing True Glucose
This is the section that distinguishes a careful explainer from a thin one. A1C is a proxy, and several conditions break the proxy. In each case the number misstates the underlying average glucose, sometimes dangerously.
Conditions that shorten red cell survival lower A1C falsely. Hemolytic anemia, recent blood loss, sickle cell trait, and treatment with erythropoietin all push the value down while glucose may be high. Conditions that lengthen red cell survival or alter hemoglobin raise it falsely. Iron deficiency anemia, vitamin B12 deficiency, and chronic kidney disease all bias upward.
Hemoglobin variants are the most underappreciated trap. Roughly 7 percent of the world's population carries a structural hemoglobin variant, and common assays such as ion-exchange chromatography can read them incorrectly. A person with sickle cell trait may show an A1C that is 0.5 to 1.5 points off the true value in either direction. Labs now run variant interference checks, but ordering clinicians must request them.
Pregnancy is its own case. Red cell mass expands and turnover accelerates, so A1C falls by roughly 0.5 points in the second trimester even at unchanged glucose. Guidelines therefore prefer the oral glucose tolerance test for gestational diabetes screening rather than A1C.
Prediabetes as a Reversible Window
The 5.7 to 6.4 percent band is the highest-value range in the entire table because it is the only one with strong evidence of reversibility. The Diabetes Prevention Program, a landmark trial reported in 2002, randomized 3,234 adults with impaired glucose tolerance to intensive lifestyle change, metformin, or placebo. Lifestyle intervention cut progression to diabetes by 58 percent over about three years. Metformin cut it by 31 percent.
The lifestyle arm's targets were concrete: 7 percent body weight loss and 150 minutes of moderate activity per week. A person at 6.2 percent who loses 12 pounds and walks briskly for 30 minutes five days a week has a realistic chance of returning below 5.7 percent. The window is not permanent, but it is real.
Progression rates clarify urgency without panic. Roughly 5 to 10 percent of people with prediabetes convert to diabetes each year. Over a decade, a majority of untreated prediabetes either progresses or holds steady, and a meaningful minority reverts to normal. The range is a fork, not a verdict.
Timeline: How A1C Moves Across a Year
A1C is not static, and the time course of change matters for interpreting a trend. The table below shows what typical shifts look like at each stage, based on the glycation weighting and the DPP progression data.
| Time frame | Typical A1C change | What drives it |
|---|---|---|
| 2 to 4 weeks | 0.2 to 0.5 points | Diet change; newest red cells dominate |
| 8 to 12 weeks | 0.5 to 1.5 points | Weight loss, new medication, activity increase |
| 3 to 6 months | Full effect of a lifestyle program | Steady-state glycation |
| 1 year untreated prediabetes | Rise of 0.1 to 0.3 points | Gradual beta-cell decline |
| 3 years on DPP lifestyle arm | Net fall of 0.3 to 0.5 points | Sustained weight loss |
| Year 10 after diagnosis | Goal drift upward | Beta-cell function declines roughly 4 percent per year |
| Post-bariatric surgery | Fall of 2 to 3 points in 6 months | Rapid insulin sensitivity gain |
The beta-cell decline row deserves emphasis. Even well-controlled diabetes tends to need intensification over time because insulin-producing capacity erodes. A rising A1C in a stable patient is often biology, not noncompliance, and treating it as a moral failing delays the medication adjustment that actually helps.
Testing Frequency by Range
How often to test depends on where the value sits and what is changing.
- Normal, below 5.7 percent: rescreen every 3 years, or sooner with risk factors such as obesity or family history.
- Prediabetes, 5.7 to 6.4 percent: retest annually.
- Diagnosed diabetes at goal, below 7.0 percent: twice yearly.
- Diagnosed diabetes above goal or after therapy change: every 3 months until stable.
- Suspected lab interference or discordant results: repeat with a hemoglobin variant screen.
The three-month interval after a therapy change is not arbitrary. It matches the time needed for the glycation average to reach a new steady state, so testing earlier wastes information.
What to Actually Do With the Number
Actions should match the range, and the first action is always to confirm before acting.
- If a single test lands at 6.5 percent or above with no symptoms, request a second confirmatory test on a different day before accepting the diagnosis.
- If the value is 5.7 to 6.4 percent, ask for a referral to a structured lifestyle program rather than relying on general advice.
- If red cell conditions, pregnancy, or a known hemoglobin trait apply, ask the lab whether an alternative assay or a fructosamine test is needed.
- If you are diagnosed and above goal, track trend direction every three months rather than reacting to one reading.
- If A1C and home glucose readings disagree persistently, raise it explicitly; the discrepancy is diagnostic information.
In plain terms: treat one A1C as a data point, three consecutive values as a trend, and a trend as the thing you act on.
Limits of the A1C Range
No threshold in this table is a hard biological boundary. The 5.7 and 6.5 cut points were chosen by committee to balance sensitivity against false positives, and risk rises continuously across the whole range rather than jumping at a number. A person at 5.6 percent carries more risk than one at 5.0 percent even though both are called normal.
The ranges also say nothing about glucose variability. Two people at 7.0 percent can have wildly different daily swings, and the person with stable glucose has fewer hypoglycemic episodes. Continuous glucose monitoring now captures that dimension, and its time-in-range metric, typically above 70 percent, complements A1C rather than replacing it.
Finally, the numbers above describe adults. Pediatric and adolescent targets differ, and targets for older adults with limited life expectancy are looser, often below 8.0 percent, because the harm of hypoglycemia outweighs the benefit of tight control. An A1C range is a population guide. The right target is the one matched to the person holding the lab slip.