A1c Calculator from Average Glucose

What HbA1c measures

Hemoglobin A1c, written HbA1c or A1C, is the percentage of hemoglobin with glucose attached. Red blood cells circulate for about 120 days, so an A1C result summarizes glucose exposure over roughly two to three months. It is a weighted average: glucose during the most recent month has more influence than glucose early in the period.

A meal, one difficult day, or a single exercise session will not determine the result.

A1C requires a whole-blood sample but does not require fasting or a timed glucose drink. The 2026 American Diabetes Association Standards identify it as a diagnostic option for nonpregnant people and as the main laboratory measure for ongoing glycemic assessment. A1C cannot show whether the same average came from steady values or repeated highs and lows. Laboratories should use an NGSP-certified method traceable to the Diabetes Control and Complications Trial reference assay, which links current results to the evidence relating A1C to long-term outcomes. A home kit or office result may support monitoring, but diagnosis requires an appropriately approved method used in a qualified laboratory setting.

[1][2][5][6]

Formula used by the calculator

This calculator applies the regression from the A1C-Derived Average Glucose study. To estimate A1C from mean glucose in mg/dL, use A1C (%) = (mean glucose + 46.7) / 28.7. The inverse equation is estimated average glucose, or eAG, in mg/dL = 28.7 × A1C − 46.7. For glucose in mmol/L, first convert to mg/dL by multiplying by 18.0182, or convert the resulting eAG back by dividing by 18.0182.

The ADAG investigators compared central laboratory A1C with about 2,700 glucose observations per participant over three months in 507 adults with type 1 diabetes, type 2 diabetes, or no diabetes. The correlation between A1C and calculated mean glucose was 0.92. That group-level relationship supports translation between units and still leaves a prediction range around any one person's estimate.

Calculator output is an estimated laboratory-equivalent percentage, not a measured A1C.

Device accuracy, missing sensor days, selective finger-stick timing, recent glucose change, and personal differences in red-cell glycation can move measured A1C away from the estimate. Use enough glucose data from a period that overlaps the laboratory measurement before comparing the two.

[2][3][4]

Worked A1C conversion examples

Suppose a meter or sensor reports a 90-day mean of 154 mg/dL. Substitution gives (154 + 46.7) / 28.7 = 6.99%, which rounds to 7.0%. For a mean of 126 mg/dL, the calculation is (126 + 46.7) / 28.7 = 6.02%. Rounding a value just below a diagnostic boundary does not mean the boundary has been crossed.

For an SI-unit example, 8.0 mmol/L × 18.0182 = 144.15 mg/dL. Then (144.15 + 46.7) / 28.7 = 6.65%, or about 6.6% to 6.7% depending on display precision.

The result describes the mean entered. It does not report how much time glucose spent below 70 mg/dL or above 180 mg/dL.

[2][3]

Diagnostic categories and confirmation

For nonpregnant people, laboratory A1C below 5.7% is outside the prediabetes range, 5.7% through 6.4% meets the A1C criterion for prediabetes, and 6.5% or higher meets one criterion for diabetes. The glucose alternatives are fasting plasma glucose, a two-hour value after a 75 g oral glucose tolerance test, or random plasma glucose in a person with classic hyperglycemia symptoms or hyperglycemic crisis.

Unless hyperglycemia is unequivocal, diagnosis requires two abnormal results. They may be two different tests collected together, or the same or a different test repeated promptly. If A1C and fasting glucose disagree, the test above its diagnostic cutoff should be repeated while the clinician checks for analytical and biological explanations. Results near a threshold often lead to repeat testing in three to six months.

An A1C estimate calculated from home glucose is not a diagnostic result.

A1C is also unsuitable as the sole diagnostic test in pregnancy, altered red-cell turnover, some hemoglobin variants, erythropoietin treatment, dialysis, and some HIV treatment contexts. The ADA directs clinicians to plasma glucose criteria when the relationship between A1C and glycemia is altered.

[1][5]

Monitoring context and personal goals

Diagnosis classifies disease. Monitoring checks whether a care plan is reaching an individualized goal without unacceptable burden or hypoglycemia. The ADA considers an A1C below 7% appropriate for many nonpregnant adults who are not experiencing severe or harmful hypoglycemia. A lower goal, such as below 6.5%, may fit someone with good health, function, and low treatment risk. Other people need a less stringent goal.

Testing frequency follows clinical stability. The 2026 Standards state that adults with type 1 or type 2 diabetes who maintain goal-range glucose may need A1C about twice yearly. People with recent treatment changes, unstable glucose, intensive plans, or results outside their goals often need assessment about every three months, with extra review when clinically indicated. A1C should be read beside blood glucose monitoring or continuous glucose monitoring when variability or hypoglycemia is a concern. CGM adds time in range, time below range, and pattern information that an average cannot supply.

A calculated A1C trend should not trigger an independent medication or insulin change.

[2][8]

Laboratory and preanalytic limitations

A1C has better preanalytic stability than plasma glucose because cells continue consuming glucose after collection unless a glucose specimen is processed promptly or kept under suitable conditions. A1C has different vulnerabilities. Blood loss, transfusion, hemolysis, erythropoietin, kidney failure, dialysis, pregnancy, and anemia can alter red-cell age or turnover, changing A1C without a matching change in mean glucose. Iron deficiency can produce a falsely high result in some circumstances.

Hemoglobin variants can cause method-specific interference. NGSP maintains a table showing whether common methods are affected by HbC, HbS, HbE, HbD, or elevated fetal hemoglobin.

A person with sickle cell trait may receive a valid A1C if the laboratory uses a method without interference from that variant. People with homozygous variants such as HbSS lack normal HbA, so A1C cannot be interpreted in the usual way.

When the estimate, laboratory A1C, symptoms, and device data do not fit together, investigate. Confirm that the glucose period overlaps the A1C draw, review sensor wear and meter technique, ask which A1C method the laboratory used, and consider red-cell conditions. Clinicians may use plasma glucose for diagnosis and CGM, self-monitoring, fructosamine, or glycated albumin for monitoring when A1C is unreliable.

[1][2][5][6]

Population factors and individual variation

The ADAG conversion came from 507 adults, 83% of whom were non-Hispanic White. Its confidence ranges show that the same A1C can correspond to different mean glucose values. Modern CGM systems also differ from the older calibrated sensors and capillary sampling used in that study. Sensor accuracy, wear duration, interstitial lag, and the timing of the laboratory draw affect comparisons.

Race and ethnicity should not be used as correction factors for A1C. The 2026 ADA Standards note that self-identified race is a poor proxy for genetic variants that influence the A1C-glucose relationship. Specific findings, such as a known hemoglobin variant or G6PD deficiency, are clinically relevant. Broad racial adjustment can hide the actual analytical or biological cause of discordance.

The formula does not adapt to age, pregnancy, treatment type, anemia, or hypoglycemia risk.

Children, older adults, pregnant people, and people with advanced kidney disease may have different goals or limitations. Those factors belong in interpretation with a clinician rather than in an arithmetic correction applied by this tool.

[1][2][3][6]

How to interpret the estimate

Check the input first. A mean in mg/dL should usually fall within the operating range of the meter or sensor that produced it. Do not enter a fasting value, a post-meal value, or one day of readings as though it were a multiweek average. If the source is CGM, note the number of valid days and whether the period included illness, steroid treatment, travel, or a recent care-plan change.

Compare like periods. A laboratory A1C drawn today is weighted toward recent glucose but includes earlier weeks. A seven-day device average may diverge after a recent improvement or deterioration without indicating that either measure is defective. A 90-day mean with adequate coverage is a closer conceptual match, although personal red-cell biology can still create a gap.

A sustained movement from an estimated 8.4% to 7.6% may support that average glucose is improving. It does not prove that a personal target has been reached safely. Review low readings, time above range, symptoms, and the measured laboratory result before decisions are made.

[2][4][8]

How to use this A1C calculator

Enter a representative average glucose and select the correct unit. Use a device average covering as much of the prior two to three months as available, with consistent sensor wear or a testing schedule that includes relevant times of day. Record the date range and the calculator result so a later comparison uses the same method. If only a few readings are available, label the output as a rough estimate.

Use the result to prepare questions for a medical visit. Ask whether your measured A1C and glucose record agree, what personal goal applies, and whether anemia, kidney disease, pregnancy, a hemoglobin variant, transfusion, or medication could change interpretation.

If the result falls near or above a diagnostic threshold and you do not have a diagnosis, arrange laboratory testing rather than repeating calculator entries.

Seek prompt medical assessment for excessive thirst, frequent urination, unexplained weight loss, vomiting, confusion, deep or difficult breathing, or a high glucose reading accompanied by illness. A calculator cannot assess diabetic ketoacidosis or hyperglycemic crisis. For routine follow-up, bring the laboratory report and a meter or CGM summary so the care team can compare averages, variability, lows, and trends.

[1][2][5][7][8]

How it works

A1c (%) ≈ (average glucose + 46.7) / 28.7 per ADAG study equation.

Frequently asked questions

• What A1c is prediabetes?5.7 to 6.4% indicates prediabetes. ≥6.5% meets diabetes diagnostic criteria.

Related calculators

References

1. American Diabetes Association Professional Practice Committee. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes 2026
2. American Diabetes Association Professional Practice Committee. Glycemic Goals, Hypoglycemia, and Hyperglycemic Crises: Standards of Care in Diabetes 2026
3. Nathan et al., ADAG Study Group. Translating the A1C assay into estimated average glucose values
4. Bergenstal et al.. Relationship between A1C and glucose levels in the ADAG study
5. National Institute of Diabetes and Digestive and Kidney Diseases. The A1C Test and Diabetes
6. NGSP. HbA1c Assay Interferences
7. Centers for Disease Control and Prevention. Diabetes Testing
8. American Diabetes Association Professional Practice Committee. Diabetes Technology: Standards of Care in Diabetes 2026
9. Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on long-term complications