About

Glycated hemoglobin (HbA_1c) reflects mean plasma glucose over the preceding 8 - 12 weeks by measuring the fraction of hemoglobin irreversibly bound to glucose via non-enzymatic glycation. A misread A1C value leads to incorrect dosing of insulin or oral hypoglycemics. The clinical consequence is either hypoglycemia or sustained hyperglycemia accelerating microvascular damage. This converter implements the regression equation from the ADAG study (Nathan et al., 2008, n = 507), which correlated A1C with continuous glucose monitoring data across 2700+ measurements per subject.

The tool classifies results against American Diabetes Association thresholds: normal below 5.7%, prediabetes from 5.7 - 6.4%, diabetes at 6.5% or above. Limitations: the ADAG equation assumes no hemoglobin variants. Conditions such as sickle cell trait, thalassemia, iron-deficiency anemia, or recent transfusion alter erythrocyte lifespan and distort A1C readings independent of true glycemia. Pregnancy (second and third trimesters) also compresses A1C values downward.

Formulas

The ADAG regression equation converts glycated hemoglobin percentage to estimated average glucose:

eAG = 28.7 × A1C − 46.7

Where eAG = estimated average glucose in mg/dL, and A1C = glycated hemoglobin expressed as a percentage (%). The regression yielded R² = 0.84 (p < 0.0001).

To convert to SI units:

eAG_mmol/L = eAG_mg/dL18.0156

Where 18.0156 g/mol is the molar mass of glucose (C₆H₁₂O₆), divided by 10 to convert dL to L.

ADA classification thresholds applied to the input A1C value:

{

Normal if A1C < 5.7%Prediabetes if 5.7% ≤ A1C ≤ 6.4%Diabetes if A1C ≥ 6.5%

Reference Data

A1C (%)	eAG (mg/dL)	eAG (mmol/L)	ADA Classification	Clinical Note
4.0	68	3.8	Normal	Hypoglycemia risk if on sulfonylureas
5.0	97	5.4	Normal	Optimal non-diabetic range
5.5	111	6.2	Normal	Upper normal boundary
5.7	117	6.5	Prediabetes	ADA screening threshold
6.0	126	7.0	Prediabetes	Lifestyle intervention recommended
6.4	137	7.6	Prediabetes	High conversion risk to T2DM
6.5	140	7.8	Diabetes	Diagnostic threshold per ADA
7.0	154	8.6	Diabetes	ADA target for most adults
7.5	169	9.4	Diabetes	Reassess therapy if sustained
8.0	183	10.2	Diabetes	Increased retinopathy risk
8.5	197	10.9	Diabetes	Consider insulin intensification
9.0	212	11.8	Diabetes	Nephropathy screening critical
9.5	226	12.5	Diabetes	High cardiovascular event risk
10.0	240	13.4	Diabetes	Urgent glycemic control needed
11.0	269	14.9	Diabetes	DKA risk in T1DM patients
12.0	298	16.5	Diabetes	Severe uncontrolled diabetes
13.0	326	18.1	Diabetes	End-organ damage likely progressing
14.0	355	19.7	Diabetes	Immediate specialist referral

Frequently Asked Questions

Older tables (e.g., the DCCT regression) were derived from fewer subjects using less accurate glucose monitoring. The ADAG equation (eAG = 28.7 × A1C − 46.7) was validated using continuous glucose monitors and 7-point daily profiles across 507 subjects spanning A1C values from 5.0% to 13.0%. The 95% confidence interval for eAG at a given A1C is approximately ±30 mg/dL, which means individual variation exists even with the updated formula.

Hemoglobin S (sickle cell trait/disease), Hemoglobin C, and Hemoglobin E can interfere depending on the assay method. Ion-exchange HPLC methods are most affected. Boronate affinity and immunoassay methods are generally unaffected by HbS and HbC but may still be influenced by HbE. The NGSP maintains a list of assays and their known interferences. If a variant is present, fructosamine or glycated albumin testing over a 2-3 week window is the recommended alternative.

Iron-deficiency anemia extends red blood cell lifespan, increasing exposure time to glucose and artificially elevating A1C by 0.5-1.0%. Conversely, hemolytic anemias and recent blood loss shorten RBC lifespan, producing falsely low A1C. After correcting the anemia (e.g., iron supplementation for 8-12 weeks), repeat the A1C to get a valid baseline.

No. The 6.5% threshold was calibrated primarily on data from European and North American cohorts. Studies show that at the same A1C, African Americans have approximately 0.4% higher A1C relative to matched fasting glucose compared to non-Hispanic whites. Some guidelines (WHO, IDF) acknowledge this and recommend confirmatory fasting plasma glucose (≥126 mg/dL) or oral glucose tolerance testing, particularly in populations with known glycation rate differences.

The DCCT/EDIC and UKPDS trials demonstrated that each 1% reduction in A1C corresponds to approximately a 37% reduction in microvascular complications (retinopathy, nephropathy, neuropathy) and a 14-21% reduction in myocardial infarction risk. In absolute glucose terms, a 1% A1C change corresponds to approximately 28.7 mg/dL (1.6 mmol/L) change in average glucose.

With caution. During the second and third trimesters, hemodilution and increased red blood cell turnover lower A1C by approximately 0.5% independent of glycemic changes. The ADA recommends A1C testing in early pregnancy but switches to self-monitored blood glucose (fasting <95 mg/dL, 1-hour postprandial <140 mg/dL) for ongoing gestational diabetes management. The ADAG formula was not validated in pregnant populations.

A1C reflects the weighted average over 8-12 weeks, with the most recent 30 days contributing approximately 50% of the value. Testing more frequently than every 3 months provides overlapping windows and limited new information. The ADA recommends quarterly testing for patients not meeting glycemic targets and biannual testing for stable patients. Allow a minimum 8-week interval after any medication change before retesting.