A1C Chart for Black Adults: What Your Number Means

An A1C below 5.7 percent is normal. From 5.7 to 6.4 percent is prediabetes. At 6.5 percent or above, that is diabetes. Those are the cutoffs your lab prints on the report. What the report does not tell you is that two inherited traits carried by millions of Black Americans, G6PD deficiency and sickle cell trait, can pull the number below your actual blood sugar. Neither one shows up unless somebody checks.

Any of these tests diagnoses the same three states, and any result other than a clear one with symptoms gets confirmed by a second test on a different day.

• Normal: A1C below 5.7 percent. Fasting plasma glucose 99 mg/dL or below. Two-hour oral glucose tolerance test 139 mg/dL or below.
• Prediabetes: A1C 5.7 to 6.4 percent. Fasting glucose 100 to 125 mg/dL. Two-hour glucose 140 to 199 mg/dL.
• Diabetes: A1C 6.5 percent or above. Fasting glucose 126 mg/dL or above. Two-hour glucose 200 mg/dL or above. Random glucose 200 mg/dL or above with symptoms.

An A1C of 5.9 sits in the middle of the prediabetes band. Not diabetes, not nothing. Inside that band the higher the number the higher the chance of progressing, which is why crossing 5.7 moves your retest from every 3 years to every year. One caveat on where the number came from: a diagnostic A1C goes to a lab running an NGSP-certified method. The analyzer that gives you a result during the visit is built for monitoring, not diagnosis.

Hemoglobin carries oxygen in your red blood cells. Glucose sticks to it and stays stuck for the life of the cell, roughly 120 days. A1C is the share of your hemoglobin carrying sugar, reported as a percentage. The 3-month window is not evenly weighted: glucose from the past 30 days moves the result more than glucose from 90 to 120 days ago, so a real change shows up on a repeat test in about 3 months. It also means A1C will not register one bad week, and a normal A1C does not rule out sharp spikes after meals.

The A1c-Derived Average Glucose study followed 507 people through 3 months of continuous glucose monitoring and fingerstick testing and produced the equation labs use: estimated average glucose in mg/dL equals 28.7 times your A1C, minus 46.7. Our arithmetic from that equation, rounded:

• A1C 5.0 percent: about 97 mg/dL
• A1C 5.7 percent: about 117 mg/dL
• A1C 6.0 percent: about 126 mg/dL
• A1C 6.5 percent: about 140 mg/dL
• A1C 7.0 percent: about 154 mg/dL
• A1C 8.0 percent: about 183 mg/dL
• A1C 9.0 percent: about 212 mg/dL
• A1C 10.0 percent: about 240 mg/dL

Read those as an average across months, not as a reading you should expect on a meter at any given moment.

Two inherited traits common in Black families change what an A1C result means. Both are usually silent, and both push the number down, which is the dangerous direction: a falsely reassuring A1C delays a diagnosis rather than causing a false alarm.

Glucose-6-phosphate dehydrogenase deficiency shortens the life of red blood cells, so sugar has less time to attach and A1C reads low while glucose sits where it always was. Using whole-exome data from UK Biobank and the Genes and Health cohort, researchers estimated that about 1 in 7 Black men carries a G6PD deficiency allele and that fewer than 1 in 50 who carry it have it in their medical notes. Male carriers averaged 0.9 percentage points lower A1C with effectively unchanged average glucose. They were diagnosed with type 2 diabetes an average of 4.1 years later, and had 1.37 times the odds of microvascular complications.

The NIH's Africans in America study found the same 0.9-point gap in people living in the United States and measured what it costs. Against the oral glucose tolerance test, A1C at the 5.7 percent cutoff identified abnormal glucose tolerance in 0 of 17 participants with G6PD deficiency. Sensitivity was zero. Work in the Million Veteran Program traced the mechanism to shortened red-cell survival, and found that in the ACCORD trial, participants with the deficiency developed retinopathy and neuropathy at higher rates at the same A1C.

About 1 in 13 Black Americans has sickle cell trait and about 2.3 percent carry hemoglobin C trait. One copy causes no symptoms, so most carriers do not know. A 2017 JAMA analysis of 4,620 African American adults in the CARDIA and Jackson Heart studies found A1C ran 0.29 percentage points lower in people with the trait at the same fasting glucose. Screening on A1C alone flagged prediabetes in 29.2 percent of observations from participants with the trait and 48.6 percent of observations from participants without it, at the same glucose levels.

Two later studies point the other way, and they deserve saying out loud. A Ugandan validation study of 182 adults with type 2 diabetes compared four A1C methods against 14 days of continuous glucose monitoring and found the trait did not alter the A1C-to-glucose relationship for any method. A 2025 follow-up in the same population found no effect either. Sickle cell trait may shift your A1C, the size of the effect is disputed, and the response is to check rather than assume.

Older A1C methods did give wrong answers in people with hemoglobin variants. Most no longer do. The NGSP publishes an interference table, updated June 2026: of the 20 most-used methods, only the Siemens DCA Vantage point-of-care analyzer carries conflicting published data on sickle cell trait, with reports of falsely high results. The rest show no clinically significant interference from hemoglobin S, C, E, or D trait. Elevated fetal hemoglobin is the bigger remaining assay problem, affecting several methods.

What to do with that. If your A1C came from a fingerstick machine in the exam room and it disagrees with your home readings or how you feel, ask for a venous draw sent to a lab, and ask which method that lab uses. If the numbers and the symptoms still disagree, ask for a fasting plasma glucose or an oral glucose tolerance test, which measure sugar directly and skip hemoglobin entirely. Fructosamine gets suggested as the alternative and has a real limit: a 2025 study in a population with a high rate of these conditions found that switching to fructosamine or glycated albumin did not improve accuracy over A1C. Direct glucose testing or a continuous glucose monitor is the better ask.

• Iron-deficiency anemia can push A1C falsely high.
• Recent blood loss, a transfusion, hemodialysis, or erythropoietin change how long red cells live, and change A1C with them.
• Kidney failure or liver disease can produce false results.
• Sickle cell disease itself (hemoglobin SS, SC, or CC, two variant genes rather than one) is not the same as trait. A1C should not be used in these patients at all. Glycated serum protein or glycated albumin is used instead.
• Pregnancy. Gestational diabetes is screened with the glucose challenge test or an OGTT between 24 and 28 weeks, not with A1C.

One more number worth keeping: repeated on the same blood sample, an A1C reported as 6.8 percent can come back anywhere from 6.4 to 7.2 percent. A shift of a tenth or two between visits is noise, not a trend.

The Diabetes Prevention Program randomized 3,234 adults with elevated glucose to a lifestyle program targeting 7 percent weight loss and at least 150 minutes of activity a week, to metformin, or to placebo. Over an average 2.8 years, new diabetes occurred at 11.0 cases per 100 person-years on placebo, 7.8 on metformin, and 4.8 with lifestyle: a 58 percent reduction with lifestyle and 31 percent with metformin. Preventing one case over 3 years took 6.9 people in the lifestyle program. Forty-five percent of participants were from minority groups.

Look AHEAD gives the size of the A1C move in people who already have type 2 diabetes. Among 5,145 adults, the intensive lifestyle arm lost an average 8.6 percent of body weight in year one and mean A1C fell from 7.3 to 6.6 percent. A 0.7-point drop in a year from weight and activity, with more of that group coming off diabetes, blood pressure, and lipid medicines than the comparison group.

Because the test weights the most recent 30 days most heavily, a change you make now shows on a repeat in about 3 months. If you are in the prediabetes band, our guide on reversing prediabetes has the specifics. If metformin is on the table, read what to expect from metformin before you start.

• Normal result: retest every 3 years.
• Prediabetes: test for type 2 diabetes every year.
• Diagnosed diabetes: A1C at least twice a year, more often if your treatment changed or you are not at target.

Routine testing starts at 35, and earlier for Black adults carrying any other risk factor, including overweight, family history, or a past pregnancy with gestational diabetes. Nationally, 8.7 million adults have diabetes and do not know it, 22.8 percent of everyone who has it.

Bring three things to the appointment: your A1C with its date, any home glucose readings, and one question. Was this run in a lab or on the office analyzer, and does the method handle hemoglobin variants? If you have sickle cell trait, G6PD deficiency, or a family history of either, say so before the blood is drawn rather than after the result comes back. You can find a Black primary care clinician or endocrinologist in our directory by city and specialty.