Learn
A1c high, but glucose normal?
An A1c in the prediabetes or diabetes range next to a normal fasting glucose is a common and confusing mix. It does not mean one test is broken. Here are the reasons it happens, and how doctors sort it out.
The short answer
- The tests measure different things. Fasting glucose is a single moment. A1c is a long-term measure of about 2 to 3 months (ADA Standards 2026, sec. 2). A result in range on one and out of range on the other is expected for some people (NIDDK).
- The A1c can also be off. Iron deficiency, kidney or liver disease and some hemoglobin variants can make it read high (NIDDK).
- A repeat test settles it. The ADA says to repeat the test that is above its cut point, and to look for interference when the gap is large and consistent (ADA Standards 2026, sec. 2).
First, the cut points
Each test has its own ranges. These are the ADA's criteria for nonpregnant adults (ADA Standards 2026, sec. 2):
| Test | Normal | Prediabetes | Diabetes |
|---|---|---|---|
| A1c | Below 5.7% | 5.7% to 6.4% | 6.5% or higher |
| Fasting plasma glucose | Below 100 mg/dL | 100 to 125 mg/dL | 126 mg/dL or higher |
| 2-hour glucose (OGTT) | Below 140 mg/dL | 140 to 199 mg/dL | 200 mg/dL or higher |
The ADA points out that the three tests reflect different parts of glucose metabolism, so their cut points identify groups of people that only partly overlap (ADA Standards 2026, sec. 2). Being "normal" on one and "prediabetes" on another is not a contradiction.
1. Fasting is one moment; A1c is months
Fasting glucose is measured after at least 8 hours without food. A1c reflects glucose bound to hemoglobin over the life of the red blood cell, about 120 days, weighted toward recent weeks (ADA Standards 2026, sec. 2). Glucose after meals counts in the A1c but never shows on a fasting test. The 2-hour glucose test (OGTT), which does capture a glucose load, diagnoses more people with prediabetes and diabetes than either fasting glucose or A1c (ADA Standards 2026, sec. 2).
2. You may be early on
NIDDK explains that people with differing results may be at an early stage, when glucose has not risen enough to show on every test. In that case a doctor may follow the person closely and repeat the test in several months (NIDDK).
3. The same A1c fits a range of average glucose
The A1c-to-average-glucose formula is an average across people. In the study it comes from, people with the same A1c had quite different average glucose (Nathan 2008):
| A1c | Estimated average glucose | 95% range in the ADAG study |
|---|---|---|
| 6.0% | 126 mg/dL | 100 to 152 mg/dL |
| 7.0% | 154 mg/dL | 123 to 185 mg/dL |
| 8.0% | 183 mg/dL | 147 to 217 mg/dL |
From Nathan 2008, Table 2 (Nathan 2008). Estimated average glucose = 28.7 × A1c − 46.7.
So two people with an A1c of 7.0% might average 123 mg/dL or 185 mg/dL. Someone at the low end of that range can have a fairly high A1c with glucose readings that look ordinary.
4. Something is raising the A1c, not the glucose
A1c depends on red blood cells and hemoglobin, so anything that changes them can move the result without any change in glucose. NIDDK and the ADA list (NIDDK) (ADA Standards 2026, sec. 2):
- Iron-deficiency anemia, which can make A1c falsely high (NIDDK).
- Kidney failure and liver disease (NIDDK), and dialysis (ADA Standards 2026, sec. 2).
- Changes in red blood cell life span: recent blood loss, transfusion, hemolysis, splenectomy, erythropoietin treatment, G6PD deficiency (ADA Standards 2026, sec. 2).
- Hemoglobin variants, more common in people of African, Mediterranean or Southeast Asian descent. Most people who carry one have no symptoms and do not know it. They affect some A1c methods but not others (NIDDK).
- HIV and pregnancy also change how A1c relates to glucose (ADA Standards 2026, sec. 2).
Some of these push A1c down rather than up. More detail in when A1c can mislead.
5. The glucose sample read low
Glucose keeps being used up by blood cells after the sample is drawn. The ADA warns that glucose results will be falsely low if samples are not processed and analyzed promptly (ADA Standards 2026, sec. 2). Fasting glucose also moves with the time of day, medicines, alcohol, smoking, illness, stress and recent activity, while A1c is barely affected by those (ADA Standards 2026, sec. 2).
6. Normal test variation
Both tests vary from day to day and from test to test. NIDDK gives the example of an A1c of 6.8% that could read anywhere from 6.4% to 7.2% on a repeat from the same blood sample (NIDDK). Near a cut point, that alone can flip a category.
How doctors sort it out
The ADA Standards of Care describe the steps (ADA Standards 2026, sec. 2):
- Repeat the test that is above its cut point. The diagnosis rests on the confirming test.
- Two A1c results of 6.5% or higher mean diabetes, even if fasting glucose is below 126 mg/dL.
- Results close to a threshold are usually repeated in 3 to 6 months.
- A large, consistent gap between A1c and glucose should prompt a search for the cause, including a problem or interference with either test.
- Where a condition changes how A1c relates to glucose, such as some hemoglobin variants, pregnancy, G6PD deficiency, HIV or red blood cell turnover problems, plasma glucose tests should be used for diagnosis instead.
If you use a CGM, a large gap between your GMI and your lab A1c is useful to bring along too.
The opposite: A1c normal, glucose high
It happens the other way round as well. NIDDK notes that a glucose test can show diabetes when an A1c does not, and the reverse (NIDDK). The same reasons apply: different time windows, an early stage, and red blood cell changes: the ADA notes that faster red blood cell turnover usually lowers A1c (ADA Standards 2026, sec. 2).
Questions to ask your doctor
- Which result should be repeated, and when?
- Could anemia, iron levels, kidney or liver health, or a hemoglobin variant be affecting my A1c?
- Would a 2-hour glucose test (OGTT) or another test help?
- Does the lab's A1c method work with hemoglobin variants?
Frequently asked questions
Can my A1c be high if my fasting glucose is normal?
Yes. A fasting glucose test is one moment after an overnight fast, while A1c reflects about 3 months, including glucose after meals. The ADA notes that the different tests pick out groups of people that only partly overlap.
Which result counts for a diagnosis?
The ADA says that when two different tests disagree, the test that is above its diagnostic cut point should be repeated. Two A1c results of 6.5% or higher mean diabetes even if fasting glucose is below 126 mg/dL.
Could my A1c be falsely high?
Yes. Iron-deficiency anemia, kidney failure, liver disease and some hemoglobin variants can make an A1c read higher than average glucose would suggest. Doctors look for this when A1c and glucose results do not match.
What happens next if my results disagree?
Usually a repeat test. The ADA suggests repeating results close to a diagnostic threshold in 3 to 6 months, and looking for a cause when the gap between A1c and glucose is large and consistent.
Sources
- American Diabetes Association Professional Practice Committee. 2. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes 2026. Diabetes Care. 2026;49(Suppl 1). doi.org/10.2337/dc26-S002
- National Institute of Diabetes and Digestive and Kidney Diseases (NIH). The A1C Test & Diabetes. www.niddk.nih.gov/health-information/diagnostic-tests/a1c-test
- Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ; A1c-Derived Average Glucose (ADAG) Study Group. Translating the A1C assay into estimated average glucose values. Diabetes Care. 2008;31(8):1473-1478. doi.org/10.2337/dc08-0545
Written and maintained by the controla1clevels.com editorial team. We do not have a medical reviewer; every claim links to its primary source above. Last reviewed: 2026-10-26. Spotted an error? See About.