HbA1c is a blood test that shows your average blood sugar level over the past two to three months. Understanding your result can help you see whether your glucose is within a healthy range, whether you may have prediabetes or diabetes, and how well your current treatment plan is working.
Written by: Dr. Albana Greca, MD, MMedSc, Family Physician and Medical Author.
Medically reviewed by: Dr. Ruden Cakoni, MD, Endocrinologist.
Last reviewed: July 2026.
HbA1c—also called A1C or glycated hemoglobin—is a blood test that estimates your average glucose exposure over approximately the previous two to three months. Unlike a finger-stick or continuous glucose monitor reading, it does not show what your glucose is at this exact moment.
For people who have not been diagnosed with diabetes, an HbA1c below 5.7% is generally classified as normal, 5.7–6.4% as prediabetes, and 6.5% or higher as diabetes when confirmed appropriately. For many nonpregnant adults who already have diabetes, a common treatment goal is below 7%, but the safest target must be individualized.
HbA1c is one of the most useful tests in diabetes care because it summarizes glucose exposure over time. It can help screen for and diagnose prediabetes or diabetes, evaluate whether treatment is working, and guide discussions about nutrition, physical activity, medication, insulin, and glucose-monitoring patterns.
However, HbA1c is not a complete report card. It cannot show daily glucose swings, the timing of highs and lows, or whether a person has frequent hypoglycemia. It can also be misleading when a medical condition changes red blood cell lifespan or interferes with the laboratory method.
Hemoglobin is the oxygen-carrying protein inside red blood cells. Glucose naturally attaches to some hemoglobin molecules in a process called glycation. The higher the glucose concentration has been over time, the larger the percentage of glycated hemoglobin usually becomes.
Red blood cells circulate for roughly 120 days, but HbA1c is commonly described as reflecting the previous two to three months. The result is weighted toward more recent glucose exposure because newer red blood cells make up a substantial part of the sample.
An HbA1c of 7% means that approximately 7% of measured hemoglobin is glycated. It does not mean that 7% of your blood is sugar.
For most nonpregnant people, the general diagnostic categories are:
| HbA1c | IFCC equivalent | General interpretation | What usually happens next |
|---|---|---|---|
| Below 5.7% | Below 39 mmol/mol | Usually classified as normal | Routine screening based on age and risk factors |
| 5.7–6.4% | 39–46 mmol/mol | Prediabetes range | Risk assessment, prevention plan, and follow-up testing |
| 6.5% or higher | 48 mmol/mol or higher | Diabetes range | Confirmation and full clinical evaluation unless hyperglycemia is unequivocal |
In the absence of unmistakable hyperglycemia or a hyperglycemic crisis, diabetes diagnosis usually requires two abnormal results. These may be the same test repeated, or two different accepted tests that are both above their diagnostic thresholds.
When HbA1c and fasting glucose disagree, the clinician may repeat the test that is above the diagnostic threshold, review possible laboratory interference, or use another test such as fasting plasma glucose or an oral glucose tolerance test.
A home glucose meter, CGM, or over-the-counter home HbA1c kit should not be used alone to diagnose diabetes.
For many nonpregnant adults with diabetes, the American Diabetes Association uses an HbA1c target below 7% as a common starting point. This is not a universal target.
A lower goal may be reasonable when it can be achieved safely without significant hypoglycemia, treatment burden, or adverse effects. A less stringent goal may be safer when the risks of intensive treatment outweigh the expected benefit.
Your target may depend on:
The safest goal is not necessarily the lowest possible number. A low HbA1c achieved through frequent hypoglycemia is not good diabetes control.
For a focused discussion, read what a good HbA1c level may be for different people.
Estimated average glucose, or eAG, converts HbA1c into units that resemble glucose-meter readings.
eAG in mg/dL = 28.7 × HbA1c − 46.7
eAG in mmol/L = eAG in mg/dL ÷ 18
| HbA1c | IFCC | Estimated average glucose | Estimated average glucose |
|---|---|---|---|
| % | mmol/mol | mg/dL | mmol/L |
| 5.0% | 31 | 97 | 5.4 |
| 5.7% | 39 | 117 | 6.5 |
| 6.0% | 42 | 126 | 7.0 |
| 6.5% | 48 | 140 | 7.8 |
| 7.0% | 53 | 154 | 8.6 |
| 8.0% | 64 | 183 | 10.2 |
| 9.0% | 75 | 212 | 11.8 |
| 10.0% | 86 | 240 | 13.3 |
| 11.0% | 97 | 269 | 14.9 |
| 12.0% | 108 | 298 | 16.5 |
These are estimates based on the ADAG relationship. Your meter or CGM average may differ because HbA1c and glucose devices measure different things, and because people often check glucose more frequently at certain times of day.
Use the dedicated tool:
A finger-stick or CGM reading shows glucose at a particular moment. HbA1c summarizes long-term exposure. Neither test replaces the other.
You can have:
When the results do not match, do not choose whichever result feels more reassuring. Ask your clinician to review the timing and quality of the tests, glucose-monitoring data, red blood cell conditions, kidney and liver health, pregnancy status, and possible hemoglobin variants.
Continuous glucose monitoring can show time in range, time above range, time below range, glucose variability, overnight patterns, and meal-related changes. These details are invisible to HbA1c.
Some CGM reports provide a glucose management indicator, or GMI. GMI estimates an A1C-like value from the CGM average, but it is not a laboratory HbA1c and may differ from the measured result.
A difference between GMI and laboratory HbA1c is not automatically a device failure. It may reflect biological variation, incomplete sensor data, recent glucose changes, anemia, altered red blood cell turnover, or other conditions.
HbA1c depends on both glucose exposure and red blood cell biology. A result may be falsely high, falsely low, or simply unreliable when either is altered.
| Situation | Possible effect | Clinical response |
|---|---|---|
| Iron-deficiency anemia | May raise HbA1c in some people | Evaluate and treat the anemia; compare with glucose data |
| Hemolysis or recent blood loss | May lower HbA1c by shortening red blood cell lifespan | Use glucose-based assessment or another suitable marker |
| Recent transfusion | May make the result difficult to interpret | Tell the clinician and laboratory when the transfusion occurred |
| Erythropoietin treatment, dialysis, or advanced kidney disease | May alter the relationship between HbA1c and true average glucose | Interpret with CGM, meter data, and kidney-team guidance |
| Hemoglobin variants or hemoglobin disorders | Some laboratory methods may read falsely high or low; HbA1c may be unavailable in some disorders | Use an appropriate certified assay or an alternative approach |
| Pregnancy | Red blood cell and glucose changes can reduce reliability for monitoring and diagnosis | Use pregnancy-specific glucose targets and testing methods |
| Rapid recent improvement or worsening | HbA1c may lag behind the new glucose pattern | Use current meter or CGM data while waiting for the next HbA1c |
Liver disease, vitamin deficiencies, certain medicines, and other blood conditions may also affect interpretation. Tell your clinician about anemia, transfusion, pregnancy, dialysis, or a known hemoglobin condition before relying on HbA1c alone.
Pregnancy uses tighter and pregnancy-specific glucose targets. HbA1c may be useful before conception and as one part of care, but it is not the best standalone test for day-to-day monitoring during pregnancy.
Gestational diabetes is usually assessed with glucose challenge or oral glucose tolerance testing rather than diagnosed from HbA1c alone. Anyone who is pregnant or planning pregnancy should review targets and medicines with the obstetric and diabetes teams promptly.
Many people who are meeting treatment goals and have stable glucose control have HbA1c checked at least twice a year. Testing is often performed about every three months when:
The schedule should be individualized. Repeating HbA1c too soon may not show the full effect of a treatment change, while waiting too long with clearly uncontrolled glucose may delay necessary care.
An HbA1c of 5.6% is below the standard prediabetes cutoff of 5.7%. It does not diagnose prediabetes.
It may still deserve attention when a person has significant risk factors, rising results, abnormal fasting glucose, previous gestational diabetes, polycystic ovary syndrome, high blood pressure, abnormal triglycerides, or a strong family history.
Read the detailed answer to whether HbA1c 5.6% should prompt additional monitoring.
Do not change medication based only on the HbA1c percentage. Review the pattern that produced it.
Your clinician may assess:
An elevated HbA1c is a reason to improve the plan—not a reason for shame. Diabetes care should be practical, collaborative, and focused on safety and long-term health.
Our blood sugar control guide explains how monitoring, meals, activity, sleep, medicines, and sick-day planning work together.
Do not increase insulin, double medication, begin a restrictive diet, or use fasting to lower HbA1c without professional guidance. A rapid reduction achieved through repeated hypoglycemia is unsafe.
Seek urgent care for confusion, seizure, unconsciousness, inability to swallow safely, repeated severe low glucose, vomiting with ketones, deep breathing, fruity-smelling breath, severe dehydration, or very high glucose with serious symptoms.
I explain HbA1c as an average—not a verdict. Before changing treatment, we need to know whether the average comes from steady glucose, repeated meal spikes, overnight highs, or a mixture of dangerous highs and lows. We also need to confirm that anemia, kidney disease, pregnancy, blood loss, transfusion, or a hemoglobin condition is not distorting the result.
HbA1c is used to estimate average glucose, but it is not a direct average of your meter or CGM readings. eAG is a calculated estimate derived from HbA1c.
No. You usually do not need to fast for HbA1c, although other blood tests ordered at the same visit may require fasting.
One meal generally will not meaningfully change HbA1c. Repeated high glucose over days and weeks can raise it.
Yes. Intermittent highs may be averaged with normal or low readings, and some medical conditions can make HbA1c falsely low. Review the pattern with your clinician.
Yes. After-meal or overnight glucose may be high, or HbA1c may be affected by iron deficiency, a laboratory issue, or another red blood cell condition.
No. It is a common goal for many adults, but the result must be interpreted with hypoglycemia, glucose variability, age, pregnancy, other conditions, and treatment burden.
HbA1c begins responding to glucose changes within weeks, but the full effect of a sustained change is usually assessed over approximately three months.
Check whether enough CGM data were collected, review recent glucose changes, and ask about anemia, kidney disease, transfusion, pregnancy, and hemoglobin variants. Do not adjust treatment from either number alone.
Educational safety note: This article provides general diabetes education and does not replace personal medical advice, laboratory interpretation, diagnosis, or treatment. Do not start, stop, or change insulin, diabetes medicines, supplements, diet, or exercise based only on an HbA1c result.