Skip to content
Join free
Back

The family's health go-to

Fasting blood sugar and HbA1c are not the same fact

One test is a single morning after a fast. The other reflects about two to three months, and some blood conditions make it the wrong test to diagnose with.

Checked against the sources ยท Updated 17 Nov 2026

In short

  • Fasting glucose is one morning. HbA1c is about two to three months. They are both valid, they do not always match, and some conditions make A1C the wrong test.
  • The 2025 ADA standards and the 2020 WHO handbook describe what each test measures, when to repeat it, and when A1C is misleading.

Fasting blood sugar and HbA1c are not the same fact

You are at the lab before the city is fully awake, holding your mother's papers in a plastic folder, and the canteen is already selling coffee. She is the one who has to fast, so the cup you do not buy is the whole lesson in miniature. If you are the person the family texts when two numbers disagree, you do not need a winner. You need to know which question each test was asked.

Fasting blood sugar is one morning. HbA1c is about two to three months. Both are valid. They do not always match. And some conditions make A1C the wrong test to diagnose with. Nothing here ranks one as the family champion, and nothing here diagnoses your parent.

What each test is looking at

Fasting plasma glucose is that morning. The ADA's definition is no caloric intake for at least eight hours. WHO's note is an overnight fast. The "no calories" line is the ADA's wording, so ask the lab what fasting means for them before you assume the canteen coffee was harmless. Water is not listed as a calorie. Coffee with milk or sugar is.

The ADA and WHO both put the diabetes line for that fasting test at 126 mg/dL, or 7.0 mmol/L, or higher.

HbA1c, which the ADA writes as A1C, is glucose stuck to hemoglobin, the protein in red blood cells. It reflects about the past two to three months, and it does not require a fast. WHO calls that an advantage, and also calls the test more costly and more indirect. The diabetes line, in both documents, is 6.5%, or 48 mmol/mol, or higher.

A two-hour test sits beside them. After a drink that contains 75 g of glucose, it looks at a different slice of the day than either the fasting morning or the months-long A1C. The two slips in your folder are not twins, and a two-hour result is a third fact again.

They miss different people

The ADA says fasting glucose, the two-hour test, and A1C reflect different parts of how the body handles glucose, so the groups they flag overlap without matching perfectly. A two-hour test after 75 g of glucose picks up more people with prediabetes and diabetes than the fasting or A1C cutoffs do. You can hold a result called normal on one paper and a high result on the other, and both papers can be doing their job.

The same ADA section notes that prevention trials were shown mainly in people who had a high two-hour result, with or without a high fasting number. They were not shown for an isolated high fasting glucose, or for prediabetes defined only by A1C. A normal A1C does not cancel a high two-hour test. If you are collecting results for a parent, the test name belongs beside the number before anyone declares the worry over.

A fasting 118 mg/dL and an A1C of 5.8% are both inside the ADA's prediabetes ranges, and they are not the same fact. One is a morning. One is a longer average.

An abnormal result should be repeated unless the picture is already obvious. The ADA's example of obvious is classic symptoms, or a crisis, plus a random glucose at or above 200 mg/dL. Otherwise the ADA wants two abnormal results, from the same draw or on two days. WHO says the same idea for people without symptoms: repeat, preferably with the same test, on another day. The crisis wording is the ADA's. WHO's random-glucose line is for use when symptoms are present. Same neighborhood, not the same sentence.

When A1C is the wrong tool

A1C rides on red blood cells living about as long as they usually do. When they do not, the months-long average can lie. The ADA's list of those situations is the long one: some hemoglobin variants, pregnancy in the second and third trimesters and after birth, glucose-6-phosphate dehydrogenase deficiency, HIV, hemodialysis, recent blood loss or transfusion, hemolysis, and erythropoietin therapy. In those settings the ADA says to use plasma glucose, not A1C, as the diagnostic test.

WHO's list is shorter, and it is the list in WHO's handbook: hemoglobin disorders, kidney failure, some anemias, and red cells that turn over quickly, as reasons HbA1c can be inaccurate. WHO also prefers plasma glucose over HbA1c when type 1 is suspected. The longer list is the ADA's. The shorter one is WHO's. Mixing them is how a family folder gets confusing.

A hemoglobin variant that does not shorten the life of the red cell, such as sickle-cell trait, needs an assay that is not thrown off by that variant. The ADA points clinicians to a list of which assays are affected. That list is for clinicians. There is no guess here about which kit on a pharmacy shelf is the certified lab method the standard requires. A pharmacy A1C is not automatically that method. A continuous glucose monitor is not a screening or diagnostic test in the 2025 standard.

How to label the paper you keep

You do not have to become the lab. You do have to stop the folder from mixing two facts into one.

  • Before a fasting test, ask what fasting means there. No calories for at least eight hours is the ADA line. The coffee with milk is a calorie. The water is not listed as one.
  • Write the test name next to every number you collect. Fasting, two-hour, or A1C. Lab or finger-stick. Whether anyone repeated it. A fasting 118 mg/dL and an A1C of 5.8% can both sit in the prediabetes ranges and still be different news.
  • Do not let a normal A1C erase a high two-hour result. The prevention trials speak most clearly to the high two-hour result. An isolated fasting high, or a label that rests only on A1C, is a different group.

Neither test wins the family argument. The useful sentence is the specific one: which test, which morning or which months, and whether the red cells can be trusted to tell the months.

When the mismatch is the reason to call

Bring the folder to a doctor when the two tests disagree, when A1C looks oddly low or high beside the glucose numbers, or when your parent is pregnant, on dialysis, living with HIV, or known to have a hemoglobin condition. Those are the settings where the ADA says plasma glucose should be the diagnostic test. You are not being asked to decide which machine was wrong. You are being asked to notice that A1C may be the wrong tool, and to let a clinician choose the test that still works. That is a kinder use of the family role than picking a favorite number.

---

Sources

  • American Diabetes Association Professional Practice Committee. 2. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes-2025. Diabetes Care. 2025. https://doi.org/10.2337/dc25-s002
  • World Health Organization. Diagnosis and management of type 2 diabetes (HEARTS-D). Geneva: WHO; 2020. WHO/UCN/NCD/20.1. https://iris.who.int/server/api/core/bitstreams/2a0b4f68-7155-4ad1-b543-945791e31830/content

General guidance, not medical advice or a diagnosis. Talk with your doctor about your own health.

How strong is the evidence?

Clear evidence

What the research says: The ADA says the tests identify overlapping but not identical groups, and that prevention trials were shown mainly in people with a high two-hour glucose, not in isolated high fasting glucose or in prediabetes defined only by A1C. A continuous glucose monitor is not for diagnosis. WHO prefers plasma glucose over HbA1c when type 1 is suspected, and says HbA1c can be inaccurate with hemoglobin disorders, kidney failure, some anemias, and fast red-cell turnover.

What it doesn't say: This does not rank one test as the family winner. It does not convert a pharmacy finger-stick into a diagnosis. It does not say a normal A1C erases a high two-hour test. Point-of-care A1C rules were not turned into a shopping list.

American Diabetes Association Professional Practice Committee. Diagnosis and classification of diabetes. Diabetes Care. 2025.

How we rate: the Flamel Method

Try this

  • Before a fasting test, ask the lab what 'fasting' means for them. The ADA definition is no caloric intake for at least eight hours. Water is not listed as a calorie. Coffee with milk or sugar is.
  • When you collect a parent's results, write the test name next to the number. A fasting 118 mg/dL and an A1C of 5.8% are both in the ADA prediabetes table, and they are not the same fact.

That counts.

See a doctor if The tests disagree, A1C looks oddly low or high beside the glucose, or your parent is pregnant, on dialysis, living with HIV, or has a hemoglobin condition. The ADA says plasma glucose should be the diagnostic test in those settings.

Get delightful living tips directly in your email

Join free

Sources

  1. American Diabetes Association Professional Practice Committee. Diagnosis and classification of diabetes. Diabetes Care. 2025.
  2. World Health Organization. Diagnosis and management of type 2 diabetes (HEARTS-D). Geneva: WHO; 2020. WHO/UCN/NCD/20.1.

Last reviewed by Flamel Staff on 17 Nov 2026

Flamel is not medical advice and does not diagnose. If you have symptoms or readings in the red zone, see a doctor.