Insulin, glucose, HOMA-IR and HbA1c: what do these results mean?

Fasting glucose is normal, but insulin looks high. HOMA-IR is above a value you found online, while HbA1c is still within range. Is this already insulin resistance? Does such a result mean prediabetes? Or is there no reason to worry?

Interpreting carbohydrate metabolism from a single parameter often leads to wrong conclusions. Glucose, insulin, HOMA-IR and glycated hemoglobin describe different elements of metabolism. They also do not all carry the same diagnostic value.

Glucose and HbA1c have established criteria for diagnosing prediabetes and diabetes. Insulin and HOMA-IR can provide additional information, but there is no single universal result that on its own would confirm insulin resistance.

Key points at a glance

• Fasting glucose shows the glucose concentration at a specific moment.
• Fasting insulin tells you how much insulin the body secretes to keep glucose at a given level, but its interpretation depends on the laboratory method and clinical situation.
• HOMA-IR is an index calculated from fasting glucose and insulin. It estimates insulin resistance, but is not a standalone diagnostic test.
• HbA1c reflects the body's average exposure to glucose over roughly the last two to three months.
• Normal glucose does not always mean the body does not have to secrete more insulin.
• Elevated insulin or HOMA-IR do not automatically mean diabetes.
• Diabetes and prediabetes are diagnosed based on glucose, HbA1c or the oral glucose tolerance test, not on insulin alone.

How does the body control glucose levels?

Glucose is one of the body's basic energy sources. Its concentration in the blood rises among other things after a meal containing carbohydrates. In response, the beta cells of the pancreas secrete insulin.

Insulin acts as a signal that allows glucose to be used or stored. It facilitates its transport into muscle and adipose tissue, limits glucose production in the liver and supports the rebuilding of glycogen stores.

When tissue sensitivity to insulin is reduced, the pancreas can produce more of it for some time. As a result, glucose can still remain normal. This state is called compensatory hyperinsulinemia.

Only when the beta cells are no longer able to secrete enough insulin relative to the body's demand does glucose start to rise more clearly. This is precisely why carbohydrate metabolism cannot always be assessed based on a single glucose measurement.

Glucose: a result from a single moment

Blood glucose shows its concentration at the time the sample is taken. The result depends on whether the test was done fasting, after a meal, during an oral glucose tolerance test or at any time of day.

Diagnostics primarily use glucose measured in venous blood plasma. A home glucometer can be very useful for monitoring diabetes, but does not serve to diagnose it on your own.

Fasting glucose

The test is performed after at least eight hours without calorie intake. During this time you may drink water. Blood is most often drawn in the morning.

How to interpret fasting glucose?

Glucose in venous blood plasma	Interpretation
below 100 mg/dl, i.e. below 5.6 mmol/l	normal value
100–125 mg/dl, i.e. 5.6–6.9 mmol/l	impaired fasting glucose, one of the prediabetic states
at least 126 mg/dl, i.e. at least 7.0 mmol/l	value meeting the criterion for diabetes

In the absence of characteristic diabetes symptoms, a single abnormal result usually requires confirmation in a subsequent test. Confirmation may be two abnormal results of the same test or abnormal results of two different diagnostic tests.

Is glucose of 99 mg/dl good and 100 mg/dl bad?

Diagnostic thresholds are needed for making clinical decisions, but metabolism does not change abruptly once a single number is crossed.

Glucose of 99 mg/dl and 100 mg/dl are very similar results. The difference may stem among other things from natural biological and laboratory variability. When interpreting a borderline result, it is worth considering earlier measurements, HbA1c, body weight, waist circumference, family history, medications and other risk factors.

Glucose in the oral glucose tolerance test

The oral glucose tolerance test, or OGTT, consists of measuring fasting glucose, drinking a solution containing 75 g of glucose and drawing blood again after two hours.

In non-pregnant people, the result after two hours is interpreted as follows:

Glucose 2 hours after OGTT	Interpretation
below 140 mg/dl, i.e. below 7.8 mmol/l	normal glucose tolerance
140–199 mg/dl, i.e. 7.8–11.0 mmol/l	impaired glucose tolerance, prediabetes
at least 200 mg/dl, i.e. at least 11.1 mmol/l	value meeting the criterion for diabetes

The OGTT can detect disturbances that fasting glucose does not show. Some people have normal morning glycemia, but too large or too prolonged a rise in glucose after a load.

The values above do not apply to the diagnosis of gestational diabetes, for which separate criteria apply.

Random glucose

Random glucose can be measured at any time, regardless of the last meal. A result of at least 200 mg/dl, i.e. 11.1 mmol/l, can be the basis for diagnosing diabetes if typical symptoms of hyperglycemia or a hyperglycemic crisis are present.

Characteristic symptoms include:

• increased thirst,
• frequent urination,
• unintended weight loss,
• weakness,
• blurred vision,
• recurrent infections.

In the absence of unambiguous symptoms, the result should be confirmed in line with diagnostic principles.

Insulin: important, but harder to interpret

Fasting insulin shows the hormone concentration at the time blood is drawn. The result can help assess how much insulin the body needs to maintain the current glucose level.

The problem is that, unlike glucose and HbA1c, no single, universally binding diagnostic range has been established for fasting insulin.

Why do insulin reference ranges differ between laboratories?

Insulin measurement results depend among other things on:

• the laboratory method used,
• the analyzer and reagents used,
• the population on which the reference range was defined,
• the patient's age and body composition,
• the time since the last meal,
• the current metabolic state,
• natural variability in insulin secretion.

Comparisons of different immunochemical assays showed that insulin results, and therefore HOMA-IR calculations, can differ depending on the method. In one analysis, the distribution of HOMA2-IR results differed by up to roughly twofold depending on the insulin assay used.

For this reason, a value found online, for example 5, 10 or 15 µIU/ml, should not be treated as a universal boundary separating health from disease.

What can elevated fasting insulin mean?

A higher insulin concentration with normal glucose may suggest that the body needs a stronger hormonal response to keep glucose within the normal range. This can occur with reduced insulin sensitivity.

This is not, however, a diagnosis based on a single result. Insulin should be assessed together with:

• fasting glucose,
• HbA1c,
• a possible OGTT result,
• waist circumference and body composition,
• the lipid profile,
• blood pressure,
• symptoms and coexisting diseases,
• medications taken.

A single result may also be temporarily altered by non-standard testing conditions or natural biological variability.

Is low insulin always beneficial?

No. Low insulin with low or normal glucose may be physiological and indicate good insulin sensitivity.

Low insulin with simultaneously high glucose may, however, mean that the pancreas is not secreting enough of the hormone. In such a situation, a doctor may consider additional tests, for example measuring C-peptide or antibodies associated with autoimmune diabetes.

Insulin results therefore cannot be interpreted by the rule: the lower, the better.

HOMA-IR: what it is and how it is calculated

HOMA-IR, or the Homeostatic Model Assessment of Insulin Resistance, is an indirect index of insulin resistance. It was created as a mathematical model describing the relationship between insulin secretion and fasting glucose concentration.

The original HOMA model was presented by Matthews and colleagues in 1985. Its aim was to estimate insulin resistance and beta-cell function based on simultaneously measured fasting glucose and insulin concentrations.

The HOMA-IR formula

If glucose is given in mmol/l:

HOMA-IR = fasting glucose × fasting insulin ÷ 22.5

If glucose is given in mg/dl:

HOMA-IR = fasting glucose × fasting insulin ÷ 405

Insulin is usually given in µIU/ml or mIU/l.

Example

Fasting glucose is 90 mg/dl and insulin is 10 µIU/ml.

HOMA-IR = 90 × 10 ÷ 405 = 2.22

Mathematically the result is therefore about 2.2. This does not mean, however, that insulin resistance can be diagnosed automatically on this basis.

Does HOMA-IR above 2 mean insulin resistance?

There is no single universal cut-off point for everyone.

Research studies used different values, often falling roughly between 2 and 3. In some populations the cut-off was below 2, and in others above 3 or even 4. The result depended among other things on age, sex, population origin, body weight, the criterion used to define insulin resistance and the insulin assay method.

Current diagnostic criteria for diabetes and prediabetes do not include HOMA-IR. The index can be an auxiliary tool, but does not replace formal medical assessment. The creators of the official HOMA2 calculator from the University of Oxford also emphasize that the model supports metabolic assessment but should not replace clinical examination.

HOMA-IR versus HOMA2-IR

HOMA-IR calculated with the simple formula is also referred to as HOMA1-IR.

HOMA2 is a newer, computer-based model that more accurately accounts for the nonlinear relationship between glucose and insulin. It can calculate:

• HOMA2-IR, i.e. estimated insulin resistance,
• HOMA2-%S, i.e. estimated insulin sensitivity,
• HOMA2-%B, i.e. estimated beta-cell function.

HOMA1-IR and HOMA2-IR are not the same result. They should not be compared directly, nor should a cut-off point intended for one model be used to interpret the other.

Limitations of HOMA-IR

HOMA-IR has several important limitations:

1. It depends on the insulin result, which is not fully comparable across all laboratories.
2. It assesses mainly fasting conditions, so it does not show the body's full response after a meal.
3. It is an indirect index, not a direct measurement of insulin action in tissues.
4. A single result can be variable. Studies on biological variability indicate that HOMA-IR can differ significantly in the same person between successive measurements.
5. It is not an appropriate tool in every situation, especially in people using insulin. The official HOMA2 model is intended for assessing people who are not treated with insulin.
6. It does not identify the site of insulin resistance. It does not clearly say whether the main problem concerns the liver, muscle or adipose tissue.

For this reason, HOMA-IR is best treated as one piece of a larger puzzle rather than a final diagnosis.

HbA1c: average glycemia from recent months

HbA1c, or glycated hemoglobin, forms when glucose binds to the hemoglobin found in red blood cells. The higher and more prolonged the glucose concentration, the greater the proportion of hemoglobin that becomes glycated.

HbA1c reflects the body's average exposure to glucose over roughly the last two to three months. Not all weeks, however, have the same influence on the result. Glucose concentrations from roughly the last 30 days affect HbA1c more strongly than values from three to four months earlier.

You do not need to be fasting for an HbA1c measurement.

How to interpret HbA1c?

HbA1c	Interpretation
below 5.7%, i.e. below 39 mmol/mol	normal range
5.7–6.4%, i.e. 39–47 mmol/mol	prediabetes
at least 6.5%, i.e. at least 48 mmol/mol	value meeting the criterion for diabetes

As with glucose, in a person without unambiguous symptoms, diagnosing diabetes usually requires confirming the result. The diagnostic test should be performed using an appropriately standardized laboratory method.

Does normal HbA1c rule out insulin resistance?

No. HbA1c describes above all the average glucose level, not the amount of insulin needed to maintain it.

At an early stage of reduced insulin sensitivity, the pancreas may secrete more insulin and keep glucose and HbA1c within the normal range for some time. Normal HbA1c therefore does not rule out compensatory hyperinsulinemia or increased metabolic risk.

This does not mean, however, that everyone with normal HbA1c should have insulin or HOMA-IR measured. The scope of diagnostics should depend on symptoms, risk factors and the doctor's decision.

Does HbA1c show glucose spikes?

Only indirectly. HbA1c is an averaged value.

Two people can have the same HbA1c result even though:

• the first has relatively stable glucose values,
• the second experiences alternating high rises and low drops.

HbA1c does not precisely show glycemia after a specific meal, the duration of rises or daily variability. In selected situations, additional information can be provided by an OGTT, glucometer measurements or continuous glucose monitoring, if there is an indication for it.

When can HbA1c be unreliable?

HbA1c depends not only on glucose, but also on the lifespan of red blood cells and the type of hemoglobin. The result may be distorted among other things by:

• iron-deficiency anemia,
• hemolytic anemia,
• recent significant hemorrhage,
• blood transfusion,
• some hemoglobinopathies,
• advanced kidney disease,
• erythropoietin treatment,
• pregnancy,
• situations causing a significantly shortened or prolonged red blood cell lifespan.

Shortening of red blood cell lifespan, for example in hemolysis or after acute blood loss, can cause a falsely low result. Iron deficiency, on the other hand, may be associated with overestimation. The influence of hemoglobin variants also depends on the method used by the laboratory.

If HbA1c clearly does not fit fasting glucose, OGTT results, the glucometer or CGM, you should check whether a factor disturbing its reliability is present.

How to interpret all the results together?

The most information comes not from a single number, but from the mutual relationship between the parameters.

Pattern of results	What it may mean
normal glucose, normal insulin, low HOMA-IR and normal HbA1c	normal carbohydrate metabolism, provided there are no other worrying symptoms or risk factors
normal glucose and HbA1c, but higher insulin and HOMA-IR	possible compensatory hyperinsulinemia or reduced insulin sensitivity; the result requires assessment in the clinical context
elevated fasting glucose, normal or borderline HbA1c	possible impaired fasting glucose; a check-up or additional test may be needed
normal fasting glucose, elevated HbA1c	possible glucose rises at other times of day or a factor overestimating HbA1c
normal fasting glucose, abnormal result 2 hours after OGTT	impaired glucose tolerance not visible in the fasting test
high glucose and high insulin	insulin is being secreted, but may not be enough to maintain normal glycemia
high glucose and low insulin	possible impaired insulin secretion; requires urgent assessment of the cause
HbA1c inconsistent with glucose measurements	possible disturbance of HbA1c reliability, glycemic variability or differences related to the period assessed by each test

The table presents possible interpretations, not ready diagnoses. The same results can have a different meaning in a young, active person without symptoms than in a patient with abdominal obesity, hypertension, PCOS, fatty liver or a strong family history.

Does the insulin curve diagnose insulin resistance?

During an OGTT, some laboratories measure not only glucose but also insulin at fasting and after one and two hours.

The insulin curve can show how intense and prolonged the pancreatic response to a glucose load is. Unlike glucose concentration, however, there are no widely accepted, uniform diagnostic criteria for insulin at 60 or 120 minutes.

Values encountered online, for example defining the maximum "normal insulin" after one hour, often come from single studies, the practice of a particular center or arbitrary interpretations. They should not be treated as official criteria for diagnosing disease.

The glucose curve has established diagnostic thresholds. The insulin curve remains an auxiliary test that should be assessed by a doctor taking into account glucose, sampling time, symptoms and laboratory methods.

How to prepare for the tests?

Glucose, insulin and HOMA-IR

To make results as comparable as possible:

• come in the morning after at least eight hours without eating,
• during the fast drink only water,
• do not change your medication dosing on your own,
• inform the staff about medications and supplements you take,
• perform subsequent measurements, if possible, under similar conditions and in the same laboratory,
• do not perform the test against your doctor's advice during an acute deterioration of health.

If you take antidiabetic medications or insulin, the way to prepare should be agreed with your doctor in advance. Skipping a dose on your own can be dangerous.

HbA1c

You do not need to be fasting for an HbA1c measurement. It can be done regardless of meal time. You should, however, inform your doctor about anemia, blood diseases, chronic kidney disease, pregnancy, recent bleeding or transfusion, as these can affect interpretation.

What to do when a result is abnormal?

An abnormal result should lead neither to panic nor to starting a very restrictive diet on your own.

The next steps depend on the type of deviation and may include:

• repeating fasting glucose,
• measuring HbA1c,
• performing an OGTT,
• assessing the lipid profile,
• measuring blood pressure and waist circumference,
• checking liver tests,
• assessing the blood count and iron status if HbA1c may be unreliable,
• reviewing the medications taken,
• in selected cases measuring C-peptide or antibodies associated with autoimmune diabetes.

In people with diagnosed prediabetes, screening for diabetes is usually recommended at least once a year. With normal results, the frequency depends on age and risk factors.

When to contact a doctor urgently?

High glucose requires rapid medical assessment, especially if it is accompanied by:

• very intense thirst,
• frequent passing of large amounts of urine,
• rapid weight loss,
• nausea or vomiting,
• abdominal pain,
• increasing drowsiness,
• impaired consciousness,
• deep or rapid breathing,
• clear dehydration.

Such symptoms can indicate significant hyperglycemia and acute metabolic disturbances. Do not wait for a scheduled check-up or try to correct the situation with diet alone.

Summary

Glucose, insulin, HOMA-IR and HbA1c describe different elements of carbohydrate metabolism.

Glucose shows the current blood sugar concentration. HbA1c presents average glucose exposure over recent months. Insulin informs about the hormonal response, while HOMA-IR is only an indirect estimate of insulin resistance based on fasting results.

The most important difference is their diagnostic significance. Glucose, HbA1c and the OGTT result have official thresholds for diagnosing prediabetes and diabetes. Insulin and HOMA-IR can be helpful, but do not have a single universal cut-off point and should not be interpreted in isolation from the whole health picture.

It is therefore not worth asking only: "Is my HOMA-IR too high?". A better question is: "What do all the results together say about my carbohydrate metabolism and metabolic risk?". If you want to see how reduced insulin sensitivity shows up day to day, read our article on insulin resistance and what to do without medication.

Frequently asked questions

Does normal glucose mean I do not have insulin resistance?

Not always. At an earlier stage of reduced insulin sensitivity, the pancreas can secrete more insulin, keeping glucose normal. Normal glycemia is a favorable result, but it should be interpreted together with risk factors, symptoms and the remaining tests.

What fasting insulin result is normal?

There is no single value appropriate for all laboratories and all patients. You should consider the reference range of the given laboratory, the assay method, the simultaneous glucose result and the clinical situation. A result within the laboratory range also does not always answer the question about metabolic risk.

At what HOMA-IR result does insulin resistance begin?

No single universal threshold has been established. Studies use different values, often around 2–3, but these are not official diagnostic criteria. The cut-off point depends on the population, the insulin assay method and the way insulin resistance is defined.

Does HbA1c of 5.7% mean diabetes?

No. A result of 5.7–6.4% corresponds to the prediabetes range. Diabetes is diagnosed at HbA1c of at least 6.5%, taking into account the need to confirm the result if there are no unambiguous symptoms of hyperglycemia.

Can stress raise glucose?

Yes. Acute physical stress, infection or serious illness can temporarily increase glucose concentration through the action of stress hormones. That is why a result obtained under unusual conditions should be interpreted taking the current state of health into account.

Does high insulin mean diabetes?

No. Diabetes is not diagnosed based on insulin concentration. High insulin can occur when compensating for reduced insulin sensitivity, but diagnosing diabetes relies on glucose, HbA1c or the OGTT result.

Does low insulin rule out metabolic problems?

No. Its meaning depends on the simultaneous glucose concentration. Low insulin with normal glucose may indicate good insulin sensitivity, whereas with high glucose it may indicate insufficient hormone secretion.

Can insulin resistance be diagnosed based on the insulin curve alone?

There are no uniform diagnostic criteria for insulin at one and two hours after an OGTT. The curve can be an auxiliary test, but its result should be assessed by a doctor together with the glucose curve and the whole clinical picture.

References

1. American Diabetes Association Professional Practice Committee. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes–2026. Diabetes Care. 2026;49(Suppl 1)–S49. doi:10.2337/dc26-S002.
2. Polish Diabetes Association. Clinical recommendations for the management of people with diabetes 2026.
3. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985;28(7):412–419. doi:10.1007/BF00280883.
4. Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment evaluation uses the computer program. Diabetes Care. 1998;21(12):2191–2192. doi:10.2337/diacare.21.12.2191.
5. Staten MA, Stern MP, Miller WG, Steffes MW, Campbell SE. Insulin assay standardization: leading to measures of insulin sensitivity and secretion for practical clinical care. Diabetes Care. 2010;33(1):205–206. doi:10.2337/dc09-1206.
6. Manley SE, Stratton IM, Clark PM, Luzio SD. Comparison of 11 human insulin assays: implications for clinical investigation and research. Clin Chem. 2007;53(5):922–932. doi:10.1373/clinchem.2006.077784.
7. Gayoso-Diz P, Otero-González A, Rodriguez-Alvarez MX, et al. Insulin resistance index HOMA-IR cut-off values and the metabolic syndrome in a general adult population: effect of gender and age. BMC Endocr Disord. 2013;13:47. doi:10.1186/1472-6823-13-47.
8. Esteghamati A, Ashraf H, Khalilzadeh O, et al. Optimal cut-off of homeostasis model assessment of insulin resistance in an Iranian population. Nutr Metab. 2010;7:26. doi:10.1186/1743-7075-7-26.
9. National Glycohemoglobin Standardization Program. Factors that Interfere with HbA1c Test Results. Updated: June 23, 2026.
10. National Glycohemoglobin Standardization Program. HbA1c and Estimated Average Glucose.
11. Sacks DB, Arnold M, Bakris GL, et al. Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus. Diabetes Care. 2023;46(10)–e199. doi:10.2337/dci23-0036.
12. Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ. Translating the A1C assay into estimated average glucose values. Diabetes Care. 2008;31(8):1473–1478. doi:10.2337/dc08-0545.