HOMA-IR Calculator
Medically reviewed by Dr. Chhaya Makhija, MD, DipABLM — September 2026
Calculate your HOMA-IR (Homeostatic Model Assessment of Insulin Resistance) from fasting glucose and fasting insulin. The most widely used research index for estimating insulin resistance. Educational reference only.
🔬 HOMA-IR Calculator
HOMA-IR Result
β-cell function %
mg/dL
µIU/mL
HOMA-IR Reference Ranges
| HOMA-IR Value | Interpretation | Clinical Context |
|---|---|---|
| < 1.0 | Optimal insulin sensitivity | Low metabolic risk |
| 1.0 – 1.9 | Normal range | Acceptable in most adults |
| 2.0 – 2.9 | Borderline | Monitor; consider lifestyle review |
| 3.0 – 4.9 | Insulin resistance likely | Discuss with provider; metabolic syndrome risk |
| ≥ 5.0 | Significant insulin resistance | High risk; clinical evaluation warranted |
Thresholds vary by study and population. Some references use ≥ 2.5 as the resistance threshold. Discuss your result in context with your provider.
HOMA-IR Formula
Developed by Matthews et al. in 1985, HOMA-IR uses a mathematical model of the fasting glucose-insulin feedback loop. The denominator 405 converts units so the result is dimensionless (normal ≈ 1.0 in a healthy non-diabetic adult). The mmol/L version uses a divisor of 22.5 instead of 405.
A note on insulin units: most US labs report fasting insulin in µIU/mL (same as µU/mL) — use the 405 formula. Labs reporting insulin in pmol/L must convert first: µIU/mL = pmol/L ÷ 6.945. So 60 pmol/L ≈ 8.6 µIU/mL. Enter the µIU/mL value into this calculator.
For the concepts behind the score — including its role in PCOS, NAFLD and prediabetes — see our guide What Is HOMA-IR?
What your HOMA-IR score is really telling you
HOMA-IR takes two numbers from a single fasting blood draw — your glucose and your insulin — and answers one practical question: how hard is your body working to keep blood sugar normal? When insulin is doing its job well, a modest amount holds glucose in range and your score stays low. When your cells start tuning insulin out, the pancreas compensates by releasing more of it. Your glucose can still look perfectly "normal" on a standard test, but that quietly elevated insulin is what pushes HOMA-IR up — which is why the score often spots a problem years before fasting glucose alone does.
That head start is the whole point. A slowly climbing HOMA-IR is frequently the first measurable hint of the drift toward prediabetes, type 2 diabetes, fatty liver, and the wider cluster of changes called metabolic syndrome. Treat it as a screening signal rather than a verdict, though — one value on one morning can't capture the full picture, and it only means something when you read it alongside your weight, blood pressure, lipids, and family history.
Why fasting properly changes your result
HOMA-IR is only valid on genuinely fasting blood. Anything with calories in the hours beforehand lifts both glucose and insulin, inflating the score and making it unreliable. The accepted standard is an 8–12 hour overnight fast with nothing but water, followed by an early-morning draw before you've eaten, exercised hard, or had coffee.
Common reasons a score reads higher than expected
A carb-heavy dinner the night before, a fast that was too short, a recent illness or poor night's sleep, acute stress, and some medications such as steroids can all nudge insulin upward temporarily. Insulin assays also differ between labs, so a small shift between tests may reflect the lab, not your body. If a result surprises you, the most useful next step is usually to repeat it under clean, consistent fasting conditions before drawing any conclusions.
HOMA1 vs HOMA2: the version problem nobody mentions
The formula above — the one this calculator uses, and the one almost every online tool uses — is HOMA1, published by Matthews and colleagues in 1985. There is a second, later model called HOMA2, and the difference matters more than most calculators let on.
HOMA1 is a linear approximation that was calibrated against the insulin assays available in the 1970s. Those assays are not the ones your lab runs today. The practical consequence, described by the model's own authors, is that HOMA1 tends to underestimate insulin sensitivity and overestimate beta-cell function when applied to modern results.
HOMA2, released in 1998 by Levy, Matthews and colleagues, is a non-linear computer model rather than a hand equation. It corrects for three things HOMA1 cannot represent: the difference between hepatic and peripheral glucose resistance, the way the insulin secretion curve changes above 10 mmol/L (180 mg/dL), and the contribution of circulating proinsulin. It can also run on C-peptide instead of insulin, which is useful if you are already on injected insulin.
| HOMA1 (1985) | HOMA2 (1998) | |
|---|---|---|
| Form | Single equation you can do by hand | Non-linear computer model |
| Assay calibration | 1970s radioimmunoassay | Recalibrated for current assays |
| Accepts C-peptide | No | Yes |
| Handles glucose > 180 mg/dL (10.0 mmol/L) | Poorly — assumes linearity | Yes — models the secretion curve |
| Accounts for proinsulin | No | Yes |
| Reports | A single index (≈1.0 is normal) | %S and %B as percentages of a normal reference population |
| Best used for | Tracking your own trend over time | Absolute values, research, comparing across groups |
What this means for you in practice. If you are watching whether your own number is moving in the right direction between two tests at the same lab, HOMA1 — this calculator — is fine, because the calibration error is consistent and cancels out of the comparison. If you need an absolute value, or you are comparing yourself against published population figures, or your fasting glucose is above 180 mg/dL, use HOMA2 instead. The official HOMA2 calculator is free from the University of Oxford's Diabetes Trials Unit, which maintains the model.
How your score compares to real populations
Reference bands like the table above are simplified for readability. The underlying population data is messier and more useful, because it shows how wide the normal spread actually is. These figures come from NHANES, the US national health survey.
| Population | Typical HOMA-IR | What it tells you |
|---|---|---|
| US adults without diabetes | Median 2.2 · mean 2.8 (SD 2.4) | The mean sits above the usual "normal" ceiling — the distribution has a long right tail |
| US adolescents, normal weight | Mean 2.3 | Adolescents run physiologically higher; puberty raises insulin resistance |
| US adolescents, obese | Mean 4.9 | More than double the normal-weight figure in the same age group |
| Asian populations | Cutoffs commonly 1.4 – 2.5 | Resistance appears at lower BMI and lower HOMA-IR than in white European cohorts |
The single most common misreading of HOMA-IR follows from that first row: a score of 2.6 is above the tidy "normal" band but is still below the average American adult without diabetes. A number slightly over the line is a prompt to look at the trend and the rest of your metabolic picture, not evidence of disease.
Why your lab's cutoff may not match ours
There is no universally accepted HOMA-IR threshold. No major endocrine society has ratified one, and the number your lab flags as abnormal depends on choices made by whoever set its reference range. Three things drive the variation.
Sex. Thresholds that separate normal from abnormal differ measurably between men and women. Work in older adults has put the cutoff for type 2 diabetes at roughly 2.25 in men and 2.03 in women, and for pre-diabetes at about 1.73 in men and 1.85 in women — note that the sexes swap order between the two conditions.
Ancestry. Asian cohorts consistently show insulin resistance at lower HOMA-IR values and lower BMI than European ones, which is why cutoffs of 1.4–2.5 are used where a US lab might use 2.5–3.0.
The assay itself. Insulin immunoassays are not standardised between manufacturers the way glucose assays are. Two labs can return meaningfully different insulin values on the same blood, and because insulin is one of only two inputs, that difference propagates straight into your score. This is the practical reason to retest at the same lab when you are tracking a trend.
| Context | Commonly used threshold |
|---|---|
| NHANES / general US screening | ≥ 2.5 |
| Type 2 diabetes, men | ≈ 2.25 |
| Type 2 diabetes, women | ≈ 2.03 |
| Pre-diabetes, men | ≈ 1.73 |
| Pre-diabetes, women | ≈ 1.85 |
| Asian population cohorts | 1.4 – 2.5 |
Steps that can lower a high HOMA-IR over time
Insulin resistance responds to the same habits that improve metabolic health overall, and the changes don't have to be dramatic to register. Losing even 5–7% of body weight, walking after meals, building muscle through resistance training, easing back on refined carbohydrates and sugary drinks, and protecting your sleep all help your cells respond to insulin again. Improvements tend to show up over weeks to months rather than days, so retest under the same fasting conditions to compare fairly — and use the trend, with your provider's input, to guide what's working.
Where clinical guidance gets more specific than "exercise more" is on the type of exercise. The combination with the strongest support is resistance training or high-intensity interval work, roughly two to three sessions a week, paired with a Mediterranean or predominantly plant-based eating pattern. Resistance training matters here for a mechanical reason: skeletal muscle is the largest site of insulin-mediated glucose disposal in the body, so adding muscle increases the tissue available to clear glucose.
If lifestyle change alone does not move the number, medication is a conversation worth having with your clinician. Metformin has the longest track record for improving insulin sensitivity; GLP-1 receptor agonists and the dual GLP-1/GIP agonists work substantially through weight reduction. None of these are self-prescribable, and none replace the lifestyle foundation — but a high HOMA-IR that stays high despite genuine effort is a reasonable trigger for that discussion rather than a reason to try harder alone.
Frequently Asked Questions
What is a normal HOMA-IR level?
A HOMA-IR under about 1.0 reflects optimal insulin sensitivity, and roughly 1.0–1.9 is normal for most healthy adults. Values of 2.0–2.9 are borderline, and 3.0 or above suggests insulin resistance. Cut-offs vary by population and lab, so interpret the result with your provider.
How is HOMA-IR calculated?
HOMA-IR = (fasting glucose in mg/dL × fasting insulin in µIU/mL) ÷ 405. If glucose is in mmol/L, divide by 22.5 instead. Both values must come from the same fasting blood sample.
What does a high HOMA-IR mean?
A high HOMA-IR means your body is producing more insulin than normal to keep glucose in range — a sign of insulin resistance, which is linked to type 2 diabetes, metabolic syndrome, and PCOS. It is a screening signal, not a diagnosis.
Do I need to fast before a HOMA-IR test?
Yes. HOMA-IR requires fasting glucose and fasting insulin, typically after 8–12 hours without food. Non-fasting values make the result invalid because eating raises both glucose and insulin.
Can HOMA-IR be normal if I still have insulin resistance?
Yes, and this is the score's most important blind spot. HOMA-IR is driven by how much insulin your pancreas is producing. In longer-standing type 2 diabetes, beta cells become exhausted and insulin output falls even though glucose stays high — so the score drops while the underlying resistance is unchanged or worse. A falling HOMA-IR alongside rising glucose is a warning sign, not an improvement. The score is most reliable in people whose beta cells are still compensating normally.
Can I use HOMA-IR if I take insulin injections?
Not with the standard formula. An insulin assay cannot distinguish the insulin you injected from the insulin you produced, so the result is meaningless. The workaround is C-peptide, which your body releases in step with its own insulin but is not present in injected insulin. HOMA2 accepts C-peptide as an input for exactly this reason, so anyone on exogenous insulin should use the HOMA2 model rather than this one.
What is the difference between HOMA-IR and HOMA2-IR?
HOMA-IR (HOMA1) is the 1985 hand equation this calculator uses. HOMA2 is a 1998 non-linear computer model that corrects for hepatic versus peripheral glucose resistance, the shape of the insulin secretion curve above 180 mg/dL (10.0 mmol/L), and circulating proinsulin. Because HOMA1 was calibrated to 1970s insulin assays, it tends to underestimate insulin sensitivity against modern lab results. Use HOMA1 to track your own trend at one lab; use HOMA2 for absolute values or population comparisons.
Does a high HOMA-IR mean I have diabetes?
No. HOMA-IR is not a diagnostic test and no endocrine society has ratified a cutoff for diagnosis. Diabetes and prediabetes are diagnosed from fasting glucose, HbA1c or an oral glucose tolerance test. A raised HOMA-IR indicates your pancreas is working harder than usual to hold glucose in range, which often precedes those diagnoses by years — it is a reason to test properly, not a result to act on alone.
Why does my lab use a different HOMA-IR cutoff?
Because there is no standard one. Cutoffs shift with sex (roughly 2.25 in men versus 2.03 in women for type 2 diabetes), with ancestry (Asian cohorts commonly use 1.4–2.5 where US labs use 2.5–3.0), and with the insulin assay itself, which is not standardised between manufacturers the way glucose assays are. Two labs can return different insulin values on the same blood sample, and that difference flows straight into the score. Retest at the same lab when you are tracking a change.
How much does a HOMA-IR test cost?
You pay for two ordinary lab tests, not for a product called HOMA-IR. The score is arithmetic performed afterwards on fasting glucose and fasting insulin. Fasting glucose is inexpensive and often bundled into a standard metabolic panel. Fasting insulin costs more because it needs an immunoassay. Direct-to-consumer lab panels in the United States typically price the pair between roughly $30 and $100 without insurance; hospital pricing and other countries vary widely. When a clinician orders both tests for a medical reason, insurance frequently covers them. Ask the lab to quote "fasting glucose" and "fasting insulin" separately — that is what you are buying.
What are the common signs of insulin resistance?
Insulin resistance often produces no symptoms at all, which is why a blood test finds it before a person notices anything. When signs do appear, clinicians commonly look for: weight that concentrates around the waist, acanthosis nigricans (dark velvety patches on the neck, armpits or groin), skin tags, strong afternoon fatigue or hunger after meals, raised triglycerides with low HDL, elevated blood pressure, and irregular periods or PCOS in women. None of these confirms insulin resistance on its own. They justify testing fasting glucose and fasting insulin, which is where HOMA-IR comes from.
HOMA-IR vs HbA1c — what is the difference?
The two measure different things at different stages. HbA1c reports how much glucose has attached to haemoglobin over the previous two to three months, so it describes glucose that has already run high. HOMA-IR compares fasting insulin against fasting glucose in a single sample, so it describes how hard your pancreas works to keep glucose normal. Insulin resistance usually rises years before HbA1c moves, because extra insulin holds glucose in range until the pancreas can no longer keep up. HbA1c diagnoses diabetes and prediabetes; HOMA-IR does neither. Use HbA1c for diagnosis, and HOMA-IR as an earlier research-grade signal. Our HbA1c calculator converts A1c to average glucose.
Sources
- Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. "Homeostasis model assessment: insulin resistance and β-cell function from fasting plasma glucose and insulin concentrations in man." Diabetologia. 1985;28(7):412–419. — the original HOMA1 model implemented by this calculator.
- Levy JC, Matthews DR, Hermans MP. "Correct homeostasis model assessment (HOMA) evaluation uses the computer program." Diabetes Care. 1998;21(12):2191–2192. — establishes that the computer model, not the 1985 equation, should be used for absolute values. Link
- Wallace TM, Levy JC, Matthews DR. "Use and Abuse of HOMA Modeling." Diabetes Care. 2004;27(6):1487–1495. — the authors' own account of how the model is misapplied; source of the HOMA1 calibration caveat.
- University of Oxford, Diabetes Trials Unit. HOMA2 Calculator. — the maintained HOMA2 model; defines %B and %S as percentages of a normal reference population.
- National Institute of Diabetes and Digestive and Kidney Diseases. Insulin Resistance & Prediabetes. — background on insulin resistance and its progression.
HOMA-IR is a screening estimate, not a diagnostic test, and no endocrine society has ratified a single universal cutoff. Where the studies above disagree on thresholds, this page reports the disagreement rather than picking one number.
Written by Ryan Mitchell · Last updated: September 2026