Skip to content
Homepage
Free · Private · Published formula

HOMA-IR Calculator

Enter a fasting glucose and a fasting insulin to get your HOMA-IR score, the index most clinics use to estimate insulin resistance from a single blood draw. The formula is the original Matthews 1985 equation, the bands are explained rather than asserted, and the mg/dL and mmol/L conversions are handled for you.

  • Calculator
  • About 3 min
  • Matthews 1985 HOMA equation
  • Runs in your browser, nothing is sent
  • Medical review: Medical review pending
Open the calculator

What this calculator works out

Fasting glucose alone stays normal for years while the pancreas quietly pushes out more and more insulin to keep it there. HOMA-IR multiplies the two numbers so that a normal glucose held up by a high insulin shows as what it is: resistance.

  1. 1

    The HOMA-IR index

    Fasting glucose in mg/dL times fasting insulin in uU/mL, divided by 405. A score near 1.0 describes a person whose glucose is held normal with little insulin; higher scores mean more insulin is needed for the same glucose. Reported to two decimals.

  2. 2

    Where the score sits

    Four educational bands: under 1.0, 1.0 to 1.9, 1.9 to 2.9, and 2.9 and above. The page explains where each cut-off comes from, which population studies support it, and why laboratories and clinics disagree on the exact lines.

  3. 3

    Units, handled

    US labs report glucose in mg/dL and insulin in uU/mL (also written uIU/mL or mIU/L). If your report uses mmol/L or pmol/L, the conversion is in the input help so the 405 constant is applied to the right numbers.

How this compares with a typical online quiz

FeatureTypical free quizRegenerated.com
Formula named and sourcedSometimesMatthews 1985, constant 405 for mg/dL
Unit conversion shownRarelymg/dL to mmol/L and uU/mL to pmol/L in the help
Bands explained, not just coloredNoYes, with the population studies behind each line
Says when HOMA-IR is misleadingNoYes: insulin treatment, high glucose with low insulin, non-fasting draws
QUICKI alternate explainedSometimesYes, as prose with its scale
Answers sent to a serverOftenNever; scored in your browser, counts-only events

Where the 1.0, 1.9 and 2.9 lines come from

The original HOMA paper did not publish cut-offs; it calibrated the model so that a healthy young adult scores about 1.0. The lines used on this page are the ones most US clinics and laboratories quote: under 1.0 insulin sensitive, 1.0 to 1.9 typical, 1.9 to 2.9 early insulin resistance likely, and 2.9 and above significant insulin resistance. They are educational markers, not guideline thresholds, and they were chosen because they bracket what population studies have found.

Those studies land in a similar place. In the Bruneck study in northern Italy, Bonora and colleagues treated the top fifth of HOMA-IR among metabolically healthy adults, a value of 2.77, as the start of insulin resistance. In the Spanish EPIRCE study, Gayoso-Diz and colleagues found the best cut-off for identifying the metabolic syndrome was about 2.05 overall, lower in men than in women and rising with age. Studies in other populations report optimal cut-offs anywhere from about 1.8 to 3.0. The honest reading is that scores below about 1.9 are reassuring, scores above about 2.9 are consistently associated with metabolic trouble, and the range in between depends on who you are and which study you trust.

QUICKI, the logarithmic cousin

QUICKI, the quantitative insulin sensitivity check index published by Katz and colleagues in 2000, uses the same two fasting numbers but takes logarithms: 1 divided by (log of fasting insulin in uU/mL plus log of fasting glucose in mg/dL). It runs the other way, with higher values meaning better insulin sensitivity. Values around 0.38 and above are typical of insulin-sensitive adults, values in the low 0.30s describe insulin resistance, and the index correlates somewhat better with the clamp method, the research reference standard, than HOMA-IR does at the resistant end of the range.

Because QUICKI is a transformation of the same inputs, it ranks people in the same order as HOMA-IR; it does not add information, it compresses the scale. For the example on this page, glucose 90 mg/dL and insulin 10 uU/mL, QUICKI is 1 divided by (1.000 plus 1.954), which is 0.339. This page computes HOMA-IR only; the QUICKI value is here so you can read a report that uses it.

What moves fasting insulin that is not insulin resistance

A single insulin value is noisier than a single glucose value. Insulin assays differ between laboratories by 20 to 30 percent, insulin rises within minutes of eating, and stress, poor sleep, acute illness, steroid medicines and some other drugs raise it. A draw that was not truly fasting (ten to twelve hours, water only) will overstate your score.

HOMA-IR is unreliable in anyone taking insulin, because the measured insulin is the injected dose rather than what the pancreas is making. It is also misleading when glucose is high and insulin is low or normal: that combination can describe a pancreas that is no longer keeping up, including type 1 or latent autoimmune diabetes in adults, and the multiplication can produce a deceptively modest score. If your fasting glucose is 126 mg/dL or higher, the glucose itself is the finding to take to a clinician, whatever the index says.

Methodology and sources

The calculator implements the homeostasis model assessment of insulin resistance published by Matthews, Hosker, Rudenski and colleagues in Diabetologia in 1985. The model was built from a mathematical description of the glucose and insulin feedback loop and calibrated against clamp and intravenous glucose tolerance data in adults, so that a normal-weight person under 35 with normal glucose scores about 1.0. The approximation used here, and by almost every laboratory, is the simple product: fasting glucose in mg/dL times fasting insulin in uU/mL divided by 405, which is the same as glucose in mmol/L times insulin divided by 22.5. Its limits are the ones Wallace, Levy and Matthews set out in their 2004 review: the sample must be fasting, the insulin assay must be a specific one (not a cross-reacting assay that counts proinsulin), the index is not valid in people on insulin, and the updated computer model (HOMA2) handles very high and very low values better than the simple product.

How this is calculated: a fasting glucose of 90 mg/dL and a fasting insulin of 10 uU/mL give 90 times 10, which is 900, divided by 405, which is 2.22. The same person in SI units has a glucose of 5.0 mmol/L (90 divided by 18) and the same insulin, so 5.0 times 10 divided by 22.5 is also 2.22. If your laboratory reports insulin in pmol/L, divide by 6.945 to get uU/mL before entering it; most US and European laboratories use that factor, although a few use 6.0, so check your report.

The bands on the result follow the lines most clinics quote: under 1.0, 1.0 to 1.9, 1.9 to 2.9, and 2.9 and above. They are educational, not diagnostic. The population evidence behind them comes from the Bruneck study (Bonora 1998, cut-off 2.77 for isolated insulin resistance) and the EPIRCE study (Gayoso-Diz 2013, optimal cut-off about 2.05 for the metabolic syndrome, varying by sex and age). Reference ranges differ between laboratories, and two fasting draws on different days tell a clinician more than one.

Medical review: Medical review pending. Last updated .

References

  1. 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
  2. Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care. 2004;27(6):1487-1495. doi:10.2337/diacare.27.6.1487
  3. Katz A, Nambi SS, Mather K, et al. Quantitative insulin sensitivity check index: a simple, accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab. 2000;85(7):2402-2410. doi:10.1210/jcem.85.7.6661
  4. Gayoso-Diz P, Otero-Gonzalez A, Rodriguez-Alvarez MX, et al. Insulin resistance (HOMA-IR) cut-off values and the metabolic syndrome in a general adult population: effect of gender and age: EPIRCE cross-sectional study. BMC Endocr Disord. 2013;13:47. doi:10.1186/1472-6823-13-47
  5. Bonora E, Kiechl S, Willeit J, et al. Prevalence of insulin resistance in metabolic disorders: the Bruneck Study. Diabetes. 1998;47(10):1643-1649. doi:10.2337/diabetes.47.10.1643

HOMA-IR Calculator FAQ

The model was calibrated so that a healthy young adult scores about 1.0. Most clinics treat scores under about 1.9 as typical and scores of 2.9 or more as significant insulin resistance, with the range between as a gray zone. Population studies put their best single cut-off between about 2.0 and 2.8 depending on the group studied, so there is no universal line. Reference ranges differ between laboratories.

Glucose in mg/dL and insulin in uU/mL, which is the same unit as uIU/mL and mIU/L. If your glucose is reported in mmol/L, multiply by 18. If insulin is in pmol/L, divide by 6.945 (a few laboratories use 6.0; the report usually says). The sample must be fasting, usually ten to twelve hours with water only.

Not if you inject insulin, because the measured insulin is the injected dose rather than your own production, and the score has no meaning. GLP-1 medicines such as semaglutide and tirzepatide do not invalidate the arithmetic, but they lower both glucose and insulin, so a score taken on treatment describes the treated state, not the underlying condition. Tell your clinician what you were taking at the time of the draw.

A high fasting glucose with a low or normal insulin can mean the pancreas is no longer producing enough insulin, which is a different problem from resistance and includes type 1 and latent autoimmune diabetes in adults. A fasting glucose of 126 mg/dL or more on two occasions meets the diabetes threshold on its own. Take the glucose to a clinician; the index is not the finding in that situation.

Yes. Your answers and the numbers you enter are scored in your browser and are never sent to us. They stay in this browser tab so you can return to your result, and closing the tab clears them. The only thing the page sends us is a counts-only event: which test was started or completed, the result band and the headline number, with no answers, no entered values and no account. There is nothing to sign up for.

Find clinics that evaluate insulin resistance properly

Browse US clinics offering peptide and GLP-1 therapy. Look for ones with physician oversight that repeat fasting labs, check HbA1c and lipids, and explain the risks and alternatives before prescribing. A listing on Regenerated.com is not an endorsement.

Browse peptide therapy clinics

A listing is not a treatment endorsement. Read each clinic's published checks and talk to your clinician before starting any treatment.

Important: This calculator is educational arithmetic on numbers you enter. It is not a diagnosis of insulin resistance, prediabetes or diabetes, and not a recommendation for or against any treatment. Insulin assays and reference ranges differ between laboratories, and only a clinician who sees your full results and history can interpret them. If your symptoms are severe or sudden, seek urgent medical care.