HOMA-IR Calculator
HOMA-IR Calculator
Calculate HOMA-IR from fasting glucose and insulin, with the derivation of the 405 divisor and an honest account of what the cut-offs can and cannot tell you.
HOMA-IR
Fasting glucose × insulinFasting glucose 105 mg/dL, fasting insulin 12 µIU/mL
Formula
HOMA-IR = insulin (µIU/mL) × glucose (mmol/L) ÷ 22.5
- 405
- the conventional-unit divisor; 22.5 × 18 = 405, because glucose in mg/dL is 18 times the figure in mmol/L, so the two forms are the same equation
- 22.5
- the SI divisor, the product of a normal fasting glucose of 4.5 mmol/L and a normal fasting insulin of 5 µIU/mL, which sets a normal result at 1.0
- fasting
- both values must come from the same genuinely fasting sample; a post-prandial insulin makes the result meaningless
- insulin assay
- assays are not standardised between manufacturers, so a HOMA-IR is comparable within a laboratory but not between laboratories
Worked example
Fasting glucose 105 mg/dL, fasting insulin 12 µIU/mL
12 × 105 = 1,260
1,260 ÷ 405 = 3.11
In SI units: 105 ÷ 18 = 5.83 mmol/L; 12 × 5.83 ÷ 22.5 = 3.11 — the same answer
Above 2.9 → significant insulin resistance
Commonly used interpretive bands
| HOMA-IR | Interpretation | Comment |
|---|---|---|
| < 1.9 | Normal insulin sensitivity | A value near 1.0 is the model’s definition of normal |
| 1.9 – 2.9 | Early insulin resistance | Read alongside waist circumference, triglycerides and HDL |
| > 2.9 | Significant insulin resistance | Common in MASLD and polycystic ovary syndrome |
The same equation in two unit systems
| Conventional | SI | |
|---|---|---|
| Glucose unit | mg/dL | mmol/L |
| Divisor | 405 | 22.5 |
| Worked example | 12 × 105 ÷ 405 = 3.11 | 12 × 5.83 ÷ 22.5 = 3.11 |
| Why they agree | mg/dL = mmol/L × 18 | 22.5 × 18 = 405 |
Where 405 comes from, and where the index stops working
HOMA-IR multiplies fasting insulin by fasting glucose and divides by a constant. The two published forms look different but are identical: the SI form divides by 22.5 with glucose in mmol/L, and the conventional form divides by 405 with glucose in mg/dL, because mg/dL is eighteen times the mmol/L figure and 22.5 × 18 = 405. The constant itself is the product of a normal fasting glucose of 4.5 mmol/L and a normal fasting insulin of 5 µIU/mL, which is what makes a normal result 1.0. Using 22.5 with a glucose in mg/dL is the commonest error here, and inflates the result eighteenfold.
The index came from a mathematical model of glucose and insulin homeostasis published by Matthews and colleagues in 1985, and was validated against the euglycaemic hyperinsulinaemic clamp, which remains the reference method for measuring insulin sensitivity. HOMA-IR is a good surrogate at a population level and a considerably rougher one in an individual, where it correlates with clamp-derived sensitivity but does not substitute for it.
The cut-offs need stating honestly. The thresholds used here, and everywhere else, are conventional rather than guideline values. They vary substantially between populations and, more importantly, between insulin assays, which are not standardised between manufacturers. A HOMA-IR of 2.6 from one laboratory is not the same quantity as a HOMA-IR of 2.6 from another. The index is therefore most useful compared against the same laboratory over time, or against a local reference population, and least useful compared against a number found online.
Two conditions make the result meaningless rather than merely imprecise. The first is exogenous insulin: the assay measures injected insulin, so HOMA-IR in a treated patient describes the prescription. The second is beta-cell failure. In type 1 diabetes, and in advanced type 2 diabetes, insulin is low because secretion has failed, not because sensitivity is good — and the equation will return a reassuringly low number for a patient who is profoundly hyperglycaemic. Both inputs must also come from the same genuinely fasting sample.
Frequently asked questions
How is HOMA-IR calculated?
Multiply fasting insulin in µIU/mL by fasting glucose in mg/dL and divide by 405. With glucose in mmol/L the divisor is 22.5 instead. Both forms give the same answer, because a glucose in mg/dL is eighteen times the mmol/L figure and 22.5 × 18 = 405.
What is a normal HOMA-IR?
The model sets a normal result at 1.0, and values below about 1.9 are usually taken as normal insulin sensitivity. These are conventional cut-offs rather than guideline thresholds, and they shift with the population studied and the insulin assay used.
Can HOMA-IR be used in someone taking insulin?
No. Insulin immunoassays detect injected insulin as well as endogenous insulin, so the result reflects the dose given rather than the patient’s own secretion. C-peptide-based measures are used instead.
Is HOMA-IR valid in type 1 diabetes?
No. The equation assumes that fasting insulin reflects the response to the prevailing glucose. In type 1 diabetes and advanced beta-cell failure, insulin is low because secretion has failed, so HOMA-IR returns a falsely reassuring value in a hyperglycaemic patient.
Why do published HOMA-IR cut-offs disagree?
Because insulin assays are not standardised between manufacturers and reference populations differ. The same serum can give different insulin values in two laboratories, and the index carries that difference through. Serial values within one laboratory are far more informative than a single value against a published threshold.
Related calculators
References
- 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.
- Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care. 2004;27(6):1487–1495.
- Knopp JL, Holder-Pearson L, Chase JG. Insulin units and conversion factors: a story of truth, boots, and faster half-truths. J Diabetes Sci Technol. 2019;13(3):597–600.
Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.