Calculating HOMA-IR: what the index says about insulin sensitivity

HOMA-IR is an index that condenses your fasting glucose and your fasting insulin into a single number. You work it out by multiplying the two values and dividing by 22.5, with glucose in mmol/l and insulin in mU/l.

That divisor of 22.5 belongs to mmol/l. Use an American formula with 405 and it expects glucose in mg/dl, so Dutch numbers will produce nonsense.

The result says something about how much insulin your body needs to keep your blood sugar normal. The higher the number, the more effort that takes.

What is the formula exactly?

HOMA-IR is fasting glucose times fasting insulin, divided by 22.5. Both values must come from the same draw, after a night of eight to twelve hours without food. A non-fasting measurement makes the result worthless.

A worked example makes it concrete.

Say your fasting glucose is 5.4 mmol/l and your fasting insulin is 14 mU/l. The calculation is 5.4 times 14 is 75.6, divided by 22.5 is 3.36. Your glucose in that example was entirely normal.

That is exactly why this index exists. It makes visible what an isolated glucose value hides.

Where does the index come from?

From a British study in 1985 that used a computer model to predict which fasting concentrations correspond to varying degrees of insulin resistance and reduced beta-cell function. The researchers showed that two fasting values are enough to estimate both (Matthews et al., 1985).

That was the breakthrough: no tolerance test needed for a rough estimate.

The gold standard for insulin sensitivity is the euglycaemic clamp, in which glucose and insulin are infused for hours while blood sugar is held constant. That test is accurate, time-consuming and unsuited to routine use.

HOMA-IR approximates that result from two tubes of blood. Less precise, endlessly more practical.

Which value is normal?

There is no internationally fixed threshold, and that is this index's main limitation. Unlike glucose, where the Dutch NHG guideline names clear cut-offs, the HOMA-IR thresholds in use differ by laboratory and by study population.

Values around 1 are generally regarded as favourable, and above roughly 2.5 reduced sensitivity is more often discussed.

Treat those numbers as direction, not diagnosis. A technical problem plays in as well: insulin assays are not interchangeable between all laboratories, because different methods give different results. Two labs can report a different insulin value on the same blood, and therefore a different index.

So prefer comparing your own values over time, measured at the same laboratory, rather than comparing your number against someone else's.

When does HOMA-IR say something, and when not?

It is usable in people without diabetes and without insulin therapy. Once your pancreas structurally makes less insulin, or you are injecting insulin, the formula no longer measures what it is meant to. The index assumes your own insulin production shapes the picture.

In advanced diabetes that assumption no longer holds.

Someone with exhausted beta-cell function can show a misleadingly favourable HOMA-IR, simply because little insulin is being made. In prediabetes and at normal glucose values, the index is at its most useful.

A cheaper alternative exists too. Triglycerides divided by HDL cholesterol, both in mmol/l, proved a usable simple marker in American research with a cut-point around 1.8 in SI units (McLaughlin et al., 2003). Those two numbers already sit on any lipid panel.

How do you have it measured?

With one fasting draw containing both glucose and insulin. Many standard blood panels include glucose but not insulin, and without that second value the index cannot be calculated. So ask for it explicitly.

Eight to twelve hours without food, water is fine.

The standalone HOMA-IR test contains exactly the two values the formula needs. Anyone wanting the wider picture will find glucose, insulin and HbA1c alongside liver, kidney and fat values in the GLP-1 baseline panel. A BIG-registered doctor assesses the result.

Discuss an abnormal result with your GP before drawing conclusions from it. The index is a tool, not a diagnosis.

What I would look at myself

The insulin on its own, alongside the index. HOMA-IR is a handy summary, but it hides which of the two numbers is doing the work. An index of 3.4 driven by a high insulin at normal glucose is a different thing from the same 3.4 driven by a climbing glucose.

What strikes me about this index is how often the wrong divisor gets used.

Online calculators rarely ask explicitly which unit you are entering. Type 5.4 into a tool expecting mg/dl and out comes a number that means nothing. Always check whether the divisor is 22.5 or 405, and which unit belongs to it.

References

• Matthews DR et al. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985. PMID 3899825
• McLaughlin T et al. Use of metabolic markers to identify overweight individuals who are insulin resistant. Ann Intern Med. 2003. PMID 14623617
• NHG guideline on type 2 diabetes, cut-off values for fasting glucose.
• RIVM, figures on diabetes mellitus in the Netherlands (2024).

Every blood test result includes a professional assessment from a BIG-registered doctor. Discuss your result with your GP before changing anything about your treatment.