Estimate Your Insulin-Resistance Score | HOMA-IR Calculator

HOMA1-IR display bands (study comparison)
HOMA-IR	Reading	What it suggests
< 1.5	Lower band	A low HOMA1-IR does not exclude abnormal glucose or impaired insulin secretion. The 1.5 split is a display choice.
1.5 to < 2.35	Middle band	This band remains below the selected study’s upper reference limit. It is not an established borderline diagnosis.
≥ 2.35	At or above study limit	At or above the Gutenberg reference limit for its selected sample and insulin assay. Compare with your own lab; this does not diagnose insulin resistance.

What HOMA-IR actually measures

HOMA-IR (the Homeostasis Model Assessment of Insulin Resistance) estimates how hard your body has to work to keep blood sugar in check. It was derived in 1985 by Matthews and colleagues in Oxford ^[1], and it has become the most widely used way to gauge insulin resistance from a routine blood test, because it needs only 2 fasting values: glucose and insulin.

The idea is simple. In a healthy person, a modest amount of insulin keeps fasting glucose normal. When cells stop responding well to insulin (insulin resistance), the pancreas compensates by secreting more, so fasting insulin creeps up, often years before fasting glucose does. HOMA-IR captures that imbalance as a single number, calibrated so a metabolically normal reference person scores 1.0 ^[1].

It is a screening surrogate, not a gold-standard measurement. The reference method (the hyperinsulinaemic-euglycaemic clamp) is impractical outside research, and HOMA-IR agrees with it only moderately at the individual level ^[2]. Think of HOMA-IR as a useful orientation, not a verdict.

How it is calculated

There are 2 algebraically identical forms of the formula. They differ only by the glucose unit:

In plain words: the more insulin it takes to keep your blood sugar normal, the more insulin-resistant you are.

Show the math

Conventional (glucose in mg/dL): HOMA-IR = fasting insulin (µU/mL) × fasting glucose (mg/dL) ÷ 405.
SI (glucose in mmol/L): HOMA-IR = fasting insulin (mU/L) × fasting glucose (mmol/L) ÷ 22.5.

405 = 22.5 × 18; the glucose conversion uses the conventional rounded factor 18. Insulin in µU/mL, µIU/mL and mU/L is numerically identical. For pmol/L, we divide by 6.0 ^[9]. Laboratories also use other factors ^[16]; use the lab’s reported activity-unit value when available. Dividing by 6.945 instead gives a result about 14% lower than dividing by 6.0, not necessarily 14% below the lab’s result.

HOMA-%B = 20 × insulin (µU/mL) ÷ (glucose (mmol/L) − 3.5) ^[1]. HOMA1-%S is calculated here as 100 ÷ HOMA1-IR. Both are relative model estimates, not direct measurements of secretion or sensitivity.

At glucose ≤3.5 mmol/L, %B is not interpretable and we show no value. Near that denominator’s zero, small glucose changes can greatly change %B. The ≤4.0 mmol/L warning and >999% display cap are calculator safeguards, not published diagnostic thresholds. %S has no such denominator but retains HOMA1’s limitations.

How to read your result

HOMA1 thresholds depend on the population, assay and outcome. EPIRCE found different thresholds when using population percentiles versus metabolic-syndrome criteria ^[7]. HOMA2 thresholds belong to a different model and must be kept separate ^[8].

Gutenberg’s selected reference sample had a median of 1.09 and a 97.5th percentile of 2.35 ^[5]. These are assay-specific study values. Our 1.5 split is editorial. Bruneck instead used 2.77, its reference group’s 80th percentile ^[6]; it does not validate our bands. Bands use the displayed value rounded to two decimals.

Because insulin immunoassays are not standardised, the same blood sample can give HOMA-IR values that differ almost 2-fold between labs ^[10]^[11]. Always read your number against your own laboratory's reference range rather than treating any fixed value as absolute.

The honest limitations

HOMA-IR is genuinely useful, but it is easy to over-read. Keep these caveats in mind:

HOMA-IR at a glance

It needs true fasting values (about 10 to 12 hours). Non-fasting numbers are meaningless.
A single sample is imprecise: insulin is secreted in pulses, and the original authors recommend the mean of 3 fasting samples drawn 5 minutes apart ^[1]^[2].
Insulin assays are not standardised: the same sample can vary almost 2-fold between labs, so values are not comparable across laboratories or studies ^[10]^[11].
Do not use this calculator for self-assessment with insulin treatment, type 1 diabetes or advanced beta-cell failure. C-peptide is not an automatic workaround; specialist interpretation is needed ^[2].
It is a screening surrogate, not a diagnosis, and does not replace a clinical work-up ^[2].
These orientation bands do not apply during pregnancy; discuss interpretation with your care team.
Cutoffs need validation in the population and for the clinical outcome where they are applied ^[7]^[8].
The original equations were calibrated with older insulin assays. Oxford recommends its updated computer model for modern assays; neither model has a universal diagnostic cutoff ^[2]^[3].

HOMA1 vs HOMA2: which this is

The formula above is HOMA1: the original 1985 linear approximation. It is fast, transparent and matches the most familiar published cutoffs, which is exactly what this calculator gives you.

HOMA2 is the later nonlinear computer model ^[3], followed by the adjustable research model iHOMA2 ^[4]. HOMA2 was recalibrated for newer insulin assays. Oxford supplies the software under a licence; it has assay-specific input limits and is intended for professional interpretation.

Never transfer cutoffs between models. BRAMS reported separate HOMA1-IR >2.7 and HOMA2-IR >1.8 thresholds in its Brazilian sample ^[8]. Those are study cutoffs, not a formula for converting an individual’s HOMA1 result to HOMA2. Extreme values can also fall outside HOMA2’s accepted inputs.

Beyond HOMA-IR: open advanced indices

The advanced panel adds research indices with different assumptions. A newer or more complex model is not automatically more accurate for an individual.

SPINA-Carb estimates receptor gain (GR) and beta-cell capacity (GBeta) from a mechanistic model; DI is their product ^[12]^[13]. The displayed ranges are preliminary research ranges ^[13]. SPINA-DI depends strongly on fasting glucose and does not replace an OGTT or independently confirm a HOMA result.

TyG = ln(triglycerides × glucose ÷ 2), with both inputs in mg/dL ^[14]^[17]. Some publications instead use ln(triglycerides × glucose) ÷ 2; their numbers and cutoffs differ. METS-IR = ln(2 × glucose + triglycerides) × BMI ÷ ln(HDL), with all blood values in mg/dL ^[15]. Its marker near 50 reflects a Mexican cohort, not a European reference limit.

For the updated model: Oxford HOMA2 information and licensing

Frequently asked questions

How do I read a HOMA-IR result?

Read the index alongside your fasting glucose and your lab’s reference range. Our lower and middle bands are display categories, not diagnoses. A high result merits clinical interpretation; a low result does not exclude impaired insulin secretion or abnormal glucose ^[2].

What is a good HOMA-IR value?

There is no universal target. Compare with an assay-matched laboratory reference and your clinical context. The 1.5 split used here is not a validated clinical cutoff ^[5].

Do I need to be fasting?

Yes. HOMA-IR is only valid on fasting (basal) glucose and insulin from the same blood draw, ideally after 10 to 12 hours without food. Non-fasting values make the result meaningless.

My lab reports insulin in pmol/L. What do I enter?

Select pmol/L and enter the lab value. We divide by 6.0 ^[9]. Laboratories may use different factors ^[16], so prefer their reported µU/mL or mU/L value when available.

Why can the same blood give different HOMA-IR at 2 labs?

Because insulin immunoassays are not standardised. Across 11 commercial assays the same samples produced almost 2-fold differences in HOMA-IR ^[10]^[11]. That is why you should read your number against the reference range of the lab that measured it.

Is HOMA-IR a diagnosis of insulin resistance or prediabetes?

No. It is a screening surrogate, not a diagnostic test, and it does not replace a clinical assessment or the gold-standard clamp study ^[2]. Treat an elevated result as a prompt to repeat the test and talk to your doctor.

Should I use HOMA1 or HOMA2?

This tool calculates the simplified HOMA1 equations. Oxford’s HOMA2 is a different, updated model for use with professional guidance. It has its own input limits; do not transfer HOMA1 cutoffs to HOMA2 ^[3]^[8].

Is this a HOMA2 calculator, and does it give HOMA-β (beta-cell function)?

This is a HOMA1 calculator: from the same fasting glucose and insulin it returns HOMA-IR plus HOMA-%B (beta-cell function, from the original 1985 model ^[1]) and HOMA-%S, the insulin-sensitivity figure (the reciprocal 100 / HOMA-IR). It is not the non-linear HOMA2 computer model. HOMA2-IR, HOMA2-%B and HOMA2-%S come from the Oxford HOMA2 program, which we link to for values at the extremes; never mix HOMA1 and HOMA2 cutoffs ^[8]. For an open, mechanistic beta-cell and sensitivity read, the advanced panel also computes SPINA-Carb natively.

What is the HOMA index, and is it the same as HOMA-IR?

“HOMA index” usually means HOMA-IR. This tool calculates the simplified HOMA1 version and displays comparison bands, whose limitations are explained above.

Can I use C-peptide instead of fasting insulin?

No. This HOMA1 equation requires fasting insulin. Oxford HOMA2 includes a separate C-peptide mode, but that does not automatically make it suitable for type 1 diabetes or insulin treatment. Ask your clinician which assessment is appropriate ^[2]^[3].

Sources

Created by Maurice Lichtenberg, Founder, Longevity Cities

Track the numbers that actually move with you

Longevity is a practice, and it is better with others. Join the community testing, learning and comparing notes on the markers that matter.

Explore the community