Your fasting insulin tells you how hard your pancreas is working. Your fasting glucose tells you the result. HOMA-IR, short for Homeostatic Model Assessment of Insulin Resistance, combines the two into a single number for how insulin resistant you are, and you can work it out yourself from one ordinary fasting blood draw. Here is what it means, how to calculate it, and what a good score looks like.
The quick answer
HOMA-IR estimates insulin resistance from your fasting glucose and fasting insulin. A low score means your body holds a normal blood sugar with little insulin, which is what you want. A high score means it takes a lot of insulin to do the same job, the signature of insulin resistance. Under about 1.0 is ideal; much above 2.0 suggests resistance is setting in. It is a research-grade screen you can run on yourself, and it often catches trouble years before glucose or HbA1c move.
What HOMA-IR actually measures
The idea is intuitive. Insulin is the hormone that moves sugar out of your blood and into your cells. If your cells respond well, a little insulin keeps blood sugar normal. If they resist, your pancreas has to pump out far more insulin to get the same result. HOMA-IR captures that relationship by multiplying your fasting glucose and fasting insulin together, so it essentially asks how much insulin you are running against the blood sugar it is holding in check. The model was built to approximate the euglycemic clamp, the elaborate gold-standard test of insulin resistance, from nothing more than a fasting sample [1].
How to calculate it
You need only two fasting numbers, both of which you can order yourself. In US units, with glucose in mg/dL and insulin in µU/mL:
HOMA-IR = (fasting glucose in mg/dL × fasting insulin in µU/mL) ÷ 405
For example, a fasting glucose of 90 and a fasting insulin of 8 give (90 × 8) ÷ 405, or about 1.8. If your lab reports glucose in mmol/L, divide by 22.5 instead of 405. The insulin units usually need no conversion, since µU/mL, mU/L, and mIU/L are all the same number; only if your insulin is reported in pmol/L do you divide it by about 6 first to get µU/mL.
Or let the calculator do it. Enter your fasting glucose and insulin below and it works out your HOMA-IR instantly, right in your browser.
Estimate your insulin resistance
Enter the numbers from a fasting blood draw (US units, mg/dL). The more you fill in, the more estimates you get. Everything is computed in your browser, nothing is sent or saved.
Fill in your numbers above to see your estimates.
These are estimates, not a diagnosis. The thresholds are rough and fit some people better than others; the TG:HDL ratio in particular is less reliable in people of African ancestry, who often run low triglycerides regardless of insulin resistance. Of the three, HOMA-IR is the most direct, because it already uses your insulin, while TG:HDL and TyG are proxies you can read without an insulin test. So if a proxy looks high and you have not measured insulin, getting a fasting insulin, and the HOMA-IR it gives you, is the more direct next step. A high TG:HDL also hints at small, dense LDL, which is a good reason to check your ApoB.
Want the validated version instead of a proxy? The NMR LipoProfile reports the LP-IR insulin-resistance score and your LDL particle count directly. It is a specialized, pricier test, but the most direct lipoprotein read.
What your score means
As a rough guide for US assays:
| HOMA-IR | What it suggests |
|---|---|
| Under ~1.0 | Insulin sensitive, the ideal zone |
| ~1.0 to 1.9 | Typical; watch the trend |
| ~2.0 and up | Insulin resistance becoming likely |
| Above ~3 | Marked insulin resistance |
Treat these as approximate. Because insulin assays are not standardized between laboratories, the exact cutoffs shift with the method used, and normal values differ by population. That makes HOMA-IR most useful for tracking your own trend over time, or comparing before and after a change, rather than as a hard diagnostic line [2].
What it is good for, and its limits
HOMA-IR's strength is catching insulin resistance early and cheaply. Because the pancreas compensates for resistance by making more insulin long before blood sugar rises, a climbing HOMA-IR can flag the problem years ahead of glucose or HbA1c [3]. Its limits are worth knowing too: it needs a true fasting draw, the insulin assay varies between labs, and it was designed as a research and screening tool rather than a formal clinical diagnostic, so no single number is a diagnosis [2]. Read it as one strong clue in a pattern, not a verdict.
How to lower it
Lowering HOMA-IR means the same thing as reversing insulin resistance, and it responds well to ordinary changes:
- Lose visceral fat. Even modest weight loss sharply improves insulin sensitivity.
- Cut refined carbohydrates and added sugar, the biggest drivers of the insulin your pancreas has to produce.
- Move your muscles. Both aerobic exercise and resistance training help muscle take up glucose with less insulin.
- Protect your sleep. Short or broken sleep worsens insulin resistance on its own.
These are the same levers that move fasting insulin and fasting glucose, which makes sense, since HOMA-IR is built from them.
How it compares to other tests
HOMA-IR, fasting insulin, and the lipid clues all circle the same problem from different angles. Fasting insulin alone is most of the signal, and HOMA-IR refines it by accounting for your glucose at the same time; see fasting insulin vs HbA1c for why insulin moves first. The triglyceride to HDL ratio is a free proxy hiding on any lipid panel. And HbA1c lags all of them, rising only once blood sugar finally climbs. To order the inputs together, the Harbor panel puts fasting insulin and HbA1c on one draw with your heart markers, and the metabolic testing guide shows how to order them.
The bottom line
HOMA-IR is one of the highest-yield numbers in metabolic health that almost no standard checkup bothers to calculate. From two fasting values you can order yourself, it estimates insulin resistance years before it shows up as high blood sugar, and you can do the arithmetic on the back of your lab report. Aim under 1, watch the trend, and treat a rising score as an early and fixable warning.
Roughly, under about 1.0 is ideal and 1.0 to 2.0 is typical, while above about 2.0 suggests insulin resistance. The exact cutoffs vary by lab and population, so your trend over time matters more than a single value.
Not exactly. It is an estimate of insulin resistance from a single fasting sample, a close stand-in for the elaborate gold-standard clamp test rather than a direct measurement. In practice it tracks insulin resistance well, which is why it is so widely used.
No. A high score means insulin resistance, which usually arrives years before blood sugar rises. Diabetes and prediabetes are diagnosed from glucose and HbA1c, not HOMA-IR, so a high number is an early and fixable warning, not a diagnosis.
Yes. Above about 2.0 suggests insulin resistance, and by roughly 2.9 to 3 it is fairly marked. It is worth acting on, and it responds well to the same changes that lower fasting insulin.
Often weeks to a few months. Because HOMA-IR reflects your current insulin rather than a long-term average, it can move fairly quickly once you cut refined carbohydrates, lose visceral fat, and start exercising.
In US units, multiply your fasting glucose (mg/dL) by your fasting insulin (µU/mL) and divide by 405. In mmol/L, divide by 22.5 instead. Both numbers must come from the same fasting blood draw.
Yes. HOMA-IR uses fasting glucose and fasting insulin, so a 9 to 12 hour overnight fast is needed for a valid result.
They answer different questions. HOMA-IR catches insulin resistance early, while HbA1c diagnoses prediabetes and diabetes once blood sugar has already risen. See the fasting insulin vs HbA1c guide.
Yes, and often quickly. Losing visceral fat, cutting refined carbohydrates, exercising, and sleeping well all improve insulin sensitivity and bring the score down.
- 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.Reaven GM. The insulin resistance syndrome: definition and dietary approaches to treatment. Annu Rev Nutr. 2005;25:391-406. doi:10.1146/annurev.nutr.24.012003.132155