Your lab report gives glucose in one unit and insulin in another, so how do you turn them into one index? The answer is a short conversion, then one small formula.
HOMA-IR equals fasting glucose times fasting insulin, divided by 22.5, with glucose in mmol/L and insulin in uIU/mL. With glucose in mg/dL, divide by 405 instead. QUICKI is 1 divided by the sum of the base-10 logs of insulin and glucose. Both are rough estimates for education, not a diagnosis.
What HOMA-IR and QUICKI Estimate
Both indices turn two fasting blood values into one score that estimates insulin resistance. HOMA stands for homeostatic model assessment. Matthews and colleagues first described it in 1985 to quantify insulin resistance and beta-cell function.
HOMA-IR needs only fasting glucose and fasting insulin. A review of insulin-sensitivity methods says it suits large studies where those two values are all you have. In the original work, the formulas tracked the euglycemic clamp, a reference lab method, with a correlation of 0.88.
QUICKI uses the same two inputs but adds a log step and a reciprocal. The review calls it virtually identical to the simple HOMA form apart from that log. Direction matters: a higher HOMA-IR points toward more insulin resistance, while a lower QUICKI points the same way.
Both inputs must be fasting values, so the sample matters as much as the math. You can skip the arithmetic with our fasting insulin index calculator. Knowing the steps still helps you catch a mistyped unit, which is the most common slip.
Convert Your Lab Units First
Units cause most errors, so start there. Glucose may arrive in mg/dL or mmol/L. Insulin may arrive in uIU/mL (the u means micro) or pmol/L.
| Value | You have | You want | Rule |
|---|---|---|---|
| Glucose | mg/dL | mmol/L | Divide by 18.0156 |
| Glucose | mmol/L | mg/dL | Multiply by 18.0156 |
| Insulin | pmol/L | uIU/mL | Divide by 6.00 |
| Insulin | uIU/mL | pmol/L | Multiply by 6.00 |
Read the unit printed beside each result, not just the number. If your lab prints glucose as 5.4 mmol/L, multiply by 18.0156 to see about 97 mg/dL, which QUICKI needs. The glucose factor comes from its molar mass of 180.156 g/mol. A Diabetes Technology Society paper gives 6.00 as the correct insulin factor for human insulin. Other factors circulate, so check which one your lab used.
The Two Formulas Side by Side
HOMA-IR multiplies the two values and divides by a constant. QUICKI adds their logs and flips the result. The table compares them.
| Feature | HOMA-IR | QUICKI |
|---|---|---|
| Inputs | Fasting glucose and insulin | Fasting glucose and insulin |
| Formula | Glucose x insulin / 22.5 (mmol/L) or / 405 (mg/dL) | 1 / (log10 insulin + log10 glucose) |
| Units needed | Either glucose unit, with the matching constant | Glucose in mg/dL, insulin in uIU/mL |
| More resistance shows as | A higher number | A lower number |
The two HOMA forms agree because 405 divided by 22.5 equals 18, the glucose conversion factor. Tiny gaps in the second decimal appear because the exact factor is 18.0156.
The 22.5 constant rests on an assumption. A normal-weight person under 35 with 100 percent beta-cell function would score exactly 1. The simple formulas are crude estimates near normal glucose and insulin, and the authors prefer their HOMA2 computer model where possible.
A Worked Example, Line by Line
Let us use illustration numbers, not a patient: fasting glucose of 90 mg/dL and fasting insulin of 10 uIU/mL.
- Convert glucose. Divide 90 by 18.0156 to get 4.996, which rounds to 5.00 mmol/L.
- Check the insulin unit. It is already in uIU/mL. In pmol/L it would be 10 x 6.00 = 60.0.
- Find HOMA-IR. Multiply 5.00 by 10 to get 50, then divide by 22.5. The result is 2.22. The mg/dL check, 90 x 10 / 405, also gives 2.22.
- Find QUICKI. The base-10 log of 10 is 1, and the log of 90 is 1.954. The sum is 2.954, and 1 divided by 2.954 is 0.338.
Now try a lab that reports SI units: glucose 5.4 mmol/L and insulin 72 pmol/L. Divide 72 by 6.00 to get 12.0 uIU/mL. Then 5.4 x 12.0 / 22.5 equals 2.88. For QUICKI, 5.4 mmol/L is about 97 mg/dL.
How the Two Indices Move Together
As the glucose-insulin product rises, HOMA-IR climbs and QUICKI falls. The chart shows four illustration pairs, not patients. Row B is the worked example above.
Compare rows A and D. The glucose-insulin product is 320 in row A and 2,750 in row D, about 8.6 times larger. HOMA-IR is also about 8.6 times larger, because it simply scales that product.
QUICKI changes far less, from 0.399 to 0.291, because logs squeeze large differences into a small range. That is why the two indices are not read on the same scale. The review also quotes one study with QUICKI group averages of 0.382, 0.331 and 0.304 for non-obese, obese and diabetic subjects. Those are one study’s averages, not cut-offs. Do not use them to grade your own result.
Why One Reading Is Only a Rough Guide
A single result is an estimate, not a verdict. The tool page shows a band of plus or minus 31 percent around the result and calls 31 percent the coefficient of variation the original authors reported. For 2.22, that band runs from 1.53 to 2.91.
Both inputs also vary from day to day. The glucose test itself is done in the morning after fasting at least 8 hours, according to NIDDK. The table shows NIDDK’s glucose ranges for people who are not pregnant.
| Fasting glucose | NIDDK label | Applies to |
|---|---|---|
| 99 mg/dL or below | Normal | The glucose test only |
| 100 to 125 mg/dL | Prediabetes | The glucose test only |
| 126 mg/dL or above | Diabetes | The glucose test only |
These ranges describe glucose, not HOMA-IR or QUICKI. A diagnosis rests on tests your doctor orders and repeats. The sources we read give no cut-off for either index, and cut-offs differ by lab, assay and population. A clinician interprets the result.
Common Calculation Mistakes
Most wrong answers come from unit slips, not from the formula itself. Watch for these.
- Mixing glucose units. Use 22.5 only with mmol/L, and 405 only with mg/dL.
- Using 22.5 with mg/dL values. That gives a wrong number.
- Mixing insulin units. Convert pmol/L to uIU/mL, and note it if you use a factor other than 6.00.
- Using a non-fasting sample. NIDDK describes fasting as at least 8 hours.
- Comparing labs. Indices from different labs and assays may not line up.
- Treating one reading as a diagnosis. It is a rough estimate, as the band above shows.
Other lab numbers have their own plain-language guides. See what your A1C number means, the cholesterol ratio explained, eGFR and kidney function, and the waist-to-height ratio.
Enter them in the fasting insulin index calculator to see HOMA-IR and QUICKI side by side. Then ask your doctor what your own results mean.
FAQs About HOMA-IR and QUICKI
What does HOMA-IR stand for?
It comes from the homeostatic model assessment. Matthews and colleagues first described it in 1985 as a way to estimate insulin resistance from fasting glucose and fasting insulin.
Can I use glucose in mg/dL in the HOMA-IR formula?
Yes. Multiply glucose by insulin and divide by 405 instead of 22.5. The two forms agree because 405 divided by 22.5 equals 18.
How do I convert insulin from pmol/L to uIU/mL?
Divide by 6.00 for human insulin. For example, 72 pmol/L is 12.0 uIU/mL. Labs sometimes use other factors, so check which one yours used.
Do I need to fast before the blood draw?
Yes. NIDDK says the fasting glucose test is done in the morning after fasting at least 8 hours. Both inputs to these indices should be fasting values.
Why do HOMA-IR and QUICKI move in opposite directions?
HOMA-IR multiplies glucose by insulin, so it rises as they rise. QUICKI takes logs and then a reciprocal, so it falls as the same product rises.
What is a normal HOMA-IR?
No single number fits everyone. Cut-offs differ by lab, assay and population, and our sources give none. Ask your doctor how to read your own result.
Can HOMA-IR diagnose insulin resistance or diabetes?
No. It is a rough estimate with a wide band around any single result. A diagnosis rests on tests your doctor orders and repeats.
Sources and Further Reading
References Used in This Article
- Wikipedia, Homeostatic model assessment (HOMA formulas, constants and background), accessed 2026-10-15.
- PubMed Central, Assessment of insulin sensitivity/resistance (review of HOMA-IR, QUICKI and Matsuda), accessed 2026-10-15.
- PubMed Central, Diabetes Technology Society paper on insulin units and conversion factors, accessed 2026-10-15.
- National Institute of Diabetes and Digestive and Kidney Diseases, Diabetes tests and diagnosis, accessed 2026-10-15.
This article is general education and cannot diagnose insulin resistance or any condition. Ask your doctor what your own results mean. Reviewed for accuracy by Dr. Abdullah Khalil, MBBS. Last updated 2026-10-15.
Author
Shakeel is the Founder and Editor-in-Chief of MultiCalculators.com, with over 15 years of experience in digital publishing and online tool development. He oversees editorial quality, formula accuracy, and usability to ensure the platform delivers reliable, practical calculators for users worldwide.




