How to Calculate HOMA-IR and QUICKI From Fasting Labs

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.

Quick Answer

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.

Unit conversions for the two fasting values
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.

From lab units to HOMA-IR Two rows of boxes. The top row converts glucose from mg/dL to mmol/L by dividing by 18.0156. The bottom row converts insulin from pmol/L to uIU/mL by dividing by 6.00. Arrows from the mmol/L box and the uIU/mL box lead to a final box showing HOMA-IR equals glucose times insulin divided by 22.5. Convert both inputs, then combine them Glucose mg/dL divide by 18.0156 Glucose mmol/L Insulin pmol/L divide by 6.00 Insulin uIU/mL HOMA-IR = G x I / 22.5 G in mmol/L, I in uIU/mL Prefer glucose in mg/dL? Skip the first step and divide by 405 instead. QUICKI uses glucose in mg/dL and insulin in uIU/mL.
Convert each input to the right unit before you multiply anything.

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.

HOMA-IR and QUICKI compared
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.

  1. Convert glucose. Divide 90 by 18.0156 to get 4.996, which rounds to 5.00 mmol/L.
  2. Check the insulin unit. It is already in uIU/mL. In pmol/L it would be 10 x 6.00 = 60.0.
  3. 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.
  4. 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.

HOMA-IR rises while QUICKI falls Two horizontal bar charts using the same four illustration pairs of glucose in mg/dL and insulin in uIU/mL. Pair A is 80 and 4, pair B is 90 and 10, pair C is 100 and 20, and pair D is 110 and 25. HOMA-IR values are 0.79, 2.22, 4.93 and 6.78. QUICKI values are 0.399, 0.338, 0.303 and 0.291. Same four pairs, two indices HOMA-IR A: 80 and 4 0.79 B: 90 and 10 2.22 C: 100 and 20 4.93 D: 110 and 25 6.78 QUICKI A: 80 and 4 0.399 B: 90 and 10 0.338 C: 100 and 20 0.303 D: 110 and 25 0.291 Each pair is glucose in mg/dL and insulin in uIU/mL. Illustration numbers, not patients.
One goes up as the other goes down, because QUICKI is a reciprocal of a log sum.

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.

The plus or minus 31 percent band around 2.22 A horizontal number line from 0 to 4. A solid bar spans 1.53 to 2.91. A dark marker at 2.22 shows the example HOMA-IR result sitting in the middle of that band. One reading carries a wide band 0 1 2 3 4 2.22 1.53 2.91 HOMA-IR scale. Band = 2.22 plus or minus 31 percent. Illustration number, not a patient.
The example result of 2.22 sits inside a band 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.

NIDDK fasting plasma glucose ranges (glucose only)
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.

Have your two fasting numbers ready?

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

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-Muzaffar
Founder & Editor-in-Chief at  ~ Web ~ More Posts

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.