Two calculations show up most often when people try to estimate insulin resistance from routine blood work. They answer a similar question with different inputs, and the right choice often comes down to which lab values you already have.
Quick answer: HOMA-IR uses fasting glucose and fasting insulin. The TyG index uses fasting triglycerides and fasting glucose, so it needs no insulin test. In research the two correlate moderately to well, and TyG has been proposed as a cheaper stand-in where insulin is not measured. Neither is a diagnosis, cutoffs vary between studies, and both need a true fasting sample.
How each is calculated
- HOMA-IR = (fasting glucose in mg/dL x fasting insulin in uIU/mL) / 405. With glucose in mmol/L, divide by 22.5 instead. Try our HOMA-IR calculator.
- TyG index = natural log of (fasting triglycerides in mg/dL x fasting glucose in mg/dL / 2). Try our TyG index calculator.
HOMA, short for homeostasis model assessment, was introduced by Matthews and colleagues in the 1980s and is still the most common insulin-resistance estimate in research. The TyG index was proposed in the late 2000s by Mexican researchers (Simental-Rodriguez, Guerrero-Romero, and colleagues), who compared it with the clamp method in follow-up work.
Worked examples
The numbers below were calculated for this article to show the arithmetic. They are illustrations, not reference ranges.
| Person | Fasting glucose | Fasting insulin | Triglycerides | HOMA-IR | TyG |
|---|---|---|---|---|---|
| A | 90 mg/dL | 6 uIU/mL | 100 mg/dL | 1.3 | 8.41 |
| B | 98 mg/dL | 8 uIU/mL | 150 mg/dL | 1.9 | 8.90 |
| C | 105 mg/dL | 15 uIU/mL | 180 mg/dL | 3.9 | 9.15 |
| D | 110 mg/dL | 22 uIU/mL | 250 mg/dL | 6.0 | 9.53 |
Step by step for Person C: HOMA-IR = 105 x 15 / 405 = 1,575 / 405 = 3.89. TyG = ln(180 x 105 / 2) = ln(9,450) = 9.15. Both climb together across the four people, which is the pattern researchers see when the estimates agree. They do not always agree for an individual, because triglycerides and insulin respond to different things.
What each tells you
HOMA-IR reflects how much insulin your body needs to hold glucose steady, which is closer to the mechanism. The TyG index reflects the fat and sugar combination that tends to accompany insulin resistance. Studies use a range of cutoffs for both, and there is no universal threshold; see our article on the TyG index explained. The practical use is to track your own values over time using comparable fasting tests.
Side by side
| HOMA-IR | TyG index | |
|---|---|---|
| Inputs | Fasting glucose and fasting insulin | Fasting triglycerides and fasting glucose |
| Needs a special test? | Yes: fasting insulin, which is not on every panel | No: uses a standard lipid panel and glucose |
| Closer to mechanism? | Yes, insulin is the hormone in question | Indirect, through triglycerides |
| Main weakness | Insulin assays differ between labs | Triglycerides swing with meals, alcohol, and weight changes |
| Universal cutoff? | No | No |
| Good for | Research and follow-up with consistent assays | Large-scale screening and routine labs |
Strengths and limits
- Availability. TyG uses a lipid panel and glucose that are nearly always ordered. Fasting insulin often is not.
- Insulin assays differ, so HOMA-IR values vary between labs.
- Triglycerides are noisy. A recent meal or drinks can inflate TyG; see fasting vs nonfasting triglycerides.
- Neither is a substitute for clinical assessment, A1C, or a glucose tolerance test.
- Not designed for everyone. They are less reliable in people who inject insulin, in type 1 diabetes, and when glucose is very high.
Lab values that feed the formulas
For orientation, the American Diabetes Association describes a fasting glucose below 100 mg/dL as normal, 100 to 125 mg/dL as prediabetes range, and 126 mg/dL or higher (confirmed) as diabetes range. Fasting triglycerides below 150 mg/dL are generally considered normal. If you have a result in mmol/L, convert before calculating: glucose mmol/L x 18.0 = mg/dL, and triglycerides mmol/L x 88.6 = mg/dL. The MedlinePlus triglycerides page explains the ranges.
Why insulin resistance is hard to measure
Insulin resistance is not a single lab value. It describes how much insulin the body needs to move glucose into cells, and it differs between liver, muscle, and fat tissue. The reference method is a hyperinsulinemic-euglycemic clamp, in which insulin is infused while glucose is adjusted to keep blood sugar steady. It takes hours and specialist staff, so it is used in research, not in clinic. Everything else, including HOMA-IR and the TyG index, is an estimate built from a fasting blood sample. That is why different estimates can disagree and why clinicians read them alongside waist size, blood pressure, A1C, and liver tests.
Reading results over time: an example
Because single values are noisy, the most useful habit is to compare like with like. The table shows three fasting draws for one fictional person, computed with the formulas above.
| Visit | Glucose | Insulin | Triglycerides | HOMA-IR | TyG |
|---|---|---|---|---|---|
| 1 | 98 mg/dL | 6 uIU/mL | 120 mg/dL | 1.45 | 8.68 |
| 2 | 101 mg/dL | 9 uIU/mL | 150 mg/dL | 2.24 | 8.93 |
| 3 | 97 mg/dL | 7 uIU/mL | 110 mg/dL | 1.68 | 8.58 |
Visit 2 looks worse on both measures, but the glucose changed by only four points while insulin and triglycerides rose. That is the kind of swing a late meal, a poor night’s sleep, a few drinks, or a different insulin assay can cause. Visit 3 moves back. A clinician would repeat the test and look at the pattern before drawing a conclusion, which is the same approach to use at home.
What can move each number
- Fasting length. Too short a fast raises both triglycerides and insulin.
- Alcohol. Drinks the night before can raise triglycerides.
- Recent illness, poor sleep, or hard exercise, which can temporarily change glucose and insulin.
- Medicines. Some drugs change triglycerides, glucose, or insulin. Ask the prescriber; do not change any medicine yourself.
- Weight change and body composition, which affect triglycerides and insulin over months.
Which estimate fits which situation
| Situation | Likely more useful | Why |
|---|---|---|
| You have a lipid panel and glucose but no insulin | TyG | Needs no insulin test |
| Your clinician ordered fasting insulin | HOMA-IR | Directly uses insulin |
| Triglycerides are known to be unstable | HOMA-IR | Less affected by meals and alcohol |
| You want a quick third view from a standard lipid panel | Triglyceride/HDL ratio | Uses only the lipid panel |
| You are comparing yourself with a study | The same estimate and cutoff the study used | Cutoffs are not interchangeable |
If you want to look at glucose from the A1C side, see the A1C to average glucose chart. And if you are looking at triglycerides specifically, diet for high triglycerides covers what the research says moves them.
Unit conversions for the inputs
Labs outside the United States often report glucose and triglycerides in mmol/L. The formulas on this page use mg/dL, so convert first.
| Glucose (mg/dL) | Glucose (mmol/L) | Triglycerides (mg/dL) | Triglycerides (mmol/L) |
|---|---|---|---|
| 90 | 5.0 | 100 | 1.13 |
| 100 | 5.6 | 150 | 1.69 |
| 110 | 6.1 | 200 | 2.26 |
| 126 | 7.0 | 500 | 5.65 |
Insulin is usually reported in uIU/mL in the United States and in pmol/L elsewhere; divide pmol/L by about 6.9 to get uIU/mL before using the mg/dL formula. If your lab reports everything in SI units, the HOMA-IR formula with glucose in mmol/L divided by 22.5 gives the same answer as the mg/dL version divided by 405.
Why published cutoffs differ
You will see different “insulin resistant” thresholds in different papers. The reasons are practical: studies use different populations (children, older adults, different ancestries), different assays, different reference standards, and different outcomes such as prediabetes, fatty liver, or heart disease. A cutoff chosen to flag one outcome in one population may misclassify people in another. For that reason this page explains how to calculate the numbers and compare them over time rather than naming a single line.
How to use them
Pick one, calculate it from comparable tests over time, and watch the trend rather than a single value. The complete guide to insulin resistance explains the wider picture, and the triglyceride/HDL ratio calculator offers a third view. The NIDDK page on prediabetes and insulin resistance is a good background reference. These calculators support a conversation with your clinician; they do not tell you whether to start or stop any treatment.
Common mistakes
- Calculating from a non-fasting sample.
- Mixing units, for example entering glucose in mmol/L into a mg/dL formula.
- Comparing a value from one lab’s insulin assay with another lab’s.
- Treating one result as a diagnosis; repeat on a comparable test.
- Changing diet, alcohol, or activity the week before and expecting the numbers to match a typical week.
Questions to bring to your appointment
- Which estimate of insulin resistance is most useful for me?
- Should we measure fasting insulin?
- Were my samples truly fasting?
- What would a meaningful change look like, and when should we retest?
- Do my other numbers, such as A1C and blood pressure, point the same way?
Frequently asked questions
Which is more accurate, TyG or HOMA-IR?
Neither is a gold standard. The reference method is a hyperinsulinemic-euglycemic clamp, which is a research procedure. Both estimates correlate moderately to well with clamp results in studies, and which looks better depends on the population and outcome. HOMA-IR uses insulin and reflects the mechanism more directly; TyG avoids an insulin assay and is easy to compute from routine labs.
Do I need a fasting insulin test for TyG?
No. The TyG index uses fasting triglycerides and fasting glucose, both of which are on most routine panels. That convenience is the main reason it has been studied as a stand-in when insulin is not measured.
What is a normal HOMA-IR?
There is no single universal normal value. Lower numbers mean less insulin is needed to keep glucose steady. Published cutoffs for insulin resistance vary by study, population, and assay, and many labs print their own range. Ask the lab or your clinician which range applies to your result.
What is a normal TyG index?
Also not universal. Published thresholds vary by population and outcome, which is why this article does not give a single cutoff. A more reliable use is to compare your own values over time using comparable fasting tests.
Do I need to fast?
Yes, for both. The standard is 8 to 12 hours with water only. A recent meal or alcohol can raise triglycerides sharply, which inflates TyG, and a non-fasting insulin level makes HOMA-IR unreliable.
Can I use these if I take insulin or diabetes medicine?
HOMA-IR is not designed for people who inject insulin, and medicines can alter glucose and insulin levels. If you take diabetes medicine, discuss with your clinician how to interpret any estimate. Do not stop or change medicine because of a calculation.
Can these replace an A1C or glucose tolerance test?
No. They estimate insulin resistance, which is different from diagnosing prediabetes or diabetes. A1C, fasting glucose, and an oral glucose tolerance test are the diagnostic tools; the estimates are context.
About the author: Charles Kirkland is the founder of Metabolic Clarity Labs and author of The Journey Back. He is not a physician. This is general education, not a medically reviewed treatment plan.
Sources checked October 31, 2026. Discuss personal decisions with a qualified clinician. Do not start, stop, or change prescribed treatment based on this article.
Leave a Reply