In This Article

The short answer: Divide your triglycerides by your HDL cholesterol (both in mg/dL) and you get a rough proxy for insulin resistance and LDL particle size that a standard lipid panel already contains. McLaughlin and colleagues found a ratio of 3.0 or higher identified insulin resistance in overweight adults, and in a Physicians' Health Study case-control analysis, people in the top quartile of the ratio carried roughly 16 times the heart attack risk of those in the bottom quartile. The ratio is not a diagnostic test. It reads reliably in white populations but has been shown to fail as a marker of insulin resistance in African Americans, so a normal ratio should not be read as reassurance on its own.

Read key takeaways →

What the triglyceride-to-HDL ratio actually measures

Every standard lipid panel reports triglycerides and HDL cholesterol as separate numbers. The ratio between them, triglycerides divided by HDL-C, both measured in mg/dL, is not a separate blood draw. It is arithmetic performed on numbers you likely already have from your last physical.

What makes the ratio useful is what it stands in for. High triglycerides paired with low HDL cholesterol is the lipid signature of insulin resistance, and that same pairing tends to travel with a shift toward smaller, denser LDL particles, the subtype most consistently linked to atherosclerosis. A single ratio cannot diagnose insulin resistance or measure LDL particle size directly, the way a clamp study or NMR-based LDL-P testing can, but it approximates both from data most people already have.

Triglycerides

Fat circulating in the blood

Rises with excess calories, refined carbohydrate, alcohol, and insulin resistance. Reported directly on every fasting lipid panel.

HDL cholesterol

Cholesterol carried by HDL particles

Tends to fall as triglycerides rise, because triglyceride-rich HDL particles are cleared from circulation faster.

TG/HDL-C ratio

A free proxy, not a new test

Calculated from two numbers you already have. Higher ratios track with insulin resistance and small, dense LDL.

How to read your ratio

McLaughlin, Abbasi, Cheal, Chu, Lamendola, and Reaven studied 258 nondiabetic overweight and obese adults, 87 percent of them non-Hispanic white, and used receiver-operating-characteristic analysis to find the ratio value that best identified insulin resistance, defined by a clamp-derived measure of steady-state plasma glucose. A triglyceride-to-HDL-C ratio of 3.0 in traditional mg/dL units, equivalent to about 1.8 when both values are converted to mmol/L, was their optimal cutoff.

The unit conversion matters because triglycerides and cholesterol do not convert from mg/dL to mmol/L by the same factor. Applying a mg/dL-derived cutoff to values reported in mmol/L, or the reverse, will give a misleading answer.

Reading the Ratio (mg/dL Units)

Well below 3.0

Consistent with the profile McLaughlin's group associated with insulin sensitivity in their overweight, mostly white cohort.

Approaching 3.0

Close enough to the cutoff that trend matters more than a single reading. Recheck alongside fasting glucose or fasting insulin.

At or above 3.0

The zone McLaughlin's ROC-optimized cutoff flagged as insulin resistant in their validation cohort. Worth a direct look at fasting insulin or HOMA-IR rather than treating the ratio alone as a verdict.

A single cutoff from one validation study is a reference point, not a hard line. Treat 3.0 as the threshold this specific study used to separate its cohort, not a universal biological boundary that applies identically to everyone.

Why triglycerides and HDL move together

The two numbers are not independent. Insulin resistance increases hepatic output of VLDL, the triglyceride-carrying particle the liver ships into circulation. Cholesteryl ester transfer protein then swaps triglycerides out of VLDL for cholesteryl esters in HDL and LDL. The result is HDL particles that are triglyceride-enriched and cholesterol-depleted, which are cleared from the bloodstream faster, and LDL particles that get remodeled by hepatic lipase into the smaller, denser subtype.

That is the mechanistic link behind the ratio: one number going up while the other goes down is not two separate problems, it is one upstream process, elevated VLDL output driven by insulin resistance, showing up in both directions at once.

The Cascade

1

Insulin resistance raises hepatic VLDL-triglyceride output

2

CETP exchanges triglycerides into HDL and LDL, cholesteryl esters out

3

Triglyceride-enriched HDL clears faster, so HDL-C falls

4

Triglyceride-enriched LDL is remodeled into small, dense particles

The misconception: one ratio, every population

The most common mistake with this ratio is treating it as a universal biological signal rather than a pattern validated in a specific population. Sumner and colleagues tested fasting triglycerides and the triglyceride-to-HDL-C ratio in 99 African American and 50 white adults between 18 and 45 years old. Fasting insulin, BMI, and waist circumference tracked with insulin resistance in both groups, but triglycerides and the triglyceride-to-HDL-C ratio tracked with insulin resistance only in the white participants, not in the African American participants.

That finding matters because the ratio's popularity comes from studies like McLaughlin's, run in cohorts that were predominantly non-Hispanic white. A normal ratio in someone from a population where the relationship does not hold is not evidence of insulin sensitivity. If you fall into that gap, a direct measure such as fasting insulin, HOMA-IR, or an oral glucose tolerance test tells you more than the ratio ever will.

A normal triglyceride-to-HDL ratio is reassuring only in the populations where the relationship has been validated. It is not a substitute for a direct insulin resistance test when your background, diet, or clinical picture puts that relationship in doubt.

What the ratio says about heart disease risk

Beyond insulin resistance, the ratio has also been studied directly against cardiovascular outcomes. Gaziano and colleagues ran a case-control study of 340 confirmed heart attack cases and 340 age-, sex-, and community-matched controls from Boston-area hospitals. People in the highest quartile of the triglyceride-to-HDL ratio had roughly 16 times the risk of heart attack compared with those in the lowest quartile, a relative risk of 16.0 with a 95 percent confidence interval of 7.7 to 33.1.

Da Luz and colleagues looked at a different outcome, the extent of existing disease rather than future events, in 374 patients undergoing coronary angiography. The triglyceride-to-HDL-C ratio was associated with more extensive coronary disease on the Friesinger index, with an odds ratio of 2.01, a similar magnitude to triglycerides and HDL cholesterol taken individually. Total cholesterol and LDL cholesterol were not significantly associated with disease extent in the same cohort.

Gaziano et al., 1997

Future heart attack risk

Case-control, 340 MI cases vs. 340 matched controls. Top vs. bottom quartile of the ratio: relative risk 16.0.

Da Luz et al., 2008

Existing disease extent

374 angiography patients. Ratio associated with more extensive coronary disease: odds ratio 2.01.

Neither study proves the ratio causes anything. Both are consistent with the same underlying story: a lipid pattern driven by insulin resistance and small, dense LDL tracks with worse cardiovascular outcomes, whether measured as a future event or existing plaque burden. For a fuller picture of what is driving that plaque burden, the full lipid panel, including ApoB, adds detail the ratio alone cannot.

What to do if your ratio is elevated

An elevated ratio is a prompt to look upstream at insulin resistance, not a number to chase in isolation. The interventions that move it are the same ones that address the mechanism behind it.

1

Cut refined carbohydrate and added sugar. These are the dietary drivers most directly linked to higher triglyceride output from the liver.

2

Lose excess visceral fat if present. Even modest weight loss improves insulin sensitivity and tends to lower triglycerides while raising HDL-C.

3

Limit alcohol. Alcohol is a direct driver of hepatic triglyceride production, independent of overall diet quality.

4

Build an aerobic training base. Regular aerobic exercise improves insulin sensitivity through mechanisms independent of weight change.

5

Retest with a fasting panel. Both triglycerides and HDL shift with the timing and content of your last meal, so a fasting draw keeps the ratio comparable across visits.

Frequently asked questions

What is a good triglyceride-to-HDL ratio?

McLaughlin and colleagues found 3.0 (mg/dL units) was the cutoff that best identified insulin resistance in their cohort of overweight, mostly white adults. A ratio well below that is consistent with the profile they associated with insulin sensitivity, but treat it as a reference point from one validation study, not a universal target.

Does the ratio replace a fasting insulin or HOMA-IR test?

No. The ratio is a free proxy calculated from numbers already on a standard lipid panel. A fasting insulin or HOMA-IR test measures insulin resistance more directly. If the two disagree, or if you fall into a population where the ratio has not been validated, trust the direct measure.

Why doesn't the ratio work the same way for everyone?

Sumner and colleagues tested the ratio in 99 African American and 50 white adults and found it tracked with insulin resistance in the white participants but not in the African American participants, even though fasting insulin and waist circumference tracked with insulin resistance in both groups. The ratio's validation studies have skewed toward white cohorts, so its reliability outside those populations is genuinely unclear.

Does the ratio predict heart disease risk on its own?

Studies have found associations, not proof of causation. Gaziano and colleagues found a 16-fold difference in heart attack risk between the top and bottom quartiles of the ratio in a case-control study, and Da Luz and colleagues found the ratio associated with more extensive coronary disease on angiography. Both are consistent with the ratio tracking an underlying insulin-resistant, small-dense-LDL pattern rather than causing risk by itself.

Should I calculate the ratio in mg/dL or mmol/L?

Use whichever units your lab reports, but do not mix a cutoff derived from one unit system with values from the other. McLaughlin's cutoff of 3.0 applies to mg/dL values; the equivalent in mmol/L is roughly 1.8, because triglycerides and cholesterol convert between the two systems by different factors.

How is the ratio connected to LDL particle size?

The same process that raises triglycerides and lowers HDL, elevated VLDL output paired with cholesteryl ester transfer protein activity, also remodels LDL particles into the smaller, denser subtype. The ratio does not measure particle size directly the way LDL-P or ApoB testing does, but it reflects the same upstream metabolic pattern.

What to Remember

  • →The triglyceride-to-HDL-C ratio is not a new test. It is triglycerides divided by HDL cholesterol, both already on a standard fasting lipid panel.
  • →McLaughlin and colleagues found a ratio of 3.0 (mg/dL units), or about 1.8 in mmol/L, best identified insulin resistance in 258 overweight and obese adults, 87 percent of them non-Hispanic white.
  • →Sumner and colleagues found the ratio tracked with insulin resistance in white adults but not in African American adults, even though fasting insulin and waist circumference tracked with insulin resistance in both groups.
  • →Gaziano and colleagues found a relative risk of 16.0 for heart attack comparing the top to the bottom quartile of the ratio in a case-control study of 340 cases and 340 matched controls.
  • →Da Luz and colleagues found the ratio associated with more extensive coronary disease on angiography (odds ratio 2.01) in 374 patients, a similar magnitude to triglycerides or HDL cholesterol alone.
  • →A high ratio is a prompt to look at insulin resistance directly, through fasting insulin or HOMA-IR, rather than a diagnosis to act on by itself.

Related on Protocol

LearnHealth

Fasting Insulin Deep Dive: The Early Warning Signal for Metabolic Drift

A direct measure of insulin resistance to run alongside the triglyceride-to-HDL ratio, especially when the ratio and your risk profile disagree.

Read
LearnHealth

How to Interpret Your Lipid Panel: LDL Particles, ApoB, and What Actually Matters

The full standard and advanced lipid panel, including how triglycerides and HDL fit alongside LDL particle measures.

Read
ProtocolHealth

The Lab Work and Biomarkers Protocol

A framework for deciding which labs to track, how often, and how to read a trend rather than a single result.

Read

Track your bloodwork trends alongside the rest of your data

Protocol connects lab results like your lipid panel to your daily wearable data, so you can see which changes actually move the numbers that matter.

Get started free

References

Key Researchers

  • Gerald Reaven (Stanford University) Pioneered the concept of insulin resistance as a unifying metabolic syndrome and senior author of the 2003 study establishing the triglyceride-to-HDL ratio cutoff.
  • Anne Sumner (National Institutes of Health) Led the research showing the triglyceride-to-HDL ratio does not reliably track insulin resistance in African Americans.

Key Studies

  • McLaughlin et al. (2003) Annals of Internal Medicine. Established the ROC-optimized cutoff of 3.0 (mg/dL) for the triglyceride-to-HDL ratio identifying insulin resistance in 258 overweight and obese, predominantly white adults.
  • Sumner et al. (2005) Archives of Internal Medicine. Found the triglyceride-to-HDL ratio tracked with insulin resistance in white adults but not African American adults.
  • Gaziano et al. (1997) Circulation. Case-control study of 340 heart attack cases and 340 matched controls finding a 16-fold difference in risk between the top and bottom quartiles of the ratio.
  • Da Luz et al. (2008) Clinics (Sao Paulo). Found the ratio associated with more extensive coronary disease on angiography in 374 patients.

Apps and Tools

  • Protocol Connects lab results, including lipid panels, to daily wearable data to track trends over time rather than isolated readings.