What is a good HDL level?

A reassuring HDL level is generally above 40 mg/dL for men and above 50 mg/dL for women, though Fishtown Medicine weighs the triglyceride-to-HDL ratio and ApoB more heavily than the HDL number alone. A low HDL by itself is not treated as a target to chase directly, since it is more a marker of the surrounding metabolic pattern than an independent cause of risk.

How long does it take for exercise to raise HDL?

Exercise typically raises HDL over 3 to 6 months of consistent training, sometimes longer, which is why it is the slowest lipid marker to respond among the standard panel. Resistance training and sustained aerobic exercise both help, largely through the same improvement in insulin sensitivity that lowers triglycerides and shrinks visceral fat.

Should I take niacin to raise my HDL?

No. Niacin reliably raises the HDL number, but 2 large trials (AIM-HIGH and HPS2-THRIVE) found that adding it to statin therapy did not lower heart attacks or strokes, and in one trial it caused meaningful side effects. Fishtown Medicine does not prescribe niacin to raise HDL, since the goal is lowering cardiovascular risk, not moving a single lab value.

Does alcohol raise HDL, and should I drink for that reason?

Alcohol can raise HDL modestly, which is the origin of the old idea that moderate drinking protects the heart. Fishtown Medicine does not recommend starting or continuing alcohol for this reason, since the overall health cost of regular drinking outweighs the small lipid benefit, and the same insulin-sensitivity work that helps HDL long term is undermined by regular alcohol use.

What is a good HDL level?

A reassuring HDL level is generally above 40 mg/dL for men and above 50 mg/dL for women, though Fishtown Medicine weighs the triglyceride-to-HDL ratio and ApoB more heavily than the HDL number alone. A low HDL by itself is not treated as a target to chase directly, since it is more a marker of the surrounding metabolic pattern than an independent cause of risk.

How long does it take for exercise to raise HDL?

Exercise typically raises HDL over 3 to 6 months of consistent training, sometimes longer, which is why it is the slowest lipid marker to respond among the standard panel. Resistance training and sustained aerobic exercise both help, largely through the same improvement in insulin sensitivity that lowers triglycerides and shrinks visceral fat.

Should I take niacin to raise my HDL?

No. Niacin reliably raises the HDL number, but 2 large trials (AIM-HIGH and HPS2-THRIVE) found that adding it to statin therapy did not lower heart attacks or strokes, and in one trial it caused meaningful side effects. Fishtown Medicine does not prescribe niacin to raise HDL, since the goal is lowering cardiovascular risk, not moving a single lab value.

Does alcohol raise HDL, and should I drink for that reason?

Alcohol can raise HDL modestly, which is the origin of the old idea that moderate drinking protects the heart. Fishtown Medicine does not recommend starting or continuing alcohol for this reason, since the overall health cost of regular drinking outweighs the small lipid benefit, and the same insulin-sensitivity work that helps HDL long term is undermined by regular alcohol use.

Why Isn't My HDL Going Up? The Lipid Marker That Lags Behind

TL;DR30-second take

HDL cholesterol is usually the slowest lipid marker to respond to lifestyle change, and it can drop even while triglycerides, LDL, and blood sugar are improving. That is not a sign the plan is failing. HDL rises mainly with muscle-building exercise and meaningful weight change more than diet fat changes alone, and it climbs after triglycerides and glucose have already moved. Fishtown Medicine also reads HDL alongside ApoB, the triglyceride-to-HDL ratio, and Lp(a) rather than chasing the HDL number on its own, since raising HDL directly with medication has not been shown to lower heart attack risk the way lowering ApoB does.

TL;DR: When someone rebuilds their metabolic health, triglycerides usually fall first, LDL and blood sugar follow, and HDL is often the last number to move, sometimes drifting the wrong way for months before it turns around. That pattern is normal rather than a sign of failure. HDL responds mainly to muscle and sustained exercise, less to diet alone, and it needs longer than the other lipids to catch up. What matters most for your risk is not the HDL number by itself, but the pattern it sits inside: the triglyceride-to-HDL ratio, ApoB, and Lp(a).

Patients bring me lab panels that tell a good story almost everywhere. Total cholesterol at its best in years, triglycerides down, LDL trending the right direction, blood sugar improving. Then one number will not move, or worse, has dropped since the last draw: HDL.

It is a fair thing to notice, and it deserves an honest answer rather than a reassurance that skips the mechanism. Here is what I want you to know about why HDL behaves this way, and what it does and does not mean for your heart.

Why does HDL stay low, or drop, while everything else improves?

HDL stays low or drops during a metabolic turnaround because it is downstream of the same process, insulin resistance, that is also driving the numbers that are improving faster.

When insulin resistance is active, the liver makes more triglyceride-rich particles (VLDL). A protein called CETP (cholesterol ester transfer protein) then trades cholesterol out of your HDL particles for those triglycerides, shrinking the HDL particle and making it clear out of the bloodstream faster. The net effect is a lower HDL number, and that exchange keeps happening as long as triglycerides run high, even while you are doing the right things.

Triglycerides tend to improve first, LDL and blood sugar follow a few months later, and HDL is usually the slowest of the group to turn around. Triglycerides respond fast to less refined carbohydrate and less alcohol, sometimes within weeks. HDL depends on the CETP exchange calming down and on new HDL particles being built, both of which take longer and respond to a narrower set of levers. A patient's HDL sliding a few points lower in the same season their triglycerides fell by 30 points is common, and it usually reflects that CETP exchange still running rather than the plan failing.

What raises HDL?

What raises HDL is a shorter list than most people expect, and diet changes alone rarely move it much on their own.

Resistance and aerobic exercise, sustained over months. This is the most reliable lever. Regular training raises HDL by improving how your body clears triglycerides and by building the muscle that soaks up glucose, which lowers the insulin resistance driving the CETP exchange in the first place.
Meaningful weight loss, particularly loss of visceral fat around the organs. Small amounts of weight change rarely move HDL; the effect shows up once insulin sensitivity itself has shifted.
Stopping smoking. Smoking lowers HDL directly, and quitting is one of the fastest, most reliable ways to see it move.
Less alcohol, generally. Alcohol can raise HDL modestly, which is where the old "a glass of wine is good for your heart" idea came from, but the overall health cost of regular drinking outweighs that small lipid benefit, so we do not recommend starting or continuing alcohol to raise HDL.
Omega-3 fatty acids, with a modest effect on their own, mostly useful as part of the broader plan rather than a fix by itself.

Diet changes that lower saturated fat can lower HDL along with LDL, since both often move together, so a falling HDL number during active diet change is not automatically bad news. Read it alongside where your triglycerides and LDL are heading.

Does raising the HDL number protect your heart?

Raising the HDL number does not reliably protect your heart on its own, and this is the part that surprises most patients.

For years, a low HDL was treated as a problem to solve directly, and pharmaceutical companies built drugs to raise it. Niacin (vitamin B3) at high doses reliably raises HDL, and 2 large trials, AIM-HIGH and HPS2-THRIVE, tested whether adding it to good statin therapy lowered heart attacks and strokes. It did not, and in the second trial it caused meaningful side effects without a benefit. Later, large genetic studies (Mendelian randomization, which uses genes that naturally raise or lower a marker to test whether that marker is a true cause of disease or just a bystander) found that people with genetically higher HDL were not protected from heart attacks the way the observational data had suggested.

A low HDL number reads more as a marker of an unhealthy metabolic pattern than as a direct cause of heart disease on its own. It travels with high triglycerides, insulin resistance, and small, dense LDL particles, and it is that whole pattern that raises risk more than the HDL number by itself. This is why we do not chase HDL directly with medication, and it is also why watching HDL drift for a few months during genuine metabolic improvement is not a reason to worry.

What should you track instead?

What you should track instead of the HDL number alone is the pattern around it, since that pattern carries the risk information HDL alone does not.

Your triglyceride-to-HDL ratio. Divide your triglycerides by your HDL. A falling ratio is one of the clearest signs your insulin sensitivity is improving, even when the HDL half of the equation is moving slowly.
ApoB. The direct count of the particles that cause plaque. This is the number with the strongest causal evidence, and it is what we treat to a target rather than HDL.
Lp(a). A largely inherited particle worth testing once in a lifetime, independent of what your other lipids are doing.
Non-HDL cholesterol. Total cholesterol minus HDL, a rough stand-in for ApoB when advanced testing is not available, that is not thrown off by a lagging HDL number.

A patient whose triglyceride-to-HDL ratio and ApoB are both improving is on the right track even in a season where HDL itself has not caught up yet.

Guidance from the clinic

"I want patients to stop treating HDL like a report card grade they failed. It is the slowest lipid marker to respond, and in someone whose triglycerides and blood sugar are already improving, a flat or slightly lower HDL is often the metabolism still catching up rather than a setback. The number that tells us the most is the pattern around it: the triglyceride-to-HDL ratio and ApoB. Keep training, keep the muscle building, and give HDL the extra months it needs."

Actionable Steps in Philly

What to do about a stuck or falling HDL.

Check your triglyceride-to-HDL ratio, not HDL alone. A falling ratio means your metabolism is improving even if HDL has not moved yet.
Add resistance training if it is not already in the plan. Building muscle is the most reliable lever for HDL, and it works through the same insulin-sensitivity pathway that is already improving your other numbers.
Give it 6 to 12 months. HDL responds slower than triglycerides or blood sugar. Checking too soon can look like failure when the trend is on track.
Ask for ApoB and Lp(a) if you have not had them. These carry more of the risk information than HDL by itself.
Don't start drinking alcohol to raise HDL. The small lipid benefit does not outweigh the broader health cost.

Tell Dr. Ash what's going on

Key Takeaways

HDL is usually the last lipid marker to improve. Triglycerides and blood sugar move first; HDL can lag by many months or drift lower before it turns around.
A falling HDL during genuine metabolic improvement is not automatically bad news. It often reflects an active CETP exchange between HDL and triglyceride-rich particles that is still settling.
Resistance and aerobic exercise, meaningful weight loss, and quitting smoking are the reliable levers. Diet changes alone move HDL far less than they move LDL or triglycerides.
Raising HDL directly with medication has not been shown to lower heart attack risk, so the HDL number is read alongside the triglyceride-to-HDL ratio, ApoB, and Lp(a) rather than treated as a target on its own.
The pattern matters more than the single number. A falling triglyceride-to-HDL ratio and improving ApoB mean the plan is working, even in a season where HDL itself has not caught up.

Scientific References

Boden WE, et al. "Niacin in Patients with Low HDL Cholesterol Levels Receiving Intensive Statin Therapy" (AIM-HIGH). New England Journal of Medicine. 2011;365(24):2255-2267.
HPS2-THRIVE Collaborative Group. "Effects of Extended-Release Niacin with Laropiprant in High-Risk Patients." New England Journal of Medicine. 2014;371(3):203-212.
Voight BF, et al. "Plasma HDL Cholesterol and Risk of Myocardial Infarction: A Mendelian Randomisation Study." The Lancet. 2012;380(9841):572-580.
Kodama S, et al. "Effect of Aerobic Exercise Training on Serum Levels of High-Density Lipoprotein Cholesterol: A Meta-analysis." Archives of Internal Medicine. 2007;167(10):999-1008.
Rader DJ, Hovingh GK. "HDL and Cardiovascular Disease." The Lancet. 2014;384(9943):618-625.

ApoB and Heart Health - the particle count that carries the strongest evidence for cardiovascular risk
Lp(a): The 'Widowmaker' Genetic Risk - the largely inherited particle worth testing once
Understanding Insulin Resistance - the metabolic root behind the whole atherogenic lipid pattern
Borderline Cholesterol: When ApoB, Lp(a), and BP Change the Plan - a patient case on reading numbers beyond the standard panel
Nervous About Statins? - the honest read on statin benefit, side effects, and who needs one

Fishtown Medicine | Cardiovascular risk

2418 E York St, Philadelphia, PA 19125(267) 360-7927 hello@fishtownmedicine.comHSA/FSA Eligible

Start Your Intake

Frequently Asked Questions

Common Questions

Your HDL can go down during a diet improvement because HDL and LDL often move together when saturated fat comes down, and because an active CETP exchange between HDL and triglyceride-rich particles is common while insulin resistance is still resolving. This is usually temporary. Watch your triglyceride-to-HDL ratio over the same period; if it is falling, your metabolic picture is moving in the right direction even while HDL lags.

Low HDL by itself carries less independent risk than the old thinking assumed. Large genetic studies have found that people with naturally higher HDL are not protected from heart attacks the way older observational data suggested, which points to HDL being a marker of an unhealthy metabolic pattern (high triglycerides, insulin resistance, small dense LDL) rather than a direct cause. The surrounding pattern, read through ApoB and the triglyceride-to-HDL ratio, carries more of the risk information.

Deep-Dive Questions

CETP (cholesterol ester transfer protein) is an enzyme that trades cholesterol out of HDL particles for triglycerides carried in VLDL particles. When triglycerides are high, that exchange runs faster, which shrinks HDL particles and speeds up how quickly they are cleared from circulation. As triglycerides come down with treatment, the CETP exchange slows, but the HDL number often takes months to reflect that change, which is why it lags behind the rest of the panel.

Niacin and early CETP-inhibitor drugs both raised the HDL number substantially in trials, and neither reduced heart attacks or strokes when added to good statin therapy; HPS2-THRIVE's niacin arm caused more side effects without benefit, and several CETP-inhibitor trials were stopped for lack of effect or safety concerns. The most likely explanation is that these drugs raised the amount of cholesterol carried by HDL without restoring the particle's protective function, which suggests HDL's benefit comes from what the particle does (moving cholesterol out of artery walls, reducing inflammation) rather than from the raw cholesterol mass measured on a standard panel.

Mendelian randomization is a research method that uses genetic variants known to raise or lower a marker as a natural experiment, since genes are assigned randomly at conception and are not affected by the disease itself. Voight and colleagues used this method for HDL and found that people with gene variants that raise HDL cholesterol throughout life were not protected from heart attacks, unlike a similar analysis for LDL, where genetically lower LDL clearly lowered risk. This is strong evidence that HDL cholesterol is a marker of cardiovascular risk rather than a direct cause of it.

A standard HDL cholesterol test measures the total cholesterol mass carried by all your HDL particles together, while advanced testing (NMR lipoprotein analysis) can count the number of HDL particles and their size distribution. Two people can have the same HDL cholesterol number with very different particle counts and sizes, and a higher number of small HDL particles is generally viewed as more favorable than fewer, larger ones. This distinction is part of why the single HDL number on a standard panel is an imperfect read of what is happening.

Insulin resistance connects these 3 findings, sometimes called the atherogenic lipid triad, through the same underlying process. A resistant liver overproduces triglyceride-rich VLDL particles, CETP then trades that triglyceride into HDL and LDL particles in exchange for cholesterol, and a separate enzyme (hepatic lipase) strips the triglyceride back out, leaving smaller, denser LDL particles and smaller, more rapidly cleared HDL particles. Treating the underlying insulin resistance is what resolves the whole triad together, which is why we treat the metabolism rather than any one lipid number in isolation. Our guide to insulin resistance covers the fuller mechanism.

Fishtown Medicine still reports and discusses HDL because, while raising it directly with medication has not shown benefit, it remains a useful marker of the underlying metabolic pattern and it feeds directly into the triglyceride-to-HDL ratio, which is one of the fastest, cheapest signals of insulin resistance available from a standard lipid panel. The instruction is not to ignore HDL, but to read it as a piece of the pattern rather than a target to chase on its own.

Ready when you are

Start your intake

Dr. Ash reads every intake himself, and answers questions personally - usually within a few hours.