The short version
If you read nothing else
- You don't mostly eat cholesterol — you make it. The liver is the factory, and it builds it because you need it: hormones, bile, vitamin D, every cell membrane.
- LDL and HDL aren't "bad" and "good" — they're directions. LDL carries cholesterol out to the body; HDL brings the surplus back to the liver.
- Insulin turns the factory up. The enzyme statins block is the same one insulin activates — which is why blood sugar belongs in a conversation about cholesterol.
- Particle size matters. Large buoyant particles behave differently from small dense ones, even at the same total number.
- You can estimate which you have from two numbers already on your standard panel: triglycerides and HDL.
- But LDL still counts. The evidence that it causes arterial disease is among the strongest in medicine — and it acts cumulatively, over years.
- Never stop a prescribed statin on your own. Understand it, question it, discuss it — with the doctor who prescribed it.
Why your body wants it
Cholesterol has the reputation of a poison. It is closer to a building material.
Every cell you own has cholesterol in its membrane — it's what keeps the wall flexible instead of brittle. Your body also uses it as the raw material for cortisol, oestrogen, testosterone, for vitamin D, and for the bile acids that let you digest fat at all.
Which is why your liver doesn't wait for you to eat it. It manufactures the bulk of it itself, around the clock, and adjusts production according to how much arrives in food. Eat less, and the factory speeds up. That's the mechanism behind a fact that surprises people: for most of us, dietary cholesterol has a fairly modest effect on blood cholesterol.
So a high number is a question, not a verdict. It might mean the factory is running fast. It might mean the return route is blocked. It might mean your thyroid is slow. Those are different problems with different answers — and one line on a report can't tell them apart.
The delivery round
The clearest way to understand LDL and HDL is to stop thinking of them as good and bad.
This is why "high HDL is good" is roughly true but badly explained. A healthy HDL usually means the collection service is working. It isn't that the HDL itself is protective in some magical way — it's that a functioning return route means less left lying about in the vessel wall.
How it's made — and where statins act
One production line, one bottleneck. Everything interesting happens at the bottleneck.
The practical consequence. If insulin is high — from frequent eating, refined carbohydrate, or insulin resistance — the factory runs faster no matter what is on your plate fat-wise. That's why "I cut out butter and nothing changed" is such a common story.
Why one number misleads
Two people can have identical LDL cholesterol and quite different particles carrying it.
The caveat that keeps this honest. Large buoyant particles are less atherogenic — not harmless. And the overall amount of LDL still matters independently: the European Atherosclerosis Society reviewed over 200 cohort studies, genetic studies and trials covering more than two million people, and concluded LDL is a causal factor, not merely a marker — acting cumulatively, so the longer it stays high, the more it counts. Anyone who tells you particle size makes LDL irrelevant is going further than the evidence does.
Your triglyceride-to-HDL ratio
You almost certainly already have the two numbers this needs. Look at your last lipid panel.
Why this works. When insulin is high, the liver makes more triglyceride-rich particles, and the exchange that follows leaves LDL smaller and denser and HDL lower. So the ratio between those two numbers carries information that LDL cholesterol alone does not. In one commonly cited analysis, around 80% of people above a ratio of 3.8 had predominantly small dense LDL, and around 80% below it had predominantly large buoyant LDL.
It is an estimate, not a test. It doesn't replace measuring particles directly, and it works less well in some populations — it has been found a poor marker of insulin resistance in people of South Asian descent, for instance. Use it as a prompt for a better conversation, not as a diagnosis.
What actually makes particles small
Almost none of it is dietary cholesterol.
Pushes toward pattern B
- Insulin resistance — the single biggest driver
- Refined carbohydrate and frequent eating — every meal is an insulin signal
- An underactive thyroid — slows the clearance of LDL from the blood
- Weight carried around the middle — visceral fat is metabolically noisy
- Trans fats — the one fat with no defenders left
- Chronic stress — cortisol raises glucose, glucose raises insulin
- Smoking — oxidises particles once they're in the wall
Pushes toward pattern A
- Fewer, better-spaced meals — fewer insulin spikes
- Less refined carbohydrate, more fibre
- Regular movement — muscle is where glucose goes
- A thyroid working properly — worth testing if lipids are stubborn
- Omega-3 fats — lower triglycerides, which shifts the ratio
- Sleep — short sleep worsens insulin sensitivity within days
Notice what these have in common. Almost every lever here is a blood-sugar lever, not a fat lever. That is the single most useful reframe in this guide — and it's why someone can cut fat for a year and watch nothing move.
Statins, honestly
Both the enthusiasm and the outrage tend to skip the same detail: who is being treated.
A statin blocks HMG-CoA reductase — the bottleneck in the diagram above — so the liver makes less cholesterol and pulls more out of the blood. That works, reliably, and the effect on LDL is not in dispute by anyone.
The argument is about how much good that does, and for whom. And here the answer genuinely differs depending on who you are.
Here the benefit is large and not seriously contested, including by most thoughtful critics. The baseline risk is high, so a proportional reduction translates into a meaningful number of events prevented.
This is the group for whom stopping without medical advice is genuinely dangerous.
This is where the honest argument lives. The relative reduction is similar, but the baseline risk is much lower — so the absolute benefit over five years is small, and reasonable people weigh it differently against side effects and the prospect of decades of medication.
This is a conversation to have with your doctor, armed with your own absolute risk — not a decision to make from a website.
Look again at the production line: coenzyme Q10 branches off the same mevalonate pathway a statin blocks. Lower cholesterol production, lower CoQ10 production. That much is simple biochemistry.
Whether supplementing CoQ10 relieves statin-associated muscle symptoms is less settled — trials disagree. But the mechanism is real, it's cheap to try, and it's a sensible thing to raise with your doctor rather than quietly stopping the tablet.
The line I won't cross, and nor should any website. Do not stop a prescribed statin because of something you read — here or anywhere. If you have doubts, take them to your prescriber: ask for your absolute risk, ask what the goal is, ask what happens if you stop. Those are excellent questions. Acting on them alone is not.
What you can change yourself
None of this competes with medication. It changes the thing medication is compensating for.
Bring insulin down
Fewer eating occasions, less refined carbohydrate, protein and fibre at each meal. This is the lever with the longest reach.
Move the muscles
Walking, and something that makes legs work. Muscle is where glucose is taken up; it is also where triglycerides get burned.
Keep bile flowing
Cholesterol leaves partly through bile. Fibre binds it so it leaves for good, and dietary fat is what prompts the gallbladder to empty at all.
Check the thyroid
An underactive thyroid raises cholesterol by slowing its clearance. If lipids won't move, this is worth testing before anything else is blamed.
Sleep and stress
Cortisol raises glucose; glucose raises insulin; insulin runs the factory. Not a soft recommendation — a biochemical one.
Re-test properly
Ask for triglycerides and HDL, not just total cholesterol. Without those two you can't see the pattern at all.
When this needs a doctor, not a website
See a doctor, and don't self-manage, if any of these apply:
- Very high cholesterol, especially from a young age, or a family history of early heart attacks — familial hypercholesterolaemia is genetic, common enough to matter, and treatable. Diet will not fix it.
- You've already had a heart attack, stroke or stent.
- Diabetes or diagnosed kidney disease.
- Chest pain, breathlessness on exertion, or pain in the calves when walking.
- You're already on a statin and want to change something. Discuss it — don't act alone.
Two lenses on the same question
This is a topic where honest, qualified people disagree — so here is both sides, fairly.
🏥 The conventional view
- LDL is causal, not incidental — genetics, epidemiology and trials all point the same way, and the effect accumulates over a lifetime.
- Lowering LDL lowers events, whichever drug does the lowering — which is hard to explain if LDL were merely a bystander.
- After a cardiac event, statins prevent a substantial number of further ones.
- Particle size is interesting but secondary; the amount of LDL still matters on its own.
🌿 The sceptical view
- Risk is multifactorial — insulin, inflammation, stress and blood pressure are underweighted next to one number.
- Benefits are usually quoted as relative risk, which flatters them; absolute benefit in low-risk primary prevention is small.
- Guidelines have been written with heavy industry involvement, which is a fair thing to notice.
- Populations exist with high cholesterol and low heart disease, and the model explains them awkwardly.
Where they agree — and it's more than either side admits: insulin resistance is bad news; triglycerides and HDL tell you things LDL alone can't; movement, sleep and food genuinely change the picture; and nobody should stop a prescribed medicine unsupervised.
If you want to read the sceptical case properly, read The Great Cholesterol Con by Dr Malcolm Kendrick — a British GP who argues it at book length and far better than a summary can. I don't agree with all of it, and the strongest genetic evidence on LDL came after it was written. But it is well argued, it is written by a doctor, and its criticism of how risk is communicated is fair. Reading the other side properly is how you form your own view rather than borrowing mine.
Keep reading — it's free
You've got the part that shows you where you stand. The rest is what to actually do about it — and I'll send it straight to your screen.
Still to come in this guide:
- What actually makes particles small
- Statins, honestly
- What you can change yourself
- When this needs a doctor, not a website
- Two lenses on the same question
No charge, ever. Just my guides and the occasional new one. Unsubscribe any time.
Was this useful?
There's a one-page printable to go with this guide — "What to ask your doctor about cholesterol": fill in your own numbers and take it to your appointment. Leave your email and I'll send it over, along with my magnesium guide, plus the occasional new guide as I write them.
Just the guides and the occasional helpful tip. Unsubscribe any time.
Or, if you'd like your own results read this closely: I'm Dr Galina Goksu, a medical doctor and nutrition consultant, working online with clients worldwide. Come and say hello.
Sources
Where the numbers and claims on this page come from.
- Ference BA et al. Low-density lipoproteins cause atherosclerotic cardiovascular disease. 1. Evidence from genetic, epidemiologic, and clinical studies. European Atherosclerosis Society Consensus Panel. Eur Heart J 2017;38(32):2459–2472. — over 200 cohort studies, Mendelian randomisation and trials, >2 million participants and >150,000 events; LDL causal and cumulative.
- Borén J et al. Low-density lipoproteins cause atherosclerotic cardiovascular disease: pathophysiological, genetic, and therapeutic insights. EAS Consensus Panel. Eur Heart J 2020;41(24):2313–2330. — the follow-up statement.
- Triglyceride/HDL-C ratio as a surrogate for LDL particle size and insulin resistance — the ~3.8 threshold, above which roughly 80% show predominantly small dense LDL.
- TG/HDL-C ratio compared with other lipid measures as a marker of insulin resistance (HOMA-IR comparisons) — including the finding that it performs poorly in people of South Asian descent.
- Kendrick M. The Great Cholesterol Con. John Blake, 2007. — the sceptical case, argued at length by a practising GP; cited here as the other lens, not as this guide's evidence base.