Both sides of the cholesterol debate overcook their case. Here's the evidence graded question by question, and why the right answer is always personal.
I'm in the "health over medicine" camp. My instinct is lifestyle first and drugs last, and I'm sceptical of an industry that profits when thresholds move down. You should know that before reading this, because it's a bias, and you can hold me to it.
But bias has to answer to evidence, whichever way it points. When I went through the LDL debate properly, I found strong evidence that supports parts of my instinct and strong evidence that doesn't. Both camps, the "LDL causes heart disease, unequivocally" camp and the "cholesterol is a myth" camp, make claims that don't survive checking. This post takes six questions in turn. For each one it gives the best case from each side, what the evidence supports, and how strong that evidence is.
The grades: Strong means consistent across several independent kinds of evidence. Moderate means good evidence with real gaps. Weak means limited or conflicting. Mechanistic means shown in cells or animals, not yet measured as an outcome in people. Unresolved means we don't know yet.
The case that it causes disease: LDL particles enter the artery wall, get trapped, and start the inflammatory process that builds plaque. The more particles in the blood, the more get in.
The case against: Healthy artery lining is sealed by tight junctions between cells. There's no way for LDL to cross it, so damage to the lining must come first, and that damage is the real cause.
What the evidence shows: Cells lining the arteries have been shown to actively carry LDL across themselves, using specific receptors called SR-B1 and ALK1. This process, transcytosis, doesn't need damage first. Most of this work is in mice and cells, with supporting findings in human arteries. Damage and inflammation increase entry, so both things are true: LDL can get in, and the state of the artery lining matters.
The case for: People born with high LDL get more heart disease, and people born with low LDL get less, whatever their lifestyle.
The case against: Familial hypercholesterolaemia (FH) may cause heart disease through other routes, such as clotting, and the high LDL is just a marker.
What the evidence shows: High LDL caused by several unrelated genes (faults in the LDL receptor, in apoB, or in PCSK9) all leads to more heart disease, and genes that lower LDL protect against it. In one US study, Black participants with PCSK9 variants that lowered LDL cholesterol by 28% from birth had 88% fewer coronary events over 15 years. That group was small, so the uncertainty around the 88% is wide. But the finding was repeated in white participants, where a variant lowering LDL by 15% was linked to 47% fewer events. Different genes, different proteins, the same outcome. Within FH, clotting factors and lipoprotein(a) clearly change individual risk, which fits LDL being one cause among several rather than refuting it.
The case for LDL: The more a drug lowers LDL, the more it reduces events.
The case against: Statins have anti-inflammatory and other effects, so the benefit may come from those, with the LDL lowering incidental. And some drugs have lowered LDL without any benefit.
What the evidence shows: Ezetimibe, PCSK9 inhibitors and bempedoic acid lower LDL by entirely different mechanisms from statins, and all three reduce cardiovascular events roughly in line with how much they lower it. That's hard to explain if LDL doesn't matter. But the picture isn't perfectly clean: bempedoic acid also lowered CRP, an inflammation marker, by about 22%, so even a non-statin drug doesn't fully separate LDL lowering from lower inflammation. The results for CETP inhibitors are mixed (one failed, one worked modestly), and the pattern fits benefit tracking the number of apoB particles rather than the amount of cholesterol they carry. Count matters more than size, too: small, dense LDL particles are linked to higher risk, but large ones are not protective.
The case for: Each 1 mmol/L reduction in LDL cholesterol cuts major cardiovascular events by about 22%.
The case against: That's a relative figure. In absolute terms the benefit is small, and relative figures are used to make drugs look impressive.
What the evidence shows: Both are true. Which matters depends on your starting risk. In someone who has already had a heart attack or stroke, or who has FH, the baseline risk is high, so a 22% relative reduction prevents a lot of events. In a healthy person at low risk, the same relative reduction might mean one or two fewer events per hundred people over five years.
A recent trial shows both sides at once. In CLEAR Outcomes, nearly 14,000 high-risk people who couldn't tolerate statins took bempedoic acid or placebo. Events fell from 13.3% to 11.7%. That's a real benefit, and it means about 63 people had to take the drug for around three and a half years for one to avoid an event. There was no difference in deaths from any cause, and gout and gallstones were more common.
The case for: The link between cholesterol and heart disease holds at every age.
The case against: Average LDL falls in older age groups while heart attacks rise, so how can LDL be the cause?
What the evidence shows: Comparing age groups like that is misleading. Age is the strongest risk factor of all, illness and frailty lower cholesterol, and people with very high LDL are less likely to reach the oldest groups. Within each age band, the link between cholesterol and coronary deaths is still present, though it weakens in older people. Trials suggest statins still reduce events in older people who already have heart disease. The evidence for starting them in healthy people over 75 is thinner.
The case for concern: LDL is LDL, whatever raised it.
The case against: Some lean, metabolically healthy people eating very low carb develop very high LDL alongside low triglycerides, high HDL and low inflammation. That may be a different situation.
What the evidence shows: There are case reports of people with very high LDL for years and no plaque on CT scans, and a small study whose interpretation is disputed. That's not enough to say it's safe, and not enough to say it's dangerous. One exception doesn't disprove a cause, just as not every smoker gets lung cancer. Nobody yet knows whether the metabolic context changes the risk.
The evidence tells us what happens on average. People aren't averages, and what a statin does in someone's life depends on how they think about it as much as on what's in the tablet.
The person a statin makes more careful. Some people start a statin and treat it as a wake-up call. They understand the benefit is modest, change their diet, start exercising, and see the tablet as one part of a plan. Side effects are real but less common than many fear. An analysis of the major placebo-controlled trials found statins cause muscle symptoms in roughly 1 in 100 people in the first year, mostly mild, with no excess after that. Put the other way round, only about 1 in 15 of the muscle complaints reported in that first year was actually caused by the statin. (A fair point from the sceptics: many of those trials weeded out people who didn't tolerate the drug early on.) Expectation also plays a part. In the SAMSON trial, 60 people who had already stopped statins because of side effects took statins, placebos and nothing in rotating months, without knowing which was which. About 90% of the symptoms they reported on statins also appeared on placebo, and half went back on statins afterwards. The symptoms were real. Much of the cause was taking a tablet at all. The small rise in diabetes risk is real too: about one extra case for every 255 people treated for four years, more in older people, and the researchers say it deserves most weight in people whose heart risk is low. Statins also lower CoQ10. That's worth measuring and raising with whoever prescribes, not a reason on its own to stop.
The person a statin makes careless. Others treat the prescription as a licence. US national survey data from 1999 to 2010 found statin users' calorie and fat intake rose over the decade while non-users' didn't, and their BMI rose faster. These were snapshots of the population, not the same people followed over time, so they show a pattern, not proof that statins made people careless. But a tablet isn't a force field, and a modest benefit is easily cancelled out by a worse diet.
The person who refuses when they shouldn't. Then there's someone like me, whose distrust of drugs becomes a blanket rule. After heavy negative media coverage of statins in 2013 and 2014, UK researchers estimated that more than 200,000 people may have stopped their statins in the following months, if the coverage was the cause. In Denmark, people who stopped statins early after negative news stories had 26% more heart attacks and 18% more cardiovascular deaths than those who carried on. Those are associations, not proof. But some of the people who stopped had FH or had already had a heart attack, and for them lifestyle alone wasn't enough. The right approach was both: lifestyle plus medication, with proper assessment and regular monitoring.
The person for whom lifestyle genuinely is enough. And there are plenty of people at low absolute risk, where the honest numbers show a very small benefit. For them, lifestyle first with monitoring is a reasonable, evidence-based choice, not a reckless one.
None of these people is right or wrong about statins in general. Each of them is right or wrong about their own situation.
Nonsense is nonsense and corruption is corruption, whichever camp it comes from. The answer for you depends on your numbers, your history and what you'll actually do with the information.
Stop navigating your health by population guidelines designed for the average patient. The TDG Five-Test Programme looks across hormones, gut function, organic acids, food sensitivity and blood chemistry — and a clinical picture designed around you.
See the TDG Programme →Read next: Beyond “Your Cholesterol”: The Heart Markers That Matter: ApoB, Lp(a), blood pressure and the rest, in both units, and what moves each one.