Huge study finds gaping holes in the cholesterol hypothesis

It’s ‘Cholesterol Week’ here in the UK. It’s been accompanied inevitably with a rash of stories about the perils of cholesterol and how we should be more aware of our cholesterol levels and do something about them if they’re raised. This will all no doubt help to swell the coffers of the drug and food companies that sell ‘solutions’ to cholesterol, but let’s spend a moment analysing the basic premises on which the cholesterol hypothesis is based:

  1. cholesterol causes heart disease
  2. reducing cholesterol levels therefore reduces the risk of heart disease

Actually, studies show that taking dietary steps to reduce cholesterol does not bring broad benefits for health or save lives. We also have drugs that reduce cholesterol that also do not appear to improve health. What this means is that the second assumption above is on distinctly shaky ground. But what about the first?

The idea that cholesterol is a risk factor for heart disease is repeated loudly and often. However, there is at least some evidence that this ‘fact’ does not stand up to scrutiny.

By way of example, let’s take a look at the result of a recent study published in the journal Atherosclerosis [1]. In it, more that 82,000 adults in England, UK were followed for an average of over 8 years. The relationship between a range of lifestyle factors and health markers and risk of stroke and heart disease was assessed.

Some of the factors that were associated with an increased risk of both stroke and heart disease were smoking, raised blood pressure, diabetes and low levels of physical activity. What about cholesterol though? For all the talk about the perilous dangers of this substance you’d expect it to be associated with an increased risk of cardiovascular disease, right? Wrong.

  • Higher total cholesterol levels were NOT associated with an increased risk of death due to heart disease.
  • What is more, higher total cholesterol levels were actually associated with a REDUCED risk of death due to stroke.

Here’s what the authors of this study made of these results:

“Consistent with our findings, the existing evidence has generally shown that cholesterol is a stronger risk factor for coronary events than for stroke, whereas high blood pressure is a better predictor of stroke. We demonstrated an inverse association between total cholesterol and stroke risk, which is consistent with prior work particularly in relation to fatal hemorrhagic stroke.”

What the authors appear not to be able to say up front is that their mammoth study found that higher cholesterol levels are not a risk factor for heart disease, though they do admit to the fact that higher cholesterol levels are associated with lower stroke risk. But then they add this: ““Nevertheless lipid-lowering therapy with statins has shown to reduce the incidence of both thrombotic stroke and coronary disease.”

What they appear to be suggesting here is that we should gloss over their findings, because cholesterol-reduction with statins reduces the risk of stroke and heart disease, so cholesterol must cause these conditions after all.

But the thing about statins is that they don’t just reduce cholesterol, but have range of effects that might reduce risk of cardiovascular disease, including anti-inflammatory and blood-thinning properties. Here’s what I believe to be a more logical and honest appraisal of their studies findings.

1. higher cholesterol levels were not associated with heart disease and were associated with a reduced risk of stroke

2. statins reduce the risk of stroke and heart disease

3. any benefits of statins in this regard cannot be due to their cholesterol-reducing properties

Don’t be expecting much in the way of this sort of honest appraisal from the scientific establishment any time soon. For whatever reason, some seem keen to twist and turn in an effort to keep the cholesterol hypothesis alive.


1. Hamer M, et al. Comparison of risk factors for fatal stroke and ischemic heart disease: A prospective follow up of the health survey for England. Atherosclerosis epub 22 August 2011.

15 Responses to Huge study finds gaping holes in the cholesterol hypothesis

  1. TerryJ 21 September 2011 at 10:30 pm #

    This phrase doesn’t seem right –

    ‘higher cholesterol levels are associated with lower cholesterol risk’

  2. John Briffa 21 September 2011 at 10:57 pm #


    You’re right – that should be ‘lower stroke risk’ – correct now. Thanks.

  3. Cordier 22 September 2011 at 10:30 am #

    Oh I wish that all the general practitoners read your intelligent article and your deductive reasoning,because when they know your cholesterol level,even if it is a little bit high,they declare:”My goodness,you need statins”!
    Regarding the thrombotic stroke,there is a compound,alpha linolenic acid,(very well studied by the French professor Serge Renaud),which lowers the fibrinogen level,signficantly!We can find it in canola,walnut,perilla,and flaxseed oils.

  4. Bev C 22 September 2011 at 3:44 pm #

    I notice the distinction between thrombotic and hemorrhagic stroke. Did the study address the risk of thrombotic stroke due to high cholesterol? And if so, what was the relation?

    Love your blog but love the podcast even more. And would love a full hour each week!

  5. Brian 23 September 2011 at 10:35 am #

    Using the (epidemiological) logic the cholesterol is highly associated with heart attacks and is therefore the cause of it, you MUST come to the same conclusion about the 3 emergency services being the CAUSE of road traffic accidents. After all the are HIGHLY associated with them for the same reason as cholesterol and heart attacks.

  6. Donald G 23 September 2011 at 8:24 pm #

    You write: “statins reduce the risk of stroke and heart disease.” I thought that statins only reduce this risk in males under 65 who have already had serious cardiac illness. Is this so?

  7. simona 29 September 2011 at 12:15 am #

    I suppose they’re not talking about Total cholesterol levels of 11 as not associated with heart disease. When is high cholesterol not too high?

  8. Actispan 29 September 2011 at 2:46 am #

    John, thanks for your thought provoking analysis. You didn’t mention the study finding that increased HDL levels were protective against IHD (e.g. heart attack). Of course, statins lower LDL-c and raise HDL-c, and the ratio of LDL:HDL is also considered an important part of the “cholesterol hypothesis”.

    Yes, the drug makers are selling something. So are the supplement makers, the book writers, the diagnostic companies, and everyone else. Let it be a meritocracy, based upon truth… and let’s all keep searching for it.

  9. Wille 29 September 2011 at 11:06 pm #

    Actispan..and also
    Saturated fat raises the good “carry away ” HDL. Carbohydrates raises the small dense LDL that get stuck underneath the endotelian cells and dietary fat raises the non-harmful boilant LDL, that float around not aware of this enourmous human problem.

    The solution is if you ask me, less starch and sugar, instead of more statins.

  10. TP 2 October 2011 at 4:38 pm #

    An interesting post, particularly with the advent of Denmark’s “fat tax” on saturated fat containing products. This strikes me as monumentally unfair, and I can’t help but think that margarine companies etc might have had some sort of influence…

  11. Christopher Palmer 8 October 2011 at 6:14 pm #

    While this is a bit long, I hope it gets to a few points and clarifies them.

    ”You didn’t mention the study finding that increased HDL levels were protective against IHD (e.g. heart attack). Of course, statins lower LDL-c and raise HDL-c, and the ratio of LDL:HDL is also considered an important part of the “cholesterol hypothesis”.” [#8]

    A very important point: the ratio of one lipoprotein to another can imply something helpful.

    Something, I feel, that gives the lie to the imprecise virtue in the cholesterol hypothesis is the imprecise way in which the boffins, and the media, have used terminology over the life of the hypothesis. The blurring of the distinction between cholesterol and lipoproteins is a case in point.

    Lipoprotein is not cholesterol; and cholesterol is not a lipoprotein. Lipoproteins association with cholesterol is that lipoproteins carry cholesterol and other fat soluble substances around the blood. Cholesterol cannot disperse in blood because it is an aqua-phobe. However referring to one as the other is a bit like confusing the occupants of a vehicle with the vehicle itself. The size, shape and formula of cholesterol is invariable. One cholesterol molecule is a precise replica of the next; whereas the lipoproteins that can transport cholesterol (and other things) can come in different sizes, or densities. Lipoproteins stand comparison with variably sized vehicles while cholesterol stands comparison with a class of passengers who are each precise clones of the others.

    Cholesterol is found in plaques that form in arteries; and cholesterol itself could have either an active role in the formation of those plaques or a more passive one. Proponents of the cholesterol hypothesis have tended to assume cholesterol has an active role in the formation of plaques without being able to explain the biochemical steps involved in the transition and exchange.

    The LDL:HDL ratio thingamabob reveals the irrationality of this ad-hoc aspect to an unconvincing hypothesis. How? Because the cholesterol in any lipoprotein is invariable. The density, and/or some other aspect, associated with the lipoprotein is the variable. How can identical cholesterol in either LDL or HDL have different effects without that some other aspect, agent, or feature, has a part to play? It cannot. So it must seem that lipoproteins themselves may possibly come in variants (or that the variants might carry agents or features) that may deserve tagging with labels that are either ‘good’ or ‘bad’, but cholesterol ain’t it (!) – because it’s the same cholesterol riding in the variant density lipoproteins. Hence …

    … The ratio of LDL to HDL (as is suggested may associate with advance of heart disease) may be informing us of something, but what it is informing has zip to do with cholesterol. Why then do the boffins blur the distinction between cholesterol and lipoproteins by referring to low-density liporotein as ‘LDL cholesterol’, or LDL-c, as in the quote? Because if they were clear and consistent in the terminology it would poke a stick in the direction of the spokes of the wheels that carry the cholesterol bandwagon, that’s why. How?

    … Some of us take train, some of us like a bus, and some of us drive our own car. While cholesterol itself doesn’t seem so fussy some fat soluble substances may have a preference for riding with different density lipoproteins. If I read Dr Kilmer McCully [1] right then homocysteine prefers to ride in LDL. Thus …

    … Homocysteine is an agent to be found in LDL and not in HDL. Are you now thinking what I think you’re thinking? Cholesterol may have a passive role in the formation of plaques while homocysteine, aggravated by deficiencies in B complex vitamins, may have a more active role. That is precisely what Dr McCully spent a career conscientiously establishing. It is a momentous advance in medical science. How was it received?

    Dr McCully published his findings. He would have continued researching and refining them in a post he had held for years. After the passage of time he was made unwelcome in the post he loved. In effect, he was hounded out of town and spent time in the wilderness of unemployment; for threatening to poke an even bigger stick at the wheels of the cholesterol bandwagon, I suppose.

    “The solution is if you ask me, less starch and sugar, instead of more statins.”

    How very true. The solution according to Dr Kilmer McCully is a reduction in consumption of vitamin and mineral depleted refined starchy carbohydrates and substitution with inclusion of sources of B complex vitamins. The rationale; B6, folate, and B12 can act to sweep away surplus homocysteine. I think, where saturates and monosaturates are concerned we can helpfully dispense with fat-phobia. During the course of evolution animal fats were an explicably expedient factor resulting in the physiological evolution of our kind.

    The cholesterol bandwagon and the statin bandwagon is the biggest single crime ever perpetrated against humanity in peacetime. ‘Cholesterol week’ is a promotional splash backed by the cholesterol ‘charity’, ‘Heart UK’.

    The origins of the cholesterol bandwagon began with an innocent, understandable, and forgivable mistake just over 100 years ago when the effects of oxy-cholesterols were confused with, and attributed to, cholesterol. Some decades later came the highly selective, perhaps even fraudulent ‘Seven Countries’ study in which pride and personal ambition clouded the objectivity of Dr Ancel Keys. Then came the low-fat paradigm and the industry and profits that go with that, with development of statins adding tank-tracks and armour plating to the bandwagon, and now the lucrative market in sterol-enriched spread or yoghurt-drinks.

    In such adversity and desperation US Army Captain John H Miller (Tom Hanks in ‘Saving Private Ryan’) might advocate use of a ‘sticky-bomb’. It would require the use of an exceptionally ‘sticky-bomb’ to disable the bandwagon(s) now.

    Other than saying cholesterol is a vitally important substance essential to animal life, certainly to human life, is an unreliable surrogate end-point where heart disease is concerned, and that statin pedlars should ‘go away’, what else can be charitably said about cholesterol?”

    Lower ranks of the army have a saying, ‘Bullshit baffles brains”. They’re not wrong, yet few are aware just how right they are.

    1, Dr Kilmer McCully is author of ‘The Homocysteine Revolution’ and ‘The Heart Revolution’.

    (Apologies for posting at length, Dr Briffa, I just hope my contribution is valid and virtuous. I can declare there are no financial interest involved) 🙂

  12. Brok 16 September 2013 at 9:00 am #

    On the 10th August I joined many thousands of others who have experienced a near death event – a heart attack.. Cutting a long and possibly familiar story short, a week’s treatment by a great team in the Coronary Care Unit saw me home with the biggest bag of drug goodies I had ever seen. Among them Atorvastatin. For a month I took this cocktail, a month of abdominal pain that made me feel I had swallowed a cheese grater, muscular aches and dizziness. I also suffer type 2 diabetes. My blood glucose was all over the place, five times readings of between 15 and 21 showed up. My main concern was the pain I was suffering and I decided to drop the statins. Within 24 hours the pain and discomfort had gone. I could face food without being nauseated by the sight of it! The cholesterol killer is, in fact a patient killer, because I suspect I am not the only person who has thought that of statins prolong life at the cost of increased pain and discomfort, I really do prefer to “leap off the twig”. Atoravastatin and Bisoprolol are both known to increase blood/glucose levels.
    I have increased my intake of tomatoes, bananas and spinach to lower blood pressure and have bought myself an exerciser and an ‘aerobics’ DVD. I firmly believe that the medical profession have it wrong as far as cholesterol is concerned. If they haven’t, a pain-free, shorter life is infinitely preferable to statins.


  1. Why a ‘raised’ cholesterol level may be a cause for celebration | Dr Briffa's Blog - A Good Look at Good Health - 6 October 2011

    […] couple of weeks back one of my blog posts highlighted a study from England in the UK which found that total cholesterol levels had no […]

  2. Högre kolesterol – mindre risk för stroke | by Anna Sparre - 8 October 2011

    […] kan du läsa mer: Share this page Tweet about this Email this link Det här inlägget postades i Ohälsa & […]

  3. Does Depression Increase Risk of Having and Dying from Stroke? - 27 February 2012

    […] the underlying mechanisms and elucidate the causal pathways that link depression and stroke.” Experts have recently found out that depression is associated with risk of developing and dying from…involved systematic review and meta-analysis of prospective cohort studies to describe association […]

Leave a Reply