A network for students interested in evidence-based health care

Cholesterol and Heart Disease – What’s the Evidence?

Posted on July 2, 2018 by Sasha Lawson-Frost

Evidence Reviews
heart with blood vessels visualising cholesterol

The idea that high cholesterol causes heart disease is so widespread you’d have thought it was pretty much an established fact by now. High cholesterol is listed as one of the five main risk factors for heart disease by the British Heart Foundation [1], and the NHS public guidelines on cholesterol claim that “your risk of developing coronary heart disease rises as your blood’s cholesterol level increases” [2]. In supermarkets, you’ll find a significant array of spreads, yoghurts and cereals which claim to improve your cholesterol levels and decrease your risk of heart disease.

But what actually is the evidence that having high cholesterol, or high LDL (low density lipoprotein) levels, increases your chance getting heart disease? You may be surprised to learn that the studies available to us do not all point towards a causal connection between high cholesterol and heart disease. Different studies point towards a number of conflicting conclusions, and ultimately the picture is a lot more complicated than most media on the topic would suggest.

The association between cholesterol and heart disease

At first glance, it seems that we have substantial reasons to believe that high cholesterol and heart disease are closely linked. For one thing, we have a plausible-sounding explanation as to why they should be. This explanation posits that LDLs are the “bad” type of cholesterol, which can stick to the inner walls of your blood vessels and potentially clog them, leading to heart conditions such as stroke, atherosclerosis, angina and coronary heart disease (CHD). Whereas HDLs (high density lipoproteins), are the “good” type of cholesterol, and these have the opposite effect. [3]

There are some reasons to suggest that this hypothesis is correct. Firstly, there is evidence of a correlation between high cholesterol/LDL levels and heart disease. The Framingham Study, a prospective cohort study beginning with over 5,000 participants, published a paper in 1977 which concluded that higher HDL levels and lower LDL levels were associated with a reduced risk of coronary heart disease. [4]

In addition to this, some medical treatments which reduce cholesterol/LDL levels also reduce your risk of heart disease. Statins, for example are effective in both lowering cholesterol levels, and reducing the rates of CHD, cardiovascular disease (CVD) and stroke. [5]

Conflicting evidence

This evidence seems to point towards there being a causal connection between cholesterol in heart disease. However, a closer examination of the issue may make us sceptical of this conclusion.

Firstly, the evidence about the association between high LDL levels and heart disease is more complicated than it first appears. A follow-up from the initial Framingham study suggested that there was only an increase in mortality, by heart disease or other causes, in people with higher cholesterol levels under the age of 50 [6]. This is significant if we want to offer medication or lifestyle advice about preventing heart disease to those over 50 years old. A recent systematic review even suggests that patients over the age of 60 actually lived longer if they had higher LDL levels, directly contradicting the hypothesis that you are more likely to die of heart disease the higher your cholesterol levels are. [7]

Moreover, despite statins showing a correlation between lowered cholesterol levels and a reduced risk of heart disease, some other medications do not show this connection. Niacin, for example, is a cholesterol lowering drug which is known to decrease LDL levels, but has been shown to cause no significant reduction in risk of heart attack, stroke or mortality by heart disease [8] . Even more worrying is the case of Torcetrapib. This experimental drug was designed to reduce rates of heart disease by lowering patients’ LDL levels, but the research was discontinued early because it was linked to an increased risk of mortality and heart disease [9].

A causal connection?

You might be wondering how we ended up at this point, with significant amounts of evidence about cholesterol and heart disease pointing in completely opposite directions. There’s no straightforward answer to why the evidence about this problem is so messy, but one thing that I think should give us pause for thought is exactly what we mean when we talk about ‘causes’ in medicine.

In the case of risk factors like cholesterol, a ‘cause’ does not mean that the presence of this risk factor will always and invariably lead to a certain pathology. This depends on a number of other factors, like age, co-morbidities, lifestyle, and possibly plain luck. What we are actually trying to establish here isn’t nearly as straightforward as claims like ‘the TB bacterium causes tuberculosis’. Instead we are looking at a complex array of inter-related mechanisms, all of which may or may not lead to someone becoming ill. And when we are unclear about exactly how these mechanisms work or how they relate to each other, making predictions about health outcomes will inevitably be difficult and confounded.

For a more in-depth look at evidence about heart disease and cholesterol, I recommend this video by Dr Daniel Aronov: The Truth About Cholesterol – Does It Cause Heart Disease?



Sasha Lawson-Frost

Sasha Lawson-Frost

I’m a Philosophy Student at UCL. I’m currently doing some research the philosophy of Evidence-Based Healthcare (EBH), particularly looking at how EBH deals with issues relating to patient well-being. Twitter: @sashalfrost View more posts from Sasha

Leave a Reply

Your email address will not be published. Required fields are marked *

5 Comments on Cholesterol and Heart Disease – What’s the Evidence?

  • Hedley Finger

    I have been reading lately that atherosclerosis is a disease of blood vessel inflammation, and that cholesterol is just an innocent bystander. Stations are therefore effective because they are anti-inflammatory, not because they also lower cholesterol.

    July 6, 2018 at 10:28 pm
    Reply to Hedley
  • Rob Seddon-Smith

    There is much to be said for your research.
    The philosophy behind EBH and behind the populist views is very interesting.
    I would disagree most strongly with your implications in this piece however – I suspect that you are making the same mistake that others have made in linking some unlinked data in ways it was not intended – this is how ‘what the fat’ succeeded – it’s actually a bit of a leap of faith.
    In particular, you need to be aware that medications that lower cholesterol do have evidence of benefit, it’s just that doing it with drugs other than statins has a low level of benefit, but not none. There is actually evidence that lowering TC makes a difference. As you say, statins are very powerful reducers of cholesterol and massively successful in preventing CHD (WOSCOPS trial is worth a look)
    As for the idea that high LDL in later life is associated with health this is not the case – a mistake here on the part of the analysts. Low LDL levels are asssociated with ill-health in older age but are a symptom not a cause (otherwise we’d see millions of early deaths from statin use)
    There is also a well studied plausible biological process by which cholesterol might cause heart disease, and endless studies (not just Framingham) associating higher levels with higher risk.
    Whilst it is true that there are concerns over the causative relationship, especially with dietary cholesterol specifically, there is still solid evidence that people who eat foods rich in cholesterol and saturated fat get more heart disease than those who do not.
    With respect, I think you have been guilty of making the same error of judgement that many others have done in seeking your evidence – you went with an idea and found affirmation, not information. If it turns out that the ‘cholesterol is innocent’ theory is correct, it will be because of luck, not judgement – there is no good evidence yet to overturn the lipid hypothesis of CHD. After all, it has the best credentials – it works, and saves a lot of lives.
    Now, in the context of your interest in philoshophy this seeking for affirmation is worthy of study. You may need to consider some psychology input too.
    I would argue that EBH suffers desperately from both affirmation bias and is also harmed by persuasive writers looking to make a name for themselves, or the well-intended who lack the knowledge to see the bigger picture.
    This affects everyone, including those who perform key reviews. Even Cochrane suffers (badly) from this bias.

    Good luck.

    July 23, 2018 at 7:39 am
    Reply to Rob
  • George Henderson

    There are questions that can be asked to clarify this matter.
    1) does higher LDL, when it is associated with CHD, also correlate with higher insulin or a high TG/HDL ratio or atherogenic index of plasma? (Yes, in every case). If so, which of these factors should we call causal – the LDL, the hyperinsulinaemia, or the impaired reverse cholesterol transport? Is there any way that LDL itself can cause the other factors (in Framingham, higher LDL is associated with a lower risk of type 2 diabetes, a disease of hyperinsulinaemia)?
    2) Is there any clinical dyslipidemia pattern, associated with CHD and thought to be causal, which features very low triglycerides? If LDL is not associated with CHD when lipoproteins are low in TGs (i.e. when there is little fat in the blood because the rate of fat use for fuel is high) what is causal?
    3) if fasting elevates LDL in healthy non-obese subjects (but not in obese or diabetic) will fasting increase the risk of CHD? Can CHD be a disease caused by eating too little?

    August 26, 2018 at 9:44 pm
    Reply to George
  • George Henderson

    “there is still solid evidence that people who eat foods rich in cholesterol and saturated fat get more heart disease than those who do not.”
    Not in the meta-analysis of Skeaff and Miller, which covers a period when total fat included trans-fats (so the correlation can be expected to be more beneficial today) and surely not if we control better for risk-taking behaviours (when healthy eating advice is to avoid fat, eating fat will be associated with all sorts of behaviours, including occupational and environmental exposure to atherogenic pollutants, that could add to CHD risk)
    Also, if sugar is associated with CHD risk, where does that fit with the idea that LDL is causal? It is generally believed that CHD rates increased in the US during the 20th century, peaking in the 1960s, and there is little evidence of CHD in medicine prior to the 19th century, but people ate animal fat before that – indeed, they hardly had access to vegetable fats prior to the 20th century – and animal fat intake did not increase.

    August 26, 2018 at 10:00 pm
    Reply to George
  • Ahmet Rasim Küçükusta

    Atherosclerosis is a chronic inflammation and main causes are sugar (too much fructose and refined carbonhydtaretes), refined grains, excess omega 6 fatty acids and suboptimal omega 3 faty acids (vegetable oil) and tarns fats (margarins) not fat!

    August 27, 2018 at 9:54 pm
    Reply to Ahmet

Subscribe to our newsletter

You will receive our monthly newsletter and free access to Trip Premium.