So today I tackle (again) the diet heart cholesterol hypothesis.
I’ll get into the details of what the hypotheses are shortly, but first let’s consider a key word here: hypothesis. So many people take these hypotheses as fact without realising they are just theories.
In science, one formulates a hypothesis and attempts to disprove it. If it cannot be disproven it is accepted as current best theory or fact.
For a really over-simplified example, I could hypothesise that a plant will only grow in the presence of sunlight. Then my test would be to take a group of plants and divide them into two groups, one group that is exposed to sunlight and one group that remains in the dark. If all plants in the sunlight grow, and all plants in the dark don’t grow, then my hypothesis may be right, but if 1 plant in either group does something only slightly unexpected, the hypothesis is disproven and we must do further investigation to find out why one plant did something unexpected or what flaw there may be in the experiment we did. This is scientific progress.
The diet heart cholesterol hypothesis is a theory that has been conclusively disproven time and time again, yet it is still widely believed. How is this possible? The short answer is politics, business and money, but that’s beyond the point of this post.
Here I attempt to disprove the hypothesis repeatedly, exactly what I shouldn’t be able to do. I don’t have to give an alternative hypothesis (although I may do a follow-up post one day) because the burden of truth is on those who promote the diet heart cholesterol hypothesis.
If I can disprove the diet heart cholesterol hypothesis once, then you should doubt the hypothesis is true, if I can repeatedly disprove the hypothesis you should completely rethink your theory.
As you will see, it’s time to rethink.
What is/are the hypotheses?
Well let’s go to to the British Heart Foundation:
The hypothesis is that high blood cholesterol increases risk of heart disease and cardiovascular disease.
Sounds reasonable enough. So let’s see what evidence there is to disprove this hypothesis. The following evidence is given in an “if X is true, then how do you explain Y” format, because if X was true, then the evidence that is Y would not exist. Let’s begin.
Disproving the Hypothesis that High Cholesterol Increases Heart Disease Risk
If cholesterol increases risk of heart disease why do the Japanese, who have average cholesterol levels of 5.2mmol/l have a heart disease rate of 45 per 100,000 per year and the U.K., who have average cholesterol levels of 5.1mmol/l have a heart disease rate of 145 per 100,000 per year, almost triple that of Japan despite similar (actually lower) cholesterol levels. A similar conundrum can be seen when compared with the US. Here’s the data in graphical format:
If high cholesterol levels increases heart disease risk why, when we look at data from the World Health Organisation and the British Heart Foundation for 86 countries do we so no correlation between mean total cholesterol levels and heart disease mortality rates. If anything it’s a negative correlation and high cholesterol decreases heart disease risk (see data below). Hypothesis disproven.
OK, so that’s some high-level country data. Well how about some smaller studies and data sets?
If high cholesterol increases heart disease risk why did the Sydney Diet Heart Study that set out to indisputably show that replacing saturated fat with unsaturated fat, thereby reducing cholesterol (which it did), results in less heart disease risk actually conclude that the change caused significant increases in heart disease, and all cause mortality? Hypothesis disproven. http://www.bmj.com/content/353/bmj.i1246
If high cholesterol increases heart disease risk why did a study of 997 elderly men and women find that twice as many people with low cholesterol died of heart attacks than those with high cholesterol? Study Link. Hypothesis disproven.
What about if you’ve already had a heart attack, is the hypothesis valid then? No. If having a second heart attack was more likely if you have high cholesterol why did a 10 year study of 120 men who had already had heart attacks show no relationship between survival rate and cholesterol levels? Hypothesis disproven. http://www.ncbi.nlm.nih.gov/m/pubmed/5476780/
If high cholesterol predicts heart disease risk, why did an 11 year study of 630 Maoris find no relationship between cholesterol and heart disease? Why did they find that low cholesterol was associated with higher mortality? Hypothesis disproven. http://www.ncbi.nlm.nih.gov/m/pubmed/7357343/
If high cholesterol increases heart disease risk, why did a study of 4576 Quebec men over 12 years conclude that cholesterol was not associated with heart disease risk. http://www.ncbi.nlm.nih.gov/m/pubmed/2310996/
If high cholesterol increases heart disease risk, why did a study of 92 women aged 60+ over 5 years, conclude that all cause mortality increased with lower cholesterol? http://www.ncbi.nlm.nih.gov/pubmed/2564950
OK, that’s enough. It’s important to remember that these are expensive, time-consuming and arduous studies, that, ideally would confirm the hypothesis. If the hypothesis were true, it would be super useful information that would help millions of people, but the fact is, the hypothesis is wrong.
Let’s ask some other difficult questions for the diet heart cholesterol hypothesis proponents:
- If cholesterol causes heart disease, why does every single cell in our body synthesise its own cholesterol? It seems counter intuitive, surely our bodies are smarter than that.
- If cholesterol causes heart disease, why is 85% of our body’s cholesterol synthesised by the body itself? Our bodies create the majority of cholesterol, rather than absorb it from food. Again, why would it do this if it’s atherogenic.
- If cholesterol causes heart disease by building up in artery walls why doesn’t it build up in veins or pulmonary arteries? Surely it’s easier for cholesterol to collect in the walls of veins and pulmonary arteries where there’s less pressure, but it doesn’t.
- If cholesterol causes heart disease why is it essential for life? Yeah, it’s in virtually every cell, it’s created by virtually every cell, it’s vital for numerous bodily functions such as synthesis of hormones and bile acids.
What about another hypothesis from the British Heart Foundation. Fat/animal fat/saturated fat intake increases heart disease risk? False again. Just look at the Sydney heart study I mentioned earlier, but let’s crack on with some other examples in the same format as before.
Disproving the Hypothesis that Fat / Saturated Fat / Animal Fat Increases Heart Disease Risk
If eating animal/saturated fat increases heart disease risk, explain why Switzerland’s animal fat intake increased by 20%, yet heart disease in males decreased by 22% and in females decreased by 40%? Hypothesis disproven. Link: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1051934/pdf/jepicomh00262-0008.pdf
If eating fat increases heart disease risk, why did a study of 228 men conclude that there was “no difference in mean calorie intake, total fat consumption or other major dietary constituents” between those with coronary disease and those without? Hypothesis disproven. (By the way, they did find that lack of physical activity & smoking increased heart disease risk.) http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1373093/?page=10
If eating fat increases heart disease risk, why did a 2015 analysis conclude that “saturated fats are not associated with all cause mortality, CVD, CHD, ischemic stroke, or type 2 diabetes.” Hypothesis disproven. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4532752/
If fat increases heart disease risk, why would an Food and Agricultural Organisation and World Health Organisation expert consultation conclude that “The available evidence from cohort and randomized controlled trials is unsatisfactory and unreliable to make judgment about and substantiate the effects of dietary fat on risk”? http://www.karger.com/Journal/Issue/250361
If dietary fat increases heart disease risk why did a 2008 meta-analysis of over 340,000 people conclude that “there is no significant evidence for concluding that dietary saturated fat is associated with an increased risk of CHD or CVD”? Hypothesis disproven. http://www.ncbi.nlm.nih.gov/m/pubmed/20071648/
If saturated fat increases heart disease risk why do this observational and non-intervening study of 5400 men comparing the top 15% of fat consumers to the lowest 15% of fat consumers find that, of the 88 people who suffered coronary events, there were 14 in high-fat, 16 in low-fat? Conclusively disproving the hypothesis. http://circ.ahajournals.org/content/28/1/20.full.pdf
… This is tiring… I’ll soldier on…
If eating more fat increases heart disease risk, why did a 1973 Minnesota Coronary Study that involved 9,000 men and women who were split into two groups, one group eating 18% calories from saturated fat and the other eating 9% or less from saturated fat, find that during the study there were 269 deaths in the low-fat group and 206 deaths in high-fat? In the low-fat there 27 cardiovascular events, and in the high-fat there were 26. All in this in spite of lowering cholesterol in a low-fat diet! Hypothesis disproven. http://www.ncbi.nlm.nih.gov/pubmed/2643423
If eating animal fat increases heart disease risk why did the Western Electric study, observing a sample of 3,102 men between 40-55 years old for (at least) 2 years while working at the Western Electric company conclude “the amount of saturated fatty acids in the diet was not significantly associated with the risk of death from CHD”. Hypothesis disproven. http://www.epi.umn.edu/cvdepi/study-synopsis/chicago-western-electric-study/
If saturated fat increases heart disease risk, why did the Framingham heart study show that replacing butter with margarine increases heart disease risk? Also notice more butter decreases heart disease risk:
Finally, if eating saturated fat increases heart disease risk then why does looking at data from the World Health Organisation and the British Heart Foundation for 88 countries show a negative correlation between average saturated fat intake and heart disease rate? Hypothesis disproven.
See Footnote 2
So What Is The Cause?
Like I said, I am not getting into that now, but if you want some pointers and more information around this cholesterol myth check out these sources:
- https://drmalcolmkendrick.org/ – Dr. Malcolm Kendrick has a blog full of great information, including a series on what actually causes heart disease
- The Great Cholesterol Con – Dr. Malcolm Kendrick’s book
- http://www.ravnskov.nu/cholesterol/ – Uffe Ravnskov has a book you can download for free call the Cholesterol Myths
- Great Cholesterol Myth – Jonny Bowden’s book on the topic.
I found this graph here but then struggled to find the actual source data (not a great Google / researcher). As an uber skeptic, I went on a hunt to replicate the graph as best I could. I ended up getting data from these two links:
I used data from 2000 for mean total cholesterol (age-standardized estimate) for men over 25 years old and age-standardized mortality rate by cause (per 100 000 population) cardiovascular diseases. I had to remove some countries that didn’t match up or exist in one data set or the other. In the end I had data for 170 countries and ended up with a similar graph showing no relationship between the two variables.
I found the saturated fat graph here: https://authoritynutrition.com/6-graphs-the-war-on-fat-was-a-mistake/ but was still concerned that I hadn’t seen the source data. As with the above, I went hunting and found the 1998 saturated fat and cardiovascular heart disease statistics at, low and behold, the british heart foundation website here: https://www.bhf.org.uk/publications/statistics/european-cardiovascular-disease-statistics-2008
I used the data from table. 5.5 for saturated fat intake and table 1.4 for mortality data. I averaged the data for men and women to get the data for total mortality for a country. All in all I ended up with this graph:
Then I hypothesised that the countries with higher mortality rates may just have worse health care due to being less well-developed countries. So I then picked the top 20 countries with highest GDP in 1998 from the dataset using this link http://appsso.eurostat.ec.europa.eu/nui/submitViewTableAction.do to find the GDP, which ended up leaving these countries:
Slovenia Portugal Spain Greece Hungary Italy Denmark Sweden Norway Ireland United Kingdom Germany Austria Finland Iceland Netherlands Switzerland France Bulgaria Romania
But the data still highlighted that saturated fat is not the problem. That’s right, I actually doctored the data to try and give the fat hypothesis more of a chance but had no luck: