Search This Site!
Nutrition Past and Future

19 Cholesterol Confusion 2 The Cause of Heart Disease 

This is an excerpt from cholesterol denier Uffe Ravnskov’s book, The Cholesterol Myths. It is a breathtaking display of ignorance. He is trying to convince the reader that serum cholesterol cannot be a real risk factor for heart disease. He says that if a “risk factor is the cause, its rise and fall must be reflected, with no exceptions, in the rise and fall in the death rate from the disease.” This is nuttiness. As I’ve shown you, cholesterol can and often does fall in populations that consume a lot of animal foods, but this doesn’t help their mortality in old age because it has nothing to do with improving health. But the real deception here is to be found in his use of the definite article, “the.” Look at that sentence in the middle again. “But if the risk factor is the cause…” He intentionally italicized the wrong word. “The” is the word doing the work for him. This is an example of the fallacy known as begging the question.

Begging the question is the fallacy that arises when someone sneaks into their argument a premise they need you to accept as true, even if that premise is quite disputable.

In this case the deception is that Ravnskov wants you to believe that there is a singular and uncomplicated cause of heart disease. He could have said, “but if the risk factor is a cause,” but he didn’t. He decided instead to argue dishonestly and slip in the word “the” hoping that you wouldn’t stop to question whether anyone even claims there is just one cause of heart disease, or one variable that affects risk.

Ancel Keys wouldn’t let himself be boxed into a position he didn’t hold. He said, “Those of us who emphasize the importance of diet have not claimed that dietary fat is the only etiological factor and, conceivably, the influence of diet could be overcome by other influences…” Ancel Keys never claimed that fatty foods or high cholesterol were the only things that mattered. That’s why the term “risk factor” is used instead of “cause.” The word “cause” can lead to some confusion when we talk about heart disease so we usually avoid it. “Risk factor,” or “determinant” work a lot better. High cholesterol is one determinant of your risk for a heart disease event. One might fairly say that high cholesterol is the cause of atherosclerosis, and that would mostly be right, but no one says that because no one measures heart disease outcomes in populations by scans of artery wall thickness. We don't keep population records of plaque thickness. We record heart disease by adverse events, and the immediate cause of those is not high cholesterol in the blood. The clot that kills you wouldn’t have ever happened without the presence of high cholesterol sometime during your life, but the cause of a fatal plaque rupture is a bit more complicated. It’s the result of a process that is affected by other factors, and the cholesterol in the artery wall is the cholesterol that matters at that point, not the cholesterol in the blood.

No responsible person has ever claimed that high cholesterol is the only risk factor. That is not to say that it is not the most important risk factor. Here you see that the CDC is well aware of the various risk factors. The point I want to make here is that noncommunicable diseases don’t necessarily have a single cause the way infectious diseases do. The concept of “cause” is more complicated in heart disease. We are better off if we think like epidemiologists and instead of staying stuck on the belief that a disease must have a cause, we should think in terms of composite cause. All the various risk factors come together in different ways for different people to make up the cause of their personal heart disease.

The prime importance of cholesterol in the causation of atherosclerosis is evident due to the fact that there does not appear to be a way to induce lesion formation without dyslipidemia. Dyslipidemia is therefore the cause of atherosclerosis. There is no other factor that explains lesion development better. There is no more plausible mechanism that has been described. Based on the evidence, this appears to be an incontrovertible fact. High cholesterol loads the gun. So what pulls the trigger? My answer is, if there is no loaded gun, there is no trigger to worry about.

A great example of this hierarchy of risk factors can be found in the Pygmies of Oceania. They practically all smoked, but heart disease was found to be practically non-existent among them. As you will see in a later video, this can be explained by their very low cholesterol levels. It is not explained by the healthfulness of smoking. Would a low carber say that smoking doesn’t cause heart disease? If one did, he’d be saying something less absurd than if he said that high cholesterol doesn’t cause heart disease.

The various risk factors come together to cause more damage than just one might do alone. Here you see a recent study that used cholesterol as one of several risk factors to measure the causes of heart disease at the population level.

And here you can see clearly how those risk factors reinforced one another to drive up risk. The worse your risk factors are, the more likely you are to have an adverse event. This is the whole idea of composite cause.

You may wonder how scientists decided what turns a factor into a risk factor and you may wonder how they figured out how important each risk factor is.

In epidemiology, the best way to think of this is in terms of the Bradford Hill criteria. Bradford Hill was a famous researcher who laid out the basic method for deciding what should be called a cause of a disease in a speech he gave in 1965. I won’t go into his criteria here. I’ll just point out that these have been around awhile so they have been improved and refined over time. Still, these criteria apply very well to heart disease research so they will be useful to you if you want to study how scientists identified the causes of heart disease.

If you want to read his original speech laying out these criteria, this paper is what you should look for. My favorite part comes at the end. It is an eloquent statement about why it is important to act to save lives before we have perfect knowledge. Cholesterol confusionists like Gary Taubes want you to believe we shouldn’t act because all the science isn’t in yet. The fact is all the science will never be in. This is the nature of science as practiced by mere humans. In life we act with incomplete knowledge all the time. We are just humans. We are not omniscient. Every single decision you make in life is made with incomplete knowledge and yet you act anyway, whether you are buying a car or crossing the street. Taubes is being totally unreasonable in his objections to efforts to save lives by educating people about proper nutrition. As you can see from my videos, there is a huge amount of science in support of prevention efforts, and the evidence weighing against those efforts is embarrassingly weak. I think you would have to have a personal or financial stake in the status quo to decide otherwise once you’ve seen the science.

Then there is the question of how the relative importance of risk factors is determined. A cranky blogger mockingly wondered how anyone could know how many lives are saved by cholesterol-lowering behaviors apparently without ever trying to find the answer for himself. I think he doesn’t seek to understand this because of the type of person he is.

I think either you are a person who understands that mathematics is a way to understand the natural world or you aren’t. Mathematical models are devised and tested for their ability to predict natural events in every science. Meteorology uses mathematical models. Economics uses mathematical models. Road design uses mathematical models. Epidemiology is no different. Perhaps the idea that statistics can be used to represent reality is simply too abstract an idea for some. There is nothing wrong with that, but such people shouldn’t pretend to understand anything in science.

All the risk factors now in use have been validated using mathematical models. They all have a stood the test of time.

Here is how this works in a study. When we see that there has been an increase in coronary heart disease in Beijing, we should understand that there are real statistics to support that. There are numbers out there that generate objective disease rates. There also are statistics that quantify the known risk factors. Examples of those are the numbers in your blood pressure reading, or the numbers from your cholesterol test. If we accept the existence of these statistics, we must also accept that they will be analyzed by people using mathematical models. Good numbers don’t go to waste. It doesn’t really matter if a random blogger can’t get his head around that. In this case here in Beijing, sophisticated mathematical models identified changes in serum cholesterol as the risk factor that increased along with deaths in the closest way, simply based on the numbers. There was no trendy Paleo ideology here. There were no opportunistic fat loss books being peddled. There was no attempt to incite a fake controversy. There was simply emotionless computer software performing calculations. We can see that this result makes sense, as the dietary habits of the people in Beijing have changed dramatically. They greatly increased their consumption of eggs and meat. The authors of this study have told us which mathematical models they used.

We can then look up those models and see that they have been validated and refined by other scientists in other populations. The calculations produced by these models turned out to have good predictive power in other situations.

We can also look at the paper about Beijing and see that their model was tested and validated by these authors using their own data. You may choose to disagree with them, but you must accept that you are dealing with an objective analysis based on hard numbers. If you really want to dispute their findings, you might criticize the models they used or the accuracy of the data they used. You would have to argue on some objective basis because the study was performed objectively. Simply calling someone names for reporting results like these is not a scientifically reasoned, or even an adult, response.

We are not stuck forever with the same old risk factors. New ones may come along. They probably won’t replace the existing risk factors, though. This recent paper received a lot of attention. It showed us that women who experience a high degree of psychological stress at work have a much greater likelihood of having a cardiovascular event. They characterized stress with methods I won’t explain here, but they came up with a way to describe jobs as “high strain.” You can see how much worse off the high strain women were. But they were worse off in large part because they exhibited other risk factors like high cholesterol. And that’s what’s interesting for our purposes here. You can see the relationship between the various risk factors and incident cardiovascular disease. High cholesterol was more damaging than any other single category they used except for their depression and anxiety category. High cholesterol explained more events than diabetes, high blood pressure, obesity, smoking, and more. So does high cholesterol cause heart disease? If we accept that something causes heart disease, and if we accept that math helps us to understand the world, then the answer must be yes.

Once we identify and measure the determinants of a disease, we can make rational decisions about what to do about them. You may not be surprised to know, for example, that air pollution contributes to heart disease. Air pollution isn’t a major contributor to heart disease, but it does contribute a little bit. It is also something that can be hard to avoid. The problem with air pollution is that there are competing interests that may affect how we deal with it. Does that little risk justify the big costs that would come with big policy changes? You know the politics of environmental regulation. For me, I don’t care about the answer to the cost question in making my own life choices. I know air pollution is bad so I’ll try to avoid it the best I can. The great thing about the risks from saturated fat and cholesterol is that I can control them almost completely while saving money. For me, avoiding them is even more of a no-brainer than is avoiding air pollution, and fortunately, I have a lot more control over my diet than I do over the air I breathe.

I doubt I’ve told you anything here that you either didn’t know before or that you don’t recognize to be a simple matter of common sense. The fact that someone like Ravnskov would try to fool us this way speaks to the frailty of his argument against the lipid hypothesis. I don’t think a fair-minded person who understands the concept of composite cause would be able to see things his way.

To me, the single cause fallacy is the most obviously contrived and manipulative line of argument used by the confusionists. However, it is not the most morally offensive. The next video might make you angry. It takes a lot of nerve to pretend that the richest of the rich are dealing with the same health challenges as the poorest of the poor. Nevertheless, the cholesterol confusionists are not above using this ugly tactic. I’ll show you how this one works in the next video.