Saturated fat…here is part three, where we will discuss the research on the topic.
In part one I discussed the problems with relying on science in general. Some people like accepting the “scientific consensus” on something because it seems safe to go along with majority opinion. The problem with this strategy is that you don’t end up doing the research yourself. When you do, you may find something the majority did not. Also, the conventional point of view regarding all public health related matters is not always based on evidence. We could debate that all day.
In part two I described the idea of food synergy, and why this is a missing link in nutrition science research.
In this final segment I will share what the actual research shows regarding saturated fat intake and cholesterol intake on your health. In addition, I will show you how to find this research easily.
What is the research question?
We will shortly go to pubmed to search for studies on saturated fat and cholesterol consumption. Before we go to pubmed, let’s first clarify our research question. The research question I am answering in this post is “what is the effect of saturated fat and cholesterol intake on cardiovascular disease (CVD) and all-cause mortality?” This is a specific question. “Is saturated fat bad for you” is NOT a specific question. In the research world you need specific questions for a few reasons.
For one, a specific question is easier to study and the results are easier to draw a conclusion from. If we were to design a research study to answer if saturated fat was good for you in general we would probably be answering a lot of other specific questions. We could assess if it’s good for your health, which leads to more specific questions about health. But we could also ask if it makes you feel good to eat it, which is an entirely different question. If we didn’t ask specific questions, and say just designed a survey that asked people if they thought saturated fat was good for you, and they could answer in an open ended way, our data would be quite jumbled and not as meaningful to interpret.
The effect of saturated fat, if any, on cardiovascular disease outcomes and all-cause mortality asks two different questions. The first question is looking at saturated fat and cardiovascular disease outcomes. We could define those outcomes as total heart attacks, deaths from heart attacks, number of strokes, deaths from stroke, and diagnosis of cardiovascular disease.
The second question asks if there is a relationship between saturated fat intake and all-cause mortality. All-cause mortality means death from any cause. Smoking, being sedentary, and eating a standard American diet (high sugar, high amount of processed foods, low antioxidant intake) increases your chances of dying in general from any cause. Generally, unhealthy things make you die sooner, so that is the idea this question is getting at.
Death from a specific cause or any cause is arguably the most important outcome from a medical standpoint. I personally care about all the things leading to death (because we may be living somewhat longer in modern times but it’s questionable if we are healthier), but getting wrapped up in the details sometimes isn’t very fruitful. Drugs for example may reduce risk factors for a disease but make you die sooner from that disease through some unknown mechanism or from some other disease, making the drug pointless.
The other research question is the same, but we will replace saturated fat with cholesterol to get this: “what is the effect of cholesterol intake on cardiovascular disease and all cause mortality?”
What is cardiovascular disease…
The next point I want to make before discussing the research is that cardiovascular disease is sometimes interchanged with heart disease and neither are a single entity. There are a multitude of afflictions affecting the heart, including coronary artery disease (CAD), peripheral artery disease (PAD), ischemic heart disease (IHD), angina pectoris, stroke, arteriosclerosis, hypertension, cardiomyopathies, aortic aneurysm, congestive heart failure, and valvular disorders. Many of them may involve atherosclerosis (plaque build up inside arteries, impeding blood flow). Cardiovascular disease affects the heart, either due to a dysfunction with the heart itself, plaque buildup in the coronary artery, or plaque buildup throughout the vasculature, making it harder for the heart to do its job.
So what is heart health?
In general, if something improves heart health it helps the heart do its job better. If you have lower extremity artery disease (LEAD), something good for promoting blood flow in the lower extremity arteries may improve the return of blood to the heart which could help it pump more effectively. In congestive heart failure, depending on what is causing it, treating that root cause will help the heart pump blood better, reducing stress on the myocytes (name of the cells composing the heart). Off the top of my head, I’m thinking maybe beet juice could help with this, as it promotes blood flow and nitric oxide signaling (many things do this), which dilates the arteries, enhancing blood flow. This might work. It depends on the type of CHF and cause, along with the individual person.
What is mostly true is that many heart diseases can be prevented by not smoking, limiting alcohol intake, exercising regularly, and eating a varied diet rich in antioxidants and nutrients. Most of the diseases that kill people most likely can be prevented this way.
Finding the research
Let’s go to pubmed now and look for some research. I’ve learned a few things from my research classes about getting a good search result here. It takes practice. So watch first what happens when I type in “saturated fat” in the search box.
Not bad actually, when you have “best match” chosen (circled). If you chose instead “most recent” you would get less relevant results. Try it out yourself. We will definitely look at articles 2 and 4 (check marked).
The thinner circles at the bottom bring up the next subject. Instead of pouring through individual studies, we can look for systematic reviews or meta-analyses. Dr. Uman defines these two terms in her article as follows:
“Systematic reviews, as the name implies, typically involve a detailed and comprehensive plan and search strategy derived a priori, with the goal of reducing bias by identifying, appraising, and synthesizing all relevant studies on a particular topic. Often, systematic reviews include a meta-analysis component which involves using statistical techniques to synthesize the data from several studies into a single quantitative estimate or summary effect size (Petticrew & Roberts, 2006).”
Reviews just review what’s out there, and a meta-analysis takes a bunch of studies, pools their data, and performs a new statistical analysis.
Refining the search
We will now search “saturated fat intake cardiovascular disease all cause mortality,” checking just systematic reviews and meta analyses. 13 results show up.
Not bad. Now we will go one step further and go to the advanced search. Click “advanced” below the search box. We are now going to tell the search feature to look for all studies matching two different search phrases. You can simultaneously search for two things at once with one search by using the phrase OR. My search query is now “saturated fat intake AND cardiovascular disease OR saturated fat intake AND all cause mortality.”
17 results! WOW. Just kidding…the 4 additional studies here aren’t really answering our question. So in this case the advanced search wasn’t necessary.
Now let’s compare this to google searching. To my surprise, there is one article (circled below) that did not show up in my pubmed search results. I’ll have to ask my professor about that. One of the other articles that shows up was in my search results.
We will talk about the article shortly. Next, let’s click on “scholarly articles” (where the arrows are pointing). To my surprise, most of the articles on the first page aren’t answering my question.
I’m not interested in flavonoid intake google! Well atually I am but not right now. I am drinking some pour over coffee I made myself and there are flavonoids and polyphenols in there.
Article discussion time
One of the articles google found was not in my pubmed search. This article is titled “Food Sources of Saturated Fat and the Association With Mortality: A Meta-Analysis.” The researchers here set out to answer a few different questions. They were interested in whether or not a high consumption of foods containing saturated fat increased all cause mortality, cardiovascular disease mortality, or cancer mortality. They found a slight inverse association with milk, a low positive association with meat, and a moderately positive association with processed meat consumption. In general they said the evidence is too weak to support the idea that eating meat promotes mortality from all causes. They reviewed 26 different studies. That study deserves a separate post. I will keep it brief here as we still have to talk about cholesterol after this.
The second study I want to discuss is titled “Intake of saturated and trans unsaturated fatty acids and risk of all cause mortality, cardiovascular disease, and type 2 diabetes: systematic review and meta-analysis of observational studies” (it’s a free article). They answered the same questions I want to answer (we can decide if their answer is definitive or not) and a few more. They look for an association with saturated and trans unsaturated fat with all cause mortality, CHD mortality, CVD mortality, CVD incidence, ischemic stroke, and diabetes incidence. They found no association with saturated fat and all cause mortality or cardiovascular disease. For trans fat though it’s a different story; they differentiated between natural trans fats (the ones found in grass fed meat and milk) and industrial trans fats (products of hydrogenation of vegetable oils commonly found in processed foods). Total trans fat intake was associated with a statistically significant increase in all cause mortality. Ruminant trans fats were inversely associated with diabetes (but no association was seen with cardiovascular disease). This association was the strongest out of any other association in the study, and warrants a future post.
There are plenty of other articles like these you can find from the pubmed search we went through. In general it seems as if there is no association whatsoever with saturated fat intake and cardiovascular disease or all cause mortality. In part four we will discuss the strength of these findings and come to a conclusion on whether or not we should agree with this.
Using the methods above, I searched for “cholesterol intake cardiovascular disease.” I found a paper titled “Dietary cholesterol and cardiovascular disease: a systematic review and meta-analysis.” This paper answered many questions. One of them was if cholesterol intake had an effect on VLDL (very low density lipoprotein). This is one of the “bad” cholesterols, but it’s not so simple. They evaluated three different trials to answer this question and basically found no association. They also didn’t find much association with cholesterol intake and other serum lipids in general.
They found six different trials investigating cholesterol intake on coronary artery disease (CAD). Only one (a study in a Japanese population) found a statistically significant association. For the rest no positive association was found. Cholesterol intake also had no effect on death from CAD, except in one study where an effect was seen in men who consumed about 430 mg of cholesterol per day. This study had more vegetarians than omnivores, so there could be something else mediating this relationship. 430 mg of cholesterol a day is about a few eggs worth. This may sound alarming but it’s not, as it was just one study, and many other factors could have potentiated this positive relationship. I will have to look into that study separately.
Other studies they evaluated found no association with cholesterol intake and any CVD. Whether its CAD or CVD the results are pretty consistent that there is no positive relationship between cholesterol intake and those diseases. I don’t think they answered all cause mortality so we will just skip over that for now and say don’t worry about it!
Conclusion: In general when studies are discussed we should look at their methods, the population studied, and the outcomes they assessed, in more detail. Here I am just sharing the overall conclusions from some of the most robust analyses on the questions we chose to answer. If there was a huge effect of eating more saturated fat, which is naturally found in milk, cheese, butter, meat, and coconuts, on your heart health, we would expect to see its effect in such large studies where tens of thousands of people are evaluated. There is strength in numbers; the larger the population size the stronger the conclusion from statistical analyses one can draw.
The downside is that individuality is lost. What about meat eating among those who follow a paleo diet versus meat eating in those who follow a standard American diet versus those who eat meat in moderation and try to be healthy but aren’t strict about following a diet? Those are all different populations and I’d have to do more digging to see if studies have been done on these populations. For now, it’s settled. Saturated fat won’t kill you. Eating cholesterol doesn’t cause heart attacks and likely has no relationship with all cause mortality. Emphasize whole food once again and try to connect to nature as much as you can. Choose grass-fed meat whenever possible. Choose pasture-raised eggs. Be proactive about what’s going in to your dinner plate. I know I could even improve in that with my busy lifestyle! I have a quiz tomorrow morning I have barely studied for!
In part four I will talk more about statistics and how to draw conclusions from studies like the ones we analyzed. Hope you learned something.
I’m also on Instagram, Twitter, and Facebook . There is no newsletter at the current time. I may start one eventually but right now it’s not necessary! If you have questions about health you can also email me (firstname.lastname@example.org).