What about Saturated Fat?

Now that dairy has been vindicated, let’s take a look at another highly controversial dietary component: saturated fat.

Here’s a brief review of the debate which began back in the 1950’s.

By the mid-20th century, heart disease had become the leading cause of death in the US. President Eisenhower had a heart attack which brought the heart disease issue to the forefront of the country’s focus. Researchers learned that cholesterol was a component in the plaques that block peoples’ arteries. They knew that dietary saturated fat (fat that comes mostly from animals and tropical oils) could increase the amount of cholesterol in peoples’ blood. Therefore, they hypothesized that saturated fat raised cholesterol in the blood which then formed plaques in the arteries which then caused heart attacks.

Nutrition researcher Ancel Keys published a now infamous study which claimed that men in countries that ate more saturated fat had much more heart disease than men in countries that ate less saturated fat. The United States Department of Agriculture (USDA) then published guidelines encouraging people to eat fewer than 10% of calories from saturated fat.

In recent decades, Keys’ study has been criticized for only reporting data on countries that supported his hypothesis. Some other studies have not been able to replicate his findings. Furthermore, we now have a better understanding of how different types of cholesterol work in the body and how other nutrients may contribute to heart disease. These 3 issues have raised controversy over whether saturated fat should continue to be limited in a healthy diet.

Initial Google Search

Most mainstream organizations such as the USDA and American Heart Association (AHA) continue to recommend reducing dietary saturated fat to 5-10% of total calories. These claims come directly from the AHA:

  • Eating foods that contain saturated fats raises the level of cholesterol in your blood. High levels of LDL cholesterol in your blood increase your risk of heart disease and stroke.
  • Decades of sound science has proven it can raise your “bad” cholesterol and put you at higher risk for heart disease.
  • Eleven authoritative bodies – including the World Health Organization; the Institute of Medicine; the governments of the United States, the United Kingdom; and the European Union – independently reviewed the scientific evidence (through November 2017) and concluded yet again that saturated fat is associated with heart disease.
  • Research shows that a diet rich in refined, simple carbohydrates is equally if not more detrimental to health than a fatty diet.
  • Not all fats are created equal. Saturated fats increase risk for heart disease, but that’s not the case with unsaturated fats known as monounsaturated and polyunsaturated fats. These fats are found in fish, nuts, seeds and oils from plants. However, it’s important to remember that unsaturated fats do contain calories. And too many calories can lead to weight gain. Still, unsaturated fats are generally better.

Groups and individuals that question these recommendations cite the following:

  • Saturated fats raise both LDL (cholesterol that can clog arteries) and HDL (cholesterol that clears away clogs in arteries) resulting in a net balance effect. Therefore, total cholesterol is not a good indicator of risk.
  • There are even different types of LDL. The smaller LDL particles are more likely to become oxidized and clog arteries while the larger LDL particles are less likely. Smaller LDL particles increase due to carbohydrate consumption. Larger LDL particles increase due to saturated fat. Therefore, carbs are more likely to cause heart disease than animal fat.
  • Correlation is not causation. Countries with higher saturated fat consumption may have more heart disease, but these same countries also have higher sugar consumption. Sugar may be more dangerous than saturated fat.
  • Many other studies and meta-analyses have found that saturated fat consumption does not affect heart disease risk.
  • Using polyunsaturated fats as a replacement for saturated fats may lower LDL, but they also lower HDL.
  • Non-saturated fats oxidize more easily when heated (during cooking) and in the body which causes inflammation and heart disease.
  • Polyunsaturated fats are primarily omega-6 fatty acids which increase inflammation when not balanced out by omega-3 fatty acids.
  • The AHA and other similar guidelines may be based on cherry-picked or poorly interpreted data.

Of note, both sides equally agree that trans fats (hydrogenated and partially hydrogenated oils) cause more harm than any other kind of fat.

After reading both sides of the argument, I really don’t know which way to lean. Luckily, tons of high quality research has been done on this topic.

To find relevant research, I used the following inclusion criteria: saturated fat (search term), heart or cardiovascular disease outcomes, literature reviews or meta-analysis, done within past 5 years, access to full text in English.

I found 14 relevant literature reviews and meta-analyses from the past 5 years that met all the criteria and compiled them here. In order to make sense of the findings, we must be familiar with two important types of research design: randomized controlled trials and prospective cohort studies.

A randomized controlled trial (RCT) is a study in which people are allocated at random either to receive the clinical intervention being tested or to be a part of the control group that will receive either no intervention or the standard intervention (not the new intervention being tested). RCTs provide the most reliable type of evidence. You can usually say that something happened as a result of the intervention without wondering whether another factor was at play. With RCTs you can be more sure that X caused Y and not just that X is associated with Y.

The second most common type of study in the research below is the prospective cohort study (PCS). These studies follow a large group of people over a long time to determine how their behaviors or characteristics affect certain outcomes. These studies can give us a lot of good information, but they cannot let us say for sure that X caused Y. We can only say that X is associated with Y. This leaves room for cofounding variables that may have been the real cause of Y.

A few other terms should be identified in order to make sense of the rest of this blog. Here they are:

  • Saturated fatty acid (SFA): fat molecules that have no double bonds. These fats are solid at room temperature. They are found mainly in meat, dairy and tropical oils.
  • Monounsaturated fatty acid (MUFA): fat molecules that have one double bond. They are liquid at room temperature and are found mainly in olive oil, olives, avocados, nuts, sunflower oil, sesame oil, and animal products.
  • Polyunsaturated fatty acid (PUFA): fat molecules that have more than one double bond. They are liquid at room temperature and found in nuts, seeds, fish, seed oils and oysters. Depending on where the first double bond lies, they are classified as either omega-3 or omega-6.
  • Cardiovascular disease (CVD): catch all term for any cause of heart disease, usually caused by damaged blood vessels.
  • Coronary heart/artery disease (CHD or CAD): heart disease involving the heart’s major blood vessels
  • Low density lipoprotein (LDL): protein that carries cholesterol through the body. LDL cholesterol is known as the “bad cholesterol” because it can build up on the walls of blood vessels and cause heart disease. LDL comes in various sizes. The smaller the particles, the more likely they are to stick to the blood vessels.
  • High density lipoprotein (HDL): protein that carries cholesterol to the liver and out of the body. HDL is known as the “good cholesterol”.
  • American Heart Association (AHA): large non-profit organization providing research and guidelines on how to reduce heart disease and stroke.
  • United States Department of Agriculture (USDA): government program responsible for developing laws related to farming, forestry and food.
  • Randomized Controlled Trial (RCT) See above
  • Prospective Cohort Study (PCS) See above

The research below begins with the AHA guidelines because this document is the most influential driver of USDA policy. The other research either aligns with this article or attempts to show that it is fully or partially inaccurate. So, let’s take a closer look.

Peer Reviewed Research (Click here to skip the research and go straight to the conclusions)

1. Dietary Fats and Cardiovascular Disease: A Presidential Advisory From the American Heart Association. (2017) https://www.ncbi.nlm.nih.gov/pubmed/28620111

This article relies heavily on 4 high quality randomized controlled trials conducted in the 1960s that directly measure replacing saturated fat with polyunsaturated fat over the course of at least 2 years. These 4 trials had important inclusion criteria. They did not include trans-fat as a major component, they controlled the dietary intake of both the test and control groups, the proved adherence to the diet by using biomarkers, and they collected information on cardiovascular or coronary disease events.

It also discusses 6 other randomized controlled trials that replaced saturated fat with nonsaturated fat, but may not have met other the other criteria. There have been no randomized controlled trials that replaced saturated fat with whole grains, fruits and vegetables instead of nonsaturated fat and no randomized controlled trials that purposefully replaced saturated fat with refined carbohydrates. There have also been no more recent randomized controlled trials that fit the criteria.

All 10 of these randomized controlled trials found that replacing saturated fat with nonsaturated fat resulted in a decrease in CVD and that LDL was lowered an average of about 30% (which is about the same as we would expect from a statin drug). The article goes on to show through meta-regression analysis that reductions in LDL cholesterol after diet change correlate well with the extent of reductions in CVD. They discuss “very strong evidence that satisfied rigorous criteria for causality” which means there is very high quality evidence leading us to believe that high LDL indeed causes CVD and that lowering LDL indeed lowers CVD.

The article also cites several other interesting studies showing that different types of saturated fat make little difference. Coconut oil and dairy fat did not have protective benefits and seemed to be similar to any other saturated fat in terms of CVD.

Additionally, they cite a study showing that consumption of saturated fat reduces the anti-inflammatory potential of HDL. This would mean that even though saturated fat raises both good and bad cholesterol, it makes the good cholesterol not work as well.

The authors identify what they say is the main reason why some meta-analyses found different results. Prospective cohort studies of PCSs (as opposed to randomized controlled trials) generally saw participants replacing saturated fat with packaged low-fat foods (refined carbohydrates). These studies found that reducing saturated fat was not associated with reducing CVD. Some of them even saw a rise in CVD. The AHA readily acknowledges that saturated fat should not be replaced with refined carbohydrates.

2. Fat, Sugar, Whole Grains and Heart Disease: 50 Years of Confusion. (2018) https://www.ncbi.nlm.nih.gov/pubmed/29300309

This article does not deny the findings from the studies in the AHA article, but it places much stronger blame on refined carbohydrates than SFAs. The review cites PCSs that show the association between saturated fat and CHD is present, but weak. The author confirms that RCTs show that increasing PUFAs while decreasing SFAs will lower the ratio of total cholesterol to HDL, lower triglycerides, and reduce CHD. However, he cites evidence suggesting that these RCTs may have been poorly controlled and less reliable than we once thought. He emphasizes that PCSs show a strong association between sugar/refined carbohydrates and CHD. He also stresses that fruits, vegetables, fish, and whole grains are inversely associated with CHD (protective against heart disease).

3. Reduction in saturated fat intake for cardiovascular disease. (2015) https://www.ncbi.nlm.nih.gov/pubmed/26068959

This meta-analysis looked at whether it is best to replace SFAs with PUFAs, MUFAs, carbohydrates or protein. They included only randomized controlled trials. The researchers found the strongest decrease in CHD when SFAs were replaced with PUFAs. Their findings align well with the AHA guidelines. This analysis did not find any reduction in CHD when SFAs were replaced by carbohydrates or protein, but it did not specify carbohydrate type. They stated that the evidence for MUFAs was inconclusive due to only one trial including them. Overall, this review is very similar to the AHA review and had similar conclusions.

4. Saturated Fats Versus Polyunsaturated Fats Versus Carbohydrates for Cardiovascular Disease Prevention and Treatment. (2015) https://www.ncbi.nlm.nih.gov/pubmed/26185980

This literature review is similar to the study 2 above. The authors start by acknowledging that replacing SFAs with PUFAs, especially omega-3’s, seems to improve cardiovascular health in several RCTs. They also state that when SFAs are replaced by carbohydrates, there is either no change, or a decrease in cardiovascular health. They cite other health problems caused by refined carbs and sugars and agree with the US Dietary Guidelines of limiting added sugars to less than 10% of calories per day. They say that urging people to eat only 5% of calories from SFAs may lead people to eat more sugar, so they tend to agree with the past recommendation of 10% of calories from SFAs.

The authors describe how fats and carbohydrates act differently in the body. SFAs increase large LDL whereas refined carbohydrates increase small LDL. The small LDL particles are more strongly associated with CVD than the large ones.

This study also looks at how different foods with SFAs act differently in the body. They cite research suggesting that dairy fat does not raise LDL the same way other sources of saturated fat might. They acknowledge that beef is associated with CVD, but that this may be attributed to another factor, carnitine-derived metabolites, rather than SFAs. They cite evidence that butter, palm oil and coconut oil do raise LDL, but perhaps not as much as we would predict given the specific types of fatty acids they are made of.

They also discuss how weight loss is the key factor to improving cardiovascular health no matter which macronutrients you eat. They conclude that overall dietary patterns are more important than focusing on specific macronutrients or fatty acids. They emphasize vegetables, fish, nuts and whole grains. Their overall tone is that the effect of SFAs on heart disease has not been well established.

5. 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. (2015) https://www.ncbi.nlm.nih.gov/pubmed/26268692

From reading the abstract, this review seems to refute the AHA’s claim that replacing SFAs with PUFAs will reduce CVD. However, from the title, we can see that this review only includes observational studies. That means they used prospective cohort studies and not randomized controlled trials. We can also see that they studied trans fats which are already known to increase CVD. Upon reading the details, the PCSs cited here saw people swapping out saturated fat for refined grains and sugar, just as the AHA article pointed out. So, upon further examination, this review actually doesn’t claim that replacing saturated fat with nonsaturated fat doesn’t help. It only claims that replacing saturated fat with refined carbohydrates doesn’t help. These findings really aren’t different from the AHA claims.

6. The Evidence for Saturated Fat and for Sugar Related to Coronary Heart Disease. (2016) https://www.ncbi.nlm.nih.gov/pubmed/26586275

This review makes the claim that sugar (especially fructose) is more likely to cause CVD than saturated fat. As we now know, this is not in contrast with the AHA guidelines. This review also cites studies that suggest certain types of saturated fat may be protective to the heart by increasing HDL or large LDL particles. There still seems to be debate over whether this is true. By the end of the review, the authors are simply urging people to eat more real food and “living botanical plants” rather than replacing saturated fats with highly processed foods. They wish the AHA would also concentrate on foods instead of individual fatty acids.

7. Dietary Fat and Risk of Cardiovascular Disease: Recent Controversies and Advances. (2017) https://www.ncbi.nlm.nih.gov/pubmed/28645222

This review reiterates what we’ve already seen. Replacing SFAs with PUFAs will decrease LDL significantly. Replacing SFAs with refined carbohydrates does not. Replacing SFAs with whole grains decreases LDL slightly. They cite two PCSs that associate dairy fat with reduced CVD risk, but then cite 3 PCSs where dairy fat was replaced by PUFAs. Replacing dairy fat with PUFAs was associated with decreased CVD risk. Another RCT found that CVD was reduced when butter was replaced by vegetable oil and full fat milk was replaced by skim milk with a vegetable oil emulsion.

Importantly, this review cites several studies refuting the claim that omega-6 PUFAs lead to inflammation. They explain how PUFAs break down in the body and are not associated with a net increase in any inflammatory markers. They cite many studies showing that omega-3 PUFAs (mainly from patients eating fatty fish) decrease inflammation and CVD risk, but the evidence on fish oil supplementation was inconclusive. The review also cites evidence showing beneficial effects of MUFAs (mainly from nuts and olive oil) on LDL and CVD. Overall, this review matches the AHA recommendations.

8. Dietary fats and health: dietary recommendations in the context of scientific evidence. (2013) https://www.ncbi.nlm.nih.gov/pubmed/23674795

This literature review is written by Glen D. Lawrence, author of “The Fats of Life”. Although his article does not discuss any RCTs, he presents convincing arguments that sugar is primarily to blame for the rise in heart disease and several other diseases including cancer and diabetes. This alone does not stand in contrast to the AHA claims. He admits that SFAs increase LDL, but stresses that they also raise HDL. Lawrence cites data supporting positive health effects of dairy fat and coconut oil and stresses the importance of high quality non-processed meats. The majority of his article discusses polyunsaturated fatty acids (PUFAs) which are what the AHA would claim to be the “good” kid of fats. He cites research claiming that PUFAs are susceptible to oxidization whereas SFAs are not. These oxidized particles (from PUFAs and carbohydrates) are what really cause atherosclerosis and inflammation. He claims that cooking meats at high temperatures causes oxidation of the PUFAs in the meat which then causes heart disease (and cancer) rather than the SFAs in the meat. He also argues that omega 3 PUFAs help mitigate the harm done by omega 6 PUFAs. To simplify, Lawrence’s article argues that SFAs from natural sources are healthy in moderate quantities, and that omega 6 PUFAs and sugar are causing heart disease.

9. It is time to revisit current dietary recommendations for saturated fat. (2014) https://www.ncbi.nlm.nih.gov/pubmed/25293492

This review attempts to point out flaws in the current AHA recommendations. They state that the impact of SFA on blood cholesterol is undisputable, but the resulting impact on CVD risk is not as straight-forward which has caused people to misinterpret data. They cite research showing that SFAs increase the large LDL particles and SFAs reduce triglycerides better than carbohydrates. They also claim that the evidence regarding PUFAs and oxidation/inflammation is unclear. They point to the same research we’ve already seen that associates dairy fat with a lower risk of CHD. In sum, these authors feel that the recommendations against SFAs have ignored data to the contrary.

10. Emerging nutrition science on fatty acids and cardiovascular disease: nutritionists’ perspectives. (2015) https://www.ncbi.nlm.nih.gov/pubmed/25979506

This review of the literature fully supports the AHA recommendations. They break down different components of saturated and unsaturated fats and describe how each affects the human body. The authors feel that the research supports replacing SFAs with PUFAs (like vegetable oil) rather than MUFAs (like olive oil). They also remind us that all oils and fat sources contain certain amounts of SFAs, PUFAs and MUFAs.

11. Evidence for and against dietary recommendations to prevent cardiovascular disease. (2015) https://www.ncbi.nlm.nih.gov/pubmed/26175635

This author of this review discusses how some SFAs may be beneficial. For example, she states that lauric acid (high in dairy and tropical oils) has a favorable impact on the total cholesterol to HDL ratio and that stearic acid (high in beef tallow and cocoa butter) has no effect on cholesterol. She also questioned why the AHA guidelines only used LDL, HDL, triglycerides and blood pressure to measure outcomes and not obesity, diabetes or metabolic syndrome. She cited the difference between small and large LDL particles as we have seen in other reviews. She credited the AHA and USDA for recommending decreased sugar intake and acknowledging that processed meats are more harmful than red meats. Her conclusion emphasized the need to focus on whole foods when making recommendations rather than specific nutrients like saturated fat.

12. The relation of saturated fatty acids with low-grade inflammation and cardiovascular disease. (2016) https://www.ncbi.nlm.nih.gov/pubmed/27692243

This review is very similar to the one above. The authors’ main argument is that SFAs are just one of many lifestyle factors that can cause heart disease. These authors come from a background of studying Paleolithic nutrition and lifestyle. They go into great detail about research questioning whether SFAs are actually responsible for heart disease. One interesting point they make is that two thirds of people admitted for “acute coronary events” suffer from metabolic syndrome, but 75% of these people have normal LDL and total cholesterol. They are questioning the importance of LDL as a risk factor. Then they go on to make the claim we have seen before that SFA increases large LDL particles, but not small ones, which truly cause CVD.

They spend much of the article discussing the importance of omega-3 PUFA and how it works synergistically with SFAs. They cite research claiming that harmful effects of SFAs are only present when omega-3 PUFAs are too low.

A few other points they make include: (1) human breastmilk contains high amounts of SFA, (2) wild animals have more MUFA and PUFA and less SFA than farm-raised animals, (3) high fat diets are better than low fat diets at improving blood pressure, HDL, triglycerides, and fasting glucose.

In the end, they call for a variety of diet and lifestyle recommendations: increased consumption of fish, vegetables, fruits, fiber, and whole grains; decreased consumption of salt, alcohol, sugar and fructose; more exercise and sleep, and less stress, anxiety, depression and pollution. These final recommendations don’t look too far off from what the AHA recommends.

13. Dietary fatty acids in the secondary prevention of coronary heart disease: a systematic review, meta-analysis and meta-regression. (2014) https://www.ncbi.nlm.nih.gov/pubmed/24747790

This meta-analysis analyzed 21 prospective cohort studies with more than 7,000 participants who had all already had a coronary event. Some of these studies found that replacing SFA with MUFA or PUFA had favorable results while other did not find significant results. The authors found that overall, the results of dietary fat manipulation were not significant when it came to a second coronary event. However, the researchers recommended replacing SFAs with omega-3 PUFAs in their conclusion.

14. Effect of the amount and type of dietary fat on cardiometabolic risk factors and risk of developing type 2 diabetes, cardiovascular diseases, and cancer: a systematic review (2014) https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4095759/

This is a systematic review that combined results from 607 studies including RCTs, PCHs and some case studies in order to figure out whether the type of fat people consumed had an effect on heart disease, diabetes and cancer. Here is a list of what they found:

Substitution of SFA with MUFA and/or PUFA decreases LDL, but there were no firm conclusions about which type (small or large LDL) was decreased.

There is convincing evidence that replacing SFA with PUFA can reduce CVD events by 10-20%.

The type of fat had no impact on body weight.

When PUFA replaced SFA or carbohydrates, the risk of type 2 diabetes was reduced.

Specific SFA from dairy may have an inverse relationship with type 2 diabetes, but results were inconclusive.

There is probable or convincing evidence linking increased body fat with most types of cancer, but the evidence linking specific types of fat with specific types of cancer is weak.

What We Know and Don’t Know

After analyzing the research, we have 4 reviews that completely align with the American Heart Association recommendations, 4 reviews that mostly align, and 5 that seem to refute the recommendations.

None of the research can deny that replacing SFA with PUFA and/or MUFA will lower LDL. The debate arises as to which type of LDL is being raised by SFA. Is it the large LDL particles that are less likely to cause heart disease, the small LDL particles which are more likely to cause heart disease, or both? There is also debate about how much LDL really matters. Some think we should focus more on triglycerides, the ratio of total cholesterol to HDL or other markers of metabolic syndrome.

These questions are not adequately answered in the research above. All we know for sure is that the RCTs (remember, this is the most reliable type of evidence) found that people had fewer coronary events (aka heart attacks) when SFA was replaced by PUFA or MUFA.

The research also consistently shows that replacing saturated fat with trans-fat or refined carbohydrates increases heart disease risk. This is probably the most important take home point, so I’ll repeat it:

Everyone agrees that refined carbohydrates increase the risk of heart disease even more than saturated fat!

Some researchers just feel that we should stop talking about SFA and only focus on reducing sugar/carbohydrate.

Some research shows that the source of the saturated fat (whole milk, butter, egg yolk, coconut oil, meat, etc.) makes little difference. However, there is still considerable debate over whether full-fat dairy could have protective effects. As we saw in my previous blog, all types of dairy have been associated with positive health outcomes.

The evidence in support of tropical oils and butter does not seem to suggest these are good for you, only that they may not be as bad as we once thought. None of the research I read specifically singled out egg yolks.

Another point of contention is oxidation and inflammation. PUFAs are the most likely type of fat to be damaged by oxidation because of their chemical structure. Some researchers argue that oxidation causes inflammation which causes heart disease. I was unable to find the AHA’s response to this claim.

Finally, some argue that the balance of omega-6 and omega-3 PUFA in approximately a 4:1 ratio is critical. Most research studies did not address this issue, but some researchers feel there is reason to believe SFAs are only harmful because people these days don’t consume enough omega-3 PUFA. Regardless of the importance of balancing types of PUFA, all the research supports consuming more omega-3 fatty acids from real foods, especially fish.

Many of the articles reviewed here stress the need to create guidelines based on actual foods rather than nutrients. Others say things like, “We should shift the focus from reducing saturated fat to reducing sugar in the diet”. In reality, the research supports reducing saturated fat AND reducing sugar/refined carbohydrates.

Conclusions and Applications

At the end of the day, all of the research I found concluded that we really should stick to vegetables, fruits, fish, lean meats, nuts, seeds, dairy, and whole grains in order to be healthy. Considering all the research together, it would seem reasonable to suggest a heart-healthy diet of 5-10% of calories from saturated fat, 5-10% of calories from sugar/refined carbohydrate (sorry alcohol is included in this category too), and as little trans-fat as possible. It would be even better to show people examples of how to eat according to these guidelines in terms of real foods.

Let’s take a look at how we might put this into practice with the typical 2,000 calorie diet. If we split the difference and have 7.5% calories from saturated fat and 7.5% calories from refined carbs, added sugar, or alcohol, we get 150 calories for each. I considered making sample menus using these guidelines, but I actually think that would make a simple message more complicated.

Many nutrition experts say you should never classify foods as “good” and “bad”. I’m going to break that rule. Based on the research I found,

Saturated fat is mostly bad for you. Sugar, refined carbohydrates, and alcohol are even worse.

All questions and comments are appreciated!