For decades, a mist of low-fat dogma has covered the nutritional landscape. Low-fat products have made their way into grocery stores all over the world, and high-fat foods have been demonized as artery-clogging foes that should be avoided at all cost. This situation arose largely because public health authorities told us that we should limit our consumption of fat, in particular the saturated kinds, and instead eat more carbohydrates. This advice may have done us more harm than good, as it has led a lot of people to replace nutrient-dense, whole-foods such as eggs, coconuts, and grass-fed muscle and organ meats with less nutritious, carbohydrate-rich products.
However, if you’ve been paying attention, you’ve probably noticed that over the last decade – and in particular the last couple of years – the pendulum has started to swing in the opposite direction. Low-carb diets have experienced an upswing in popularity, and more and more authors, diet bloggers, journalists, and dietitians make the case that it’s long overdue to “end the war on fat” (1, 2, 3). Some even go as far as to say that we should derive a substantial portion of our calories from saturated fat. This idea has long been popular in the low-carb community, but it wasn’t until recently that it started to make its way into the more conventional parts of the nutritional scene.
When I first got properly introduced to Paleolithic nutrition, ancestral health, and low-carb dieting about a decade ago, I too was led to believe that conventional wisdom surrounding saturated fat was all wrong, and that butter, bacon, cream, and cheese were all healthy foods that could be eaten in fairly large quantities without any ill effects. However, I quickly noticed that this dietary regime didn’t provide me with good health, and when I started to dig into the actual science on the topic and examined the evolutionary evidence, I realised that I had been led astray.
I have nothing against low-carb diets per se. Actually, I think most people benefit from eating less carbohydrate than what the official dietary guidelines recommend. That said, I think it’s more important to focus on eating healthy foods, than to strive for a specific macronutrient ratio.
As I’ve pointed out in many of my articles over the years, I don’t think there’s any reason to fear eggs, grass-fed organ meats, coconuts, and other Paleo-approved whole foods just because they contain saturated fat and/or cholesterol. However, when it comes to evolutionarily novel foods with a very high concentration of saturated fat, such as butter, ghee, cream, bacon and other processed meats, and cheese, I would urge moderation. Here’s why…
1. Saturated fat did not make up a major part of the ancestral diets that conditioned the human genetic make-up
It’s a myth that our preagricultural ancestors ate large amounts of saturated fat. We don’t even need to look at the scientific evidence to understand that it would have been virtually impossible for our primal forebears to take in as much saturated fat as many low-carb dieters do today. All we need is a basic understanding of nutrition and some common sense.
Recent innovations and advancements in food production and processing have allowed us to create products with macronutrient compositions not found in nature. This has been especially apparent during the last centuries, as products with a very high concentration of carbohydrate (e.g., pastries, chocolate), protein (e.g., whey protein supplements), and fat (e.g., refined vegetable oils, ghee, cream) have made their way into the human diet. When compared with the wild plants and animals our preagricultural ancestors consumed, these modern products have an abnormal nutrient composition.
None of the foods that were available to Paleolithic hunter-gatherers contained anywhere near as much saturated fat as for example butter, cheese, ghee, and coconut oil. When compared to the most important sources of fat in the typical hunter-gatherer diet (e.g., meat, nuts), these evolutionarily novel foods have a supernormal concentration of saturated fat and calories, low protein content, and poor micronutrient profile.
Studies of modern-day hunter-gatherers further dispel the notion that saturated fat made up a large part of the diets that supported the evolution of our species. The Hadza, a group of hunter-gatherers that live in East Africa, a part of the world believed to be the cradle of Homo sapiens, obviously don’t sit around eating bacon, cheese, butter, ghee, cream, and similar high-fat foods that often pass for “Paleo” these days. Rather, they derive most of their energy from tubers, berries, wild meat, baobab, and honey.
Even the Inuits, who are known for their extremely high fat intake, don’t consume that much saturated fat, as they derive most of their energy from sea-dwelling animals, which primarily contain unsaturated fats.
Typically, hunter-gatherers derive most of their fat from animal source foods. Even though they eat the organs and fattiest parts of the animals they kill, their intake of saturated fat remains modest, because wild animals tend to be markedly leaner and lower in saturated fat than domesticated animals (4, 5). And it isn’t just a small difference; grain-fed animals contain as much as 2-3 times more saturated fats than game meat, and much less of the essential omega-3 fatty acids (4).
Dr. Loren Cordain estimates in a paper that “the normal intake of saturated fat that conditioned our species genome likely fell between 10 to 15% of total energy” (6). This is slightly higher than the intake level recommended by most public health authorities (<10%), but much less than what you get from a typical very-low carbohydrate diet or “Paleo diet” containing ample amounts of high-fat dairy, dark chocolate, and bacon.
It should be noted that Loren Cordain’s estimates are based on the assumption that our Paleolithic ancestors ate fairly high amounts of animal source foods. While there’s no doubt that meat played a key role in terms of supporting the evolution of our large brains and complex bodies, recent evidence indicates that plant foods may have been a more important part of Paleolithic human diets than previously thought (7, 8, 9). If this was the case, the intake of saturated fat may have been, on average, even lower than 10%.
Some traditional populations, including a couple of the societies Dr. Weston A. Price visited on his journeys around the world, have seemingly maintained good health despite eating fairly high quantities of saturated fat. What is important to remember is that these populations are the exeptions rather than the norm, and that their diets are not a good representation of the diets that supported the evolution of the body and brain of our ancient ancestors. Moreover, a high intake of saturated fat may have very different consequences in the context of an ancestral diet and lifestyle than a modern diet and lifestyle.
For example, the Maasai, a Nilotic ethnic group of semi-nomadic people inhabiting southern Kenya and northern Tanzania, don’t eat coconut oil, butter, and bacon. Rather, they get almost all of their calories from meat, milk, and blood – all of which are derived from animals that live in a habitat that is more congruent with their genetic make-up than what is the case for most domesticated animals today. This is important to keep in mind, because meat and dairy products obtained from wild and domesticated ruminants have different fatty acid compositions. Also, the Maasai lead an ancestral-type lifestyle, are very physically active, and carry genetic mutations that may help them better tolerate their unusual diet (10).
Many of the same things apply to healthy, non-westernized populations that rely on coconut fat as an important source of calories (e.g., the Kitavans, the Tokelauans). These peoples don’t eat refined or highly processed coconut products. Rather, they primarily eat unprocessed coconut meat, which doesn’t just contain saturated fat, but also fiber, water, and other nutrients. Moreover, as I’ll touch on later, the primary fatty acid found in coconut – lauric acid – has some unique characteristics that separate it from longer-chain saturated fatty acids such as palmitic and myristic acid.
Last but not least, contrary to what some people seem to think, we can’t draw causal conclusions about diet and health from observational studies. For example, just because a group of people seemingly maintain good health “despite” eating a diet that is high in saturated fat doesn’t mean that it’s healthy to eat a lot of saturated fat. It may be that these people would have been even healthier if they had replaced some of the saturated fat in their diet with other nutrients and/or it may be that confounding variables are blurring the whole picture.
2. Many of the studies and systematic reviews used to rebut the association between saturated fat consumption and chronic disease risk have several flaws and limitations
Virtually every nutrition student in the world today learns that the public health recommendation to eat less saturated fat is supported by solid evidence. It’s therefore no surprise that most nutritionists and dietitians agree that a high intake of saturated fat is detrimental to health.
Recently, this conventional belief has been brought into question by a couple of systematic reviews and meta-analyses which indicate that the vilification of saturated fat may have been unfounded. While the list of scientific papers supporting this idea is relatively small in comparison to the stack of papers showing that a high intake of saturated fat is indeed harmful, it shouldn’t just be swept away as insignificant.
That said, it shouldn’t be blown out of proportions, either, which is what has happened recently. Some bloggers and published authors make the case that saturated fat is harmless, go on to cite the conclusions of a couple of these recently published papers, and then leave it at that. Often, they neglect to mention that the evidence in this area is not clear-cut, and perhaps more importantly, that it’s always important to look at the weaknesses, strengths, and limitations of a study before jumping to conclusions.
I don’t claim to have all the answers, but if there’s one thing I’ve learned (but still sometimes have to remind myself), it’s that it’s important to be cautious when interpreting and drawing conclusions from scientific studies.
Obviously, I won’t be able to carefully analyze all of the important studies and review papers on saturated fat in this article – without turning it into a long book (Drop me a comment in the comment section if there is a study you’d like me to comment on). What I’ll do instead is to focus on the meta-analysis that is most frequently cited as support for the idea that saturated fat is, in the grand scheme of things, harmless. The paper I’m talking about is the 2010 paper titled “Meta-analysis of prospective cohort studies evaluating the association of saturated fat with cardiovascular disease” (11).
The authors of this study conclude that “there is no significant evidence for concluding that dietary saturated fat is associated with an increased risk of CHD [Coronary Heart Disease] or CVD [Cardiovascular Disease]”. They also point out that “more data are needed to elucidate whether CVD risks are likely to be influenced by the specific nutrients used to replace saturated fat”.
Systematic reviews and meta-analyses are considered to be on top of the hierarchy of scientific evidence, since they combine the results from several different studies that examine similar research hypotheses and thereby attain greater power than a single study. There’s no doubt that we can learn a lot about diet and health from these types of research articles; however, what is important to remember is that a meta-analysis or systematic review is only as good as the studies it’s based on. While the aforementioned meta-analysis does provide some interesting information regarding the link between dietary saturated fat and cardiovascular disease, I would be very hesitant to draw any conclusions from it.
First of all, the data for the meta-analysis were pulled from prospective cohort studies. While these types of studies are useful for finding associations between diet and disease, they can’t prove causality. There are a number of reasons why this is the case, with perhaps the most important being that cohort studies are prone to confounding. Researchers typically try to control for confounders by carrying out multivariate analysis, but there is always a limit to how much you can control for. Residual and unmeasured confounding is a major issue in many observational studies.
Second, other researchers have criticized the study, pointing out that the meta-analysis could not distinguish between different types of substitutions for saturated fat; imprecise dietary methods were used in many of the included studies; there were undisclosed conflicts of interest; and adjustments were made for dietary cholesterol (a mediator, not a confounder) in half of the studies.
Third, the fact that some studies indicate that dietary saturated fat is not associated with an increased risk of CHD, doesn’t mean – as some people seem to think – that saturated fat is harmless. For example, the cohort studies included in the aforementioned meta-analysis only focus on the link between saturated fat consumption and cardiovascular disease incidence; they tell us nothing about the impact saturated fat has on other outcomes.
Fourth, as I’ll show in the next section, other meta-analyses based on higher quality data from RCTs don’t support the conclusions of this study.
3. A high intake of saturated fat may unfavourably impact the blood lipid profile and increase the risk of cardiovascular disease
Within the conventional nutritional community, it’s widely agreed that saturated fat unfavourably impacts the blood lipid profile. However, within the Paleo and low-carb communities, many authors and bloggers argue that this is not the case, and some even question the idea that saturated fat and blood cholesterol levels have anything to do with chronic disease.
If you’re a regular reader of this site, you know I don’t always agree with conventional diet wisdom. However, when it comes to the relationship between saturated fat, the blood lipid profile, and heart disease, I’m inclined to believe that the mainstream nutritional community is more on the right tracks than the low-carb enthusiasts.
One of the most comprehensive review papers on this topic was published in 2003 in the American Journal of Clinical Nutrition (12). The authors of this meta-analysis came to the following conclusion after combining the results from 60 (!) controlled trials:
The situation is much clearer for replacement of SFAs [Saturated Fatty Acids] with cis unsaturated fatty acids. In that case, the effects on surrogate lipid markers (Tables 1⇑ and 2⇑), the epidemiologic findings (89), and the results of controlled clinical trials (104) all suggest that replacement of SFAs with cis unsaturated fatty acids reduces CAD [Cardiovascular Disease] risk.
One of the most interesting findings of this meta-analysis is that lauric acid, the predominant fatty acid in coconut, decreased the ratio of total to HDL cholesterol. It greatly increased total cholesterol, but most of the effect was on HDL cholesterol (the “good” cholesterol). This makes complete sense to me and suggests that coconut fat may be protective against heart disease. (This doesn’t mean that adding spoonfuls of coconut oil to your dinner is a good idea)
As the authors of the study above point out, there’s good evidence to show that a high ratio of total to HDL cholesterol is a risk factor for cardiovascular disease. That said, blood lipid profile does not equal CVD risk. To really be able to elucidate the links between saturated fat consumption and chronic disease risk, we have to look at studies that include hard endpoints. As I pointed out in section 2, several controlled trials have found that individuals who eat a lot of saturated fat are at increased risk of developing cardiovascular disease.
For example, a 2015 review paper published in “The Cochrane database of systematic reviews.” concluded the following:
The findings of this updated review are suggestive of a small but potentially important reduction in cardiovascular risk on reduction of saturated fat intake. Replacing the energy from saturated fat with polyunsaturated fat appears to be a useful strategy, and replacement with carbohydrate appears less useful, but effects of replacement with monounsaturated fat were unclear due to inclusion of only one small trial. This effect did not appear to alter by study duration, sex or baseline level of cardiovascular risk. Lifestyle advice to all those at risk of cardiovascular disease and to lower risk population groups should continue to include permanent reduction of dietary saturated fat and partial replacement by unsaturated fats. The ideal type of unsaturated fat is unclear. (13)
This meta-analysis only included RCTs, a type of study that is much stronger in terms of proving causality than cohort studies. The results of another meta-analysis based on RCTs, this one published in 2010 in PLoS Medicine, provides further evidence that replacement of saturated fat with unsaturated fatty acids reduces the risk of cardiovascular disease (14).
I believe the evidence as a whole clearly indicates that a high intake of saturated fat is problematic. Moreover, for me, this is the most obvious conclusion if I consider everything I know about the impact saturated fat has on human health. That said, I’m not convinced that we have all the mechanisms right. While I strongly believe a high intake of saturated fat is deleterious in terms of health and longevity, I’m not convinced that it’s the impact on cholesterol levels that is the key here.
Also, I’ll be the first to admit that when it comes to the diet-heart hypothesis, the evidence is not clear-cut. This is to be expected, as it’s difficult to adequately assess the relationship between an isolated variable such as saturated fat intake and the risk of a degenerative disease. Furthermore, as mentioned, many of the studies in this area are of the observational kind; a study type that is particularly prone to bias.
This is why I’m always reluctant to draw conclusions about diet and health just based on the results from RCTs, meta-analyses, systematic reviews, and other similar sources of data. It’s of course important to consider the results from these types of studies, but they shouldn’t be our only source of information. To really be able to say something useful about the relationship between diet and health, we also have to examine the evolutionary evidence, look at the composition and characteristics of the foods we eat, and consider the biological mechanisms.
4. Foods with a high concentration of saturated fat (e.g., ghee, cream, oils, bacon) have a low satiety index score (poor satiating capacities)
One of the main problems with highly concentrated sources of fat such as ghee, cream, and vegetable oils is that these foods are less satiating (on a calorie-by-calorie basis) than organ meats, whole coconuts, nuts, and other completely unprocessed foods (15). In general, products with a high energy density (calories per gram of food) are less filling than foods with a low energy density, partly because the latter tend to contain more water, fiber, and protein.
While the low satiety index score of the aforementioned high-fat foods isn’t a problem for a 240 pound muscular guy who’s striving to get enough energy into his body, it can certainly be an issue for someone who’s struggling to lose weight. Just imagine how much easier it is to get 300 kcal from butter (∼717 kcal/100g) than from potatoes (∼77 kcal/100g).
Online, you’ll often see claims that fat is the most satiating macronutrient. I’ll argue that this notion is not supported by the scientific evidence. I’m by no means opposed to the consumption of all high-fat foods (I actually eat a diet that is fairly high in fat myself), but I don’t think it’s a good idea for someone who wants to stay satiated and lose weight to eat a lot of butter, bacon, cream, and similar high-fat foods. If increased satiety is what you’re looking for, protein may be your best bet, because this macronutrient increases satiety and thermogenesis to a greater extent than the other macronutrients, may improve leptin sensitivity in the central nervous system, and can help boost weight loss.
5. A high intake of saturated fat can cause endotoxemia and chronic low-grade inflammation
Atherosclerosis, cardiovascular disease, obesity, and colon cancer are just some of the many chronic diseases that are characterized by elevated levels of inflammatory biomarkers in plasma. Chronic low-grade inflammation can be a consequence of chronic health problems, such as in obesity where fat tissue releases many inflammatory mediators, but it’s also clear that inflammation can be a cause of disease.
In light of all the research linking inflammation to chronic disease, the million dollar question becomes: Where does this inflammation stem from? My belief is that the gastrointestinal tract plays an essential role. Evolutionarily novel food products with an abnormal nutrient composition and/or very high concentration of fat, protein, or carbohydrate are known to contribute to the internal fire that is present in the inflamed modern man.
Some types of gram-negative bacteria contain a substance in their outer wall called Lipopolyssacharide (LPS). Pathogen-associated molecular patterns such as LPS are molecules that are commonly part of or shed by microbes and can trigger an inflammatory response in the body if they are allowed to enter systemic circulation. Chronically increased plasma LPS concentrations (endotoxemia) can initiate low-grade chronic inflammation and obesity, diabetes, and cardiovascular disease (16, 17, 18).
Since chylomicrons (fat transporters) can promote intestinal absorption of lipopolysaccharides, it’s no surprise that the amount of fat you eat can have a significant impact on the levels of LPS that are circulating in your blood (19). LPS can also enter systemic circulation by other means, like when bacterial imbalances in the gut have allowed the gut wall to become leaky.
It could be that that the increased lipid concentration in the intestine following a high-fat meal changes the environment in such a way that LPS-containing bacteria are allowed to flourish. This alteration of the microbiota may then set the stage for decreased expression of tight junctions proteins and increased translocation of substances like LPS.
Let’s have a look at what the studies in this area have found…
- Consumption of saturated fat impairs the anti-inflammatory properties of high-density lipoproteins and endothelial function (20).
- A high-fat meal can induce low-grade endotoxemia and inflammation (21, 22, 23).
- Saturated fats may play an important role in the pathogenesis of postprandial inflammation (22, 24, 25).
- Red wine, prebiotics, and orange juice may help prevent the postprandial inflammation that sometimes accompany a high-fat meal (26, 27, 28).
Here’s what a 2015 review paper had to say about the relationship between saturated fat and chronic inflammation:
Toll-like receptors (TLR) mediate infection-induced inflammation and sterile inflammation by endogenous molecules. Among the TLR family, TLR4 is the best understood. However, while its downstream signaling pathways have been well defined, not all ligands of TLR4 are currently known. Current evidence suggests that saturated fatty acids (SFA) act as non-microbial TLR4 agonists, and trigger its inflammatory response. Thus, our present review provides a new perspective on the potential mechanism by which SFAs could modulate TLR4-induced inflammatory responses: (1) SFAs can be recognized by CD14-TLR4-MD2 complex and trigger inflammatory pathways, similar to lipopolysaccharide (LPS). (2) SFAs lead to modification of gut microbiota with an overproduction of LPS after a high-fat intake, enhancing this natural TLR4 ligand. (3) In addition, this metabolic endotoxemia leads to an oxidative stress thereby producing atherogenic lipids – oxLDL and oxidized phospholipids – which trigger CD36-TLR4-TLR6 inflammatory response. (4) Also, the high SFA consumption increases the lipemia and the mmLDL and oxLDL formation through oxidative modifications of LDL. The mmLDL, unlike oxLDL, is involved in activation of the CD14-TLR4-MD2 inflammatory pathway. Those molecules can induce TLR4 inflammatory response by MyD88-dependent and/or MyD88-independent pathways that, in turn, promotes the expression of proinflammatory transcript factors such as factor nuclear kappa B (NF-κB), which plays a crucial role in the induction of inflammatory mediators (cytokines, chemokines, or costimulatory molecules) implicated in the development and progression of many chronic diseases. (29)
In general, I think it’s safe to say that if your goal is to combat inflammation and achieve a long, healthy lifespan, it’s not a good idea to eat a diet that is very high in saturated fat. That said, I want to make it clear that it’s important to be cautious when interpreting the scientific data in this area. Most (but not all) of the studies I included in this post are done in humans; however, you can find a lot of animal studies on this topic as well. One of the reasons we shouldn’t use these studies as our primary source of information is that the animals may be given diets and meals that differ markedly from what a free-living human would ever consume. Oftentimes, the researchers will use a high-fat diet that is very low in plant foods and contains extremely high concentrations of just one or two sources of fat, largely because this makes it easier to actually detect an effect.
It’s possible that this response to a high-fat meal is a mechanism we have evolved because it allowed our ancestors to store more fat in times of food abundance. The mechanism being that endotoxemia may set the stage for insulin and leptin resistance and increased fat storage. When our hunter-gatherer ancestors killed large game and suddenly had the opportunity to gorge on fatty organ meats, the postprandial enxotoxemia that followed could have allowed them to store more fat for scarcer times.
But what does all of this mean exactly? Should we stop eating fat altogether to reduce the absorption of endotoxins? Of course not. Lipids have been an important part of the human diet for hundreds of thousands of years, and there’s no reason to believe that we suddenly aren’t adapted to digest foods that are high in fat. If we simply look at the studies it might seem that all low-carb dieters have elevated levels of inflammatory substances in their blood, and while some probably do, there’s clearly something else going on here.
As I see it, the problem isn’t the fat per se, but rather the extremely high concentration of saturated fat (and to a lesser extent, unsaturated fats) in some of our modern foods. When you consume a meal that is high in for example bacon and butter you may get a fairly rapid rise in the intraluminal concentration of saturated fat in the small intestine. This rise may initiate the aforementioned inflammatory cascade, particularly if it’s combined with a lack of prebiotic fiber and a gut microbiota that has been damaged by for example antibiotics or long-term consumption of processed food.
6. Foods that are high in saturated fat have a relatively poor nutrient profile
So far, I’ve largely focused on the saturated fat component of high-fat foods and the problems associated with eating saturated fat-rich food products. What we have to remember, though, is that we’re not eating fat – we’re eating food. To be able to say anything meaningful about the relationship between saturated fat consumption and human health, we can’t just look at the amount of saturated fat we’re eating, we also have to look at the overall nutrient composition and characteristics of the foods we derive these fatty acids from.
As I’ve pointed out previously, there’s no reason to fear grass-fed meats, coconut, eggs, and other Paleo-approved whole foods that contain saturated fat and/or cholesterol. However, when it comes to evolutionarily novel foods such as ghee, cream, oils, bacon, etc., a more restricted approach is warranted, in part because these foods have an unhealthful nutrient profile when compared with foods such as fruits, vegetables, unprocessed meat, and nuts. Perhaps most importantly, they are lower in fiber, water, protein, omega-3 fatty acids, and most vitamins and minerals (on a calorie-by-calorie basis).
7. Foods that are high in saturated fat tend to have a very high energy density
Highly concentrated sources of fat such as ghee and butter have a higher caloric density than any of the foods our preagricultural ancestors ate. Organ meats and fatty fish, which were some of the richest sources of calories in the diet of our Paleolithic ancestors, are lightweights in comparison to ghee, butter, vegetable oils, and similar products introduced after the Agricultural Revolution. Since very fatty foods are so high in calories, they take up a lot of space that could otherwise have been filled by calories from foods that contain more fiber, protein, and other nutrients that a lot of people could benefit from eating more of.
Contrary to what you might have heard from a proponent of very low-carbohydrate diets (VLCDs), you can’t stuff yourself with as much fat as you want as long as you limit your carbohydrate intake. Ketogenic diets and VLCDs have been shown to be effective for weight loss, but they are not be the best choice from a health standpoint (at least over the long term), as they are typically high in butter, bacon, ghee, oils, etc. and low in plant foods and fiber.
If you just skimmed this article or read the headings, you may have gotten the impression that you would be best off removing all foods that contain saturated fat from your diet. This is – as you know if you read the whole article and/or follow this blog – not the case. Saturated fat is a natural component of healthy, whole foods such as grass-fed muscle and organ meat, organic eggs, and coconuts. I would argue that most people would actually benefit from eating more of these types of foods.
However, when it comes to evolutionarily novel high-fat foods, such as cream, cheese, bacon and other processed meats, and ghee, a more restricted approach is warranted. When compared with fish, unprocessed meat, avocados, and other “Paleo-approved” whole foods, these post-agricultural food products have a very high content of saturated fat, poor satiety index score, very high energy density, and poor nutrient profile. Moreover, a high intake of these foods may unfavourably impact the blood lipid profile, cause endotoxemia and chronic-low grade inflammation, and increase the risk of chronic disease.
Picture: Creative Commons pictures. Picture 1 by Larry, 2 by Frank Vassen, 3 by Isaac Wedin 4 by Charles Haynes.
Note: A previous version of this article of mine was published in Paleo Magazine, the first, and only print magazine dedicated to the Paleo lifestyle and ancestral health. You can subscribe to Paleo Magazine here!