After reviewing four major studies, the American Heart Association (AHA) recently issued a “presidential advisory” stating that saturated fats, including coconut oil, should be avoided, recommending that they be replaced with polyunsaturated fats like canola oil, soybean oil, and corn oil (1). A presidential advisory is a report initiated by the AHA president to “address a topic of special current importance.” While this has been all over the news, the report is really nothing new. In fact, the AHA has consistently recommended the reduction in dietary saturated fat to reduce the risk of cardiovascular disease since 1961 (2, 3).
I’ve previously written an extensive report on why saturated fat does not cause heart disease. I’m not going to rehash all of that in this article, so be sure to check it out if you haven’t. Instead, this article will zero in on the issues with this particular advisory and focus on the coconut oil claims in particular.
Cherry-Picking Studies Is Effective
There have been at least 17 systematic reviews and meta-analyses conducted in recent years that have not found a clear link between saturated fat consumption and heart disease. The authors themselves note in the introduction:
In the past few years, meta-analyses of observational studies and randomized clinical trials have come to discordant conclusions about the relationship between dietary saturated fat and risk of CVD (1).
They then proceed to pick four core studies (4, 5, 6, 7) from these meta-analyses that they deemed to be most “well performed.” This is certainly not the first time that cherry-picking has occurred in the history of saturated fat or nutrition research. You might recall the famous “seven countries study” in which Ancel Keys, father of the saturated fat–heart disease hypothesis, picked seven of 22 countries to demonstrate an initial relationship between saturated fat and cardiovascular disease (8).
Furthermore, the publication dates of these four core studies were 1969, 1970, 1968, and 1979! The authors report that no definitive large-scale clinical trials have been conducted since then, which is true, but this represents a major public health issue. If we continue to pull up the results of the same old studies year after year to shape today’s nutrition recommendations, the AHA will continue to promote 40-year-old nutritional science for the next several decades.
Worried about the AHA’s statement on saturated fat? Don’t be.
Industry Influence in Nutritional Sciences
Unfortunately, nutritional sciences are rife with industry influence, and the AHA is no exception. While the AHA is a nonprofit organization, it receives significant funds from industry leaders. Representatives from companies like Nestlé, Coca-Cola, The Sugar Association, the United Soybean Board, and the US Canola Association also serve on its “Industry Nutrition Advisory Panel.” From the AHA website:
The American Heart Association’s Industry Nutrition Advisory Panel (INAP) is a unique, strategic relationship between the American Heart Association’s Nutrition Committee and food industry leaders. In existence since 1995, INAP provides a platform for open dialogue, sharing of information and planning cooperative programs in areas of mutual interest such as diet and nutrition and cardiovascular disease (emphasis mine).
You can easily imagine how a few pushy industry leaders might be able to influence AHA recommendations.
The Difference between Statistical Significance and Clinical Significance
Furthermore, in interpreting the findings of any study used to inform human health, it’s important to distinguish between statistical significance and clinical significance (9).
In this case, the authors looked at clinical significance by assessing the number of actual cardiovascular events, but only for a handful of studies. The remainder of their report was built around changes in LDL cholesterol, which is only a risk factor for heart disease. They present several studies that observed changes in cholesterol ranging from 0.6 to 2.1 mg/dL after altering saturated fat intake. Yes, this is statistically significant, but can that tiny change in cholesterol make a difference in the number of clinical outcomes?
Consider the variability of cholesterol tests. One research group took repeated blood samples from the same people for cholesterol analysis over several days, without any intervention. They found that LDL cholesterol fluctuated more than 20 percent (about 17 mg/dL) in 95 percent of subjects and more than 40 percent in 45 percent of subjects (10). In other words, you could get your cholesterol tested today and be told you are in perfect health and get tested tomorrow and be recommended statins.
LDL-Cholesterol versus LDL-Particle Number
Additionally, the AHA report only discusses LDL cholesterol (LDL-C). One study found that of 136,905 coronary artery disease hospitalizations, almost half of patients presented with normal LDL-C (11). The number of LDL particles (LDL-P) is a much stronger predictor of cardiovascular disease risk than LDL cholesterol, and it’s possible to have normal LDL-C and high LDL-P (12, 13, 14). Preliminary studies comparing lipid profiles after subjects followed a low-carb, high-fat diet and a high-carb, low-fat diet suggest that saturated fat does not increase LDL-P (15).
Why We Can’t Rely on Animal Models to Study Lipid Metabolism
The report next turns to a number of studies suggesting that saturated fat feeding in rodent and primate models results in the development of atherosclerosis. However, there are major differences in lipid metabolism between humans and rodents, and even between humans and nonhuman primates. This is a major problem with animal nutrition research.
This book chapter provides an excellent review on the evolution of human lipid metabolism:
To accommodate the high energy demands of our large brains, humans consume diets that are of much higher quality (i.e., more dense in energy and fat) than those of our primate kin (Leonard and Robertson, 1992, 1994). On average, we consume higher levels of dietary fat than other primates (Popovich et al., 1997).
The need for an energy-rich diet also appears to have shaped our ability to detect and metabolize high-fat foods. […] compared to large-bodied apes, humans have an enhanced capacity to digest and metabolize higher fat diets. Our gastrointestinal (GI) tract, with its expanded small intestine and reduced colon, is quite different from those of chimpanzees and gorillas and is consistent with the consumption of a high-quality diet with large amounts of animal food (Milton, 1987). Finch and Stanford (2004) have recently shown that the evolution of key “meat-adaptive” genes in hominid evolution were critical to promoting enhanced lipid metabolism necessary for subsisting on diets with greater levels of animal material (16).
Simply put, we need to be very careful about translating results from animal models of lipid metabolism to humans.
Paleolithic Diet Improves Lipid Profile and CVD Risk
It’s also worth noting that field studies of 20th-century hunter–gatherers found them to be largely free of cardiovascular disease symptoms and risk factors. This is despite consuming a diet that is 28 to 58 percent fat by energy, with as much as half of this coming from saturated fat (17).
In a study published in the Journal of Nutrition, after six weeks of a ketogenic diet, fasting serum total and LDL cholesterol and oxidized LDL were unchanged, but HDL cholesterol tended to increase (21). This suggests a more favorable lipid profile, though due to small sample size and short duration, clinical significance could not be determined.
The Proven Benefits of Coconut Oil
Bringing our discussion back to coconut oil, the claims made in the AHA report simply don’t stand up to research. It appears the authors performed the following logic:
Premise: 4 core studies from the 1960s and ’70s → saturated fat causes heart disease
Premise: coconut oil has a high content of saturated fat
Conclusion: coconut oil causes heart disease and should be avoided
This logic relies on some faulty assumptions, namely that saturated fat causes heart disease (it doesn’t) and that all saturated fats are the same (they’re not).
Most importantly, studies have shown that coconut oil:
- Improves blood lipid profile (22, 23)
- Reduces weight circumference (24, 25)
- Is anti-inflammatory (26)
- Is antimicrobial (27)
- May promote weight loss (28)
- May improve antioxidant status (29, 30)
- May protect against Alzheimer’s disease (31, 32)
So, to sum up, don’t stress about eating coconut oil. The stress is probably more likely to give you heart disease!