The American Heart Association, U.S. Dietary Guidelines and most doctors and nutritionists say that if you eat more "healthy fats" from vegetable and seed oils and less "bad fats" from red meat and dairy products, you're on your way to better cardiovascular health.
It turns out that may not be supported by the highest standards of scientific evidence. A new analysis of never-before-published trial data from the 1960s and '70s pokes holes at the notion that we can stave off heart attack and stroke by eating more polyunsaturated fat (the "healthy" kind). Instead, it suggests that some people who eat more of this fat from vegetable and and seed oils -- specifically, those that are high in omega-6 fatty acids — actually have a higher risk of death than those who have a diet high in saturated fat.
The study's findings were never published in full, perhaps because they went against the emerging and increasingly popular hypothesis that saturated fat in foods like red meat and dairy causes cholesterol levels in blood to rise, increasing the risk of cardiovascular disease and death. This hypothesis, called the Lipid Hypothesis, is currently a generally accepted piece of conventional wisdom in the nutrition science community. With help from the son of one of the original investigators, two researchers recently unearthed and re-analyzed the previously unpublished data. What they found contradicts the Lipid Hypothesis, and could change the way we view "healthy" polyunsaturated fats.
The analysis, published Tuesday in the British medical journal BMJ, is a sequel of sorts for lead researchers Daisy Zamora of the University of North Carolina, Chapel Hill and Christopher Ramsden of the National Institutes of Health. Back in 2013, they performed a similar analysis on another set of previously unpublished data, known as the Sydney Diet Heart Study, and came to similar conclusions: Substituting saturated fat with oils high in omega-6 acids, also known as linoleic acids, actually increases the risk of death from all causes, coronary heart disease and cardiovascular disease.
"Our research highlights the paucity of evidence supporting the dietary replacement of saturated fat with vegetable oils rich in linoleic acid for heart health," Zamora told The Huffington Post.
Experts say that, combined, these two trials could upend the way we view healthy eating -- especially when it comes to our relationship with so-called “healthy” oils.
This new finding is based on data from 1968
Back in 1968, the U.S. Public Health Service and National Heart Institute funded a five-year study called the Minnesota Coronary Experiment to see whether replacing saturated fat with vegetable oil would prevent heart attacks, stroke and death. To have complete control over what study participants ate and didn’t eat, researchers Ivan Frantz and Ancel Keys -- who famously made the connection between a Mediterranean diet and better heart health -- divided over 9,000 adults in nursing homes and mental hospitals into two groups in a double blind, randomized trial.
The intervention group reduced their saturated fat intake about 50 percent compared to their normal diet, and upped their vegetable oil intake more than 280 percent in the form of corn oil. The control group was put on a diet much lower in polyunsaturated fats and higher in saturated fat -- in other words, more typical of the average American diet.
These experimental diets lasted between 41 to 56 months, depending on the institution. Then Frantz and Keys followed up with roughly 2,403 participants who stayed on their new diets for a year or more, and measured their cholesterol levels several times over an average of three years. They also conducted autopsies participants who died during the experiment to see whether there were any differences in their arterial or heart health at death.
The unexpected (and perhaps unwanted) results
Frantz and Keys found that the intervention group participants, who swapped more corn oil for saturated fat, did indeed lower the levels of cholesterol in their blood compared to the control group and baseline levels. The intervention worked as they predicted.
However, later analyses reported in a 1981 master's thesis paper by now-retired biostatistician Steven K. Broste found that the though the intervention group lowered their cholesterol levels, they didn’t lower their risk of death. In fact, participants who were over 65 when they started the experiment actually had a higher risk of death if they had been part of the group eating more vegetable oils.
The autopsies Frantz and Keys conducted were even more disquieting. Among the 149 autopsies for which there was complete information, 41 percent of the intervention group showed signs of at least one heart attack, while only 22 percent of the participants in the control group -- the high saturated fat group -- showed the same. Deceased people from the intervention group also had similar levels of fatty plaque buildup in the arteries and hardened arteries compared to the control group.
After collecting all the data from the Minnesota trial, Zamora and Ramsden combined it with four published randomized, controlled trials that also substituted saturated fat with vegetable oils high in linoleic acid. Like the Minnesota trial, this additional meta-analysis found there was no benefit to switching to these oils when it came to death by coronary heart disease or death by any cause.
"It's truly a shame that Americans are being recommended to consume these industrial seed oils, as it’s in direct contradiction to the highest level of evidence in the literature," observed James DiNicolantonio, a cardiovascular research scientists at St. Luke’s Mid America Heart Institute who wasn't involved in these studies.
For unknown reasons, Frantz and Keys never published the full results of this experiment. Their study has the distinction of being one of the very few randomized, controlled trials -- considered the highest standard of medical evidence -- to evaluate the effects of lowering saturated fat on heart health.
In their 2016 re-analysis, Zamora and Ramsden speculate that perhaps the researchers didn’t publish the findings in full because they were so “contrary” to the popular scientific beliefs of the time.
"There would have been little or no scientific or clinical trial literature at the time to support findings that were so contrary to prevailing beliefs and public policy,” they wrote. "And, finally, it is possible that medical journal reviewers would not have accepted study results for the reasons cited above."
Some vegetable oils are better for you than others
It’s important to note that Frantz and Keys’ trial was conducted when Americans were eating more corn oil and margarine made from corn oil, and that corn oil was the only polyunsaturated fat that was used in the experiment. Corn oil is high in omega-6 fatty acids, and while our bodies need some amount of omega-6 fatty acids to function properly, Zamora and Ramsden note that too much of it could have a “wide range of biochemical consequences” that could perhaps be negative. This nuance isn't reflected in the most up-to-date nutritional advice in the federal dietary guidelines, which state simply we should eat less saturated fat and more polyunsaturated fat without mention of omega-6 acids.
Rebecca Blake, a registered dietitian and administrative director for medicine at Mt. Sinai-Beth Israel, put it this way: “We don’t usually consider corn oil the gold standard of healthy oils.
“It’s fine -- it’s a vegetable oil,” she continued. “But it’s not olive oil, and it’s not like having a diet that’s high in healthy fats like fatty fish.”
Some other sources of polyunsaturated fats, like fatty fish, contain both omega-6 and omega-3 fatty acids. It’s this combination that could be the key to protecting people’s hearts, rather than just consuming polyunsaturated fats as a class, notes Tom Brenna, a professor of human nutrition at Cornell University.
“This gives us strong reason to think carefully about the composition of polyunsaturated fatty acids,” said Brenna. “The analysis does not say that polyunsaturated fatty acids in general are a bad substitute, but that they’re seeing negative things exclusively with omega-6 fatty acids."
He also expressed regret that nutrition experts have not been able to effectively communicate this to the general public. Brenna, who served on the committee of scientists who helped create the 2015-2020 U.S. Dietary Guidelines, says it isn't as simple as rattling off the names of certain plant oils. Different types of oils have different fatty acid profiles, which all act on your body in different ways.
"You can no longer just say eat one kind of oil or another kind of oil," he said. "You have to specify which one it is, and we have not, as a society, figured out how to talk about that yet."
For the record, the only oils that Brenna can wholeheartedly recommend are olive oil, avocado oil and high-oleic sunflower oil.
“It gets so confusing for people, but it shouldn’t be,” Brenna concluded. “We should figure out a way to deal with this behind the scenes, so that people don’t have to become chemists in order to eat."
CORRECTION: A previous version of this story incorrectly stated that the researchers added data from an additional five randomized controlled trials. In fact, they added data from four trials. Also, a portion of a quote from an outside researcher that appeared to misinterpret Zamora and Ramsden's analysis has been removed.