The seed oil debate is raging all over social media, from Instagram and Facebook to LInkedIn. I see ‘health influencers’ fear-mongering with utter nonsense about the supposed evil of seed oils; pure drivel is being spewed left and right. Generally, these people probably don’t even know what a PMID (PubMed ID) is—I haven’t seen a single one supplied as evidence of their outlandish claims. I respect anyone’s right to have an opinion, but when you make big claims, I expect to see some evidence. They’re not just claiming that they prefer to cook with butter and tallow, they’re saying seed oils are poison and that tallow and butter are good for you. Madness.
The general population is being fed false information about seed oils, and it appears that diners are becoming increasingly fearful. I’ll be direct here: The current seed oil scare is based on misinformation, and I’m going to lay out evidence—lots of it—showing you why. I am also going to show where their claims fall flat and the inconsistent logic for the claims made by fear-mongers. You can switch to different oils if you choose, but my purpose here is to clear up the misinformation and to help you answer questions from members about what oil you cook with confidently.
The entire premise of the anti-seed-oil movement is based on the idea that fats in seed oils—predominately polyunsaturated fatty acids—aren’t heat stable, causing oxidation. I’ll speak to this a little bit more later, but I want to cover some other things first. Quite a bit of the data we’re going to look at will compare seed oils to saturated fat because that’s what anti-seed-oil people suggest is better, typically in the form of tallow or butter (or sometimes lard). The health outcome data on saturated fat is notoriously terrible, which I’ll share.
When we talk about seed oils, the main ones are vegetable (aka soybean), corn, canolo (or rapeseed), grapeseed, sunflower and safflower oils. This is not all-encompassing, but it covers the seed oils you are most likely to see in typical kitchens or at the store. Below is a table comparing the fatty acid profiles of the previously mentioned oils. We have data for saturated fat, mono and polyunsaturated fat and omega 3 fatty acids listed.
Before diving in, let’s take a moment to look at some general trends in these oil types. None of the seed oils have a saturated fat content above 15%. Monounsaturated fatty acid content ranges pretty widely between these oils, with canola being by far the highest at 63% and safflower being the lowest at 13%. Polyunsaturated fat content also varies quite widely, with grapeseed oil and safflower oil being the highest at 73% and 79%, respectively. Canola oil is by far the lowest PUFA-containing seed oil, at 28%. Most of the oils were very low in omega 3 fatty acids, known for its positive health benefits, with most oils being at or below 1%. Canola and soybean were the exceptions with omega 3 content of 10 and 7%, respectively. Interestingly enough, canola oil appears to be the most vilified of all the seed oils. Meanwhile, of the major ones listed, it has the highest monounsaturated fat content, the highest omega 3 content, and the lowest polyunsaturated fat content of any of the oils listed. Canola oil is also the lowest in saturated fat. Everyone agrees that omega 3s are healthy, and the entire premise of the seed oil fear-mongering is that polyunsaturated fatty acids are evil. Remember this.
Let’s talk about these fatty acids and what they are. Saturated( SFA), mono unsaturated (MUFA) and polyunsaturated( PUFA) are the three most common forms of fatty acids—basically broad classifications of lipids, aka what “fats” are made of. These types of fatty acids are differentiated by their chemical structure, with saturated fat being “saturated,” hydrogen and monounsaturated having a single carbon double bond, and poly—as the name suggests—having two or more carbon double bonds. Polyunsaturated fatty acid structures offer the possible fourth fatty acid configuration—trans fatty acids (usually referred to as simply trans fat). Trans fat is most commonly from partially hydrogenated oils, but there are naturally occurring trans fats, such as CLA. Note that partially hydrogenated oils are not used anymore. In 2015, the FDA reclassified partially hydrogenated oils as no longer being Generally Recognized as Safe(GRAS), effectively banning their use in food. That said, fully hydrogenated oils are still used; full hydrogenation converts oil into saturated fat, instead of trans fat. If you use or have creamy fryer oil or similar products, this addition of altered oil with a higher saturated fat content is what makes the oil creamy as opposed to clear.
Saturated fat has a long and well-documented history of being bad for you. Scientists and the medical community focused on cholesterol for a long time. They knew cholesterol in your blood tended to be bad well before they understood how dietary cholesterol worked. It turns out that your body doesn’t really absorb dietary cholesterol (Lecerf, Lorgeril 2011). Short of eating bowls of pure cholesterol, you can probably disregard it. Additionally, a meta-analysis of 40 studies with over 350,000 participants concluded that no statistical association of coronary artery disease existed from dietary cholesterol (Berger et al, 2015). Also in 2015, the USDA revised its dietary guidelines, notably no longer listing cholesterol as a “nutrient of concern” (Mozaffarian, Ludwig 2018). Their paper states that, while this change surprised people, it was “concordant with scientific evidence demonstrating no appreciable relationship between dietary cholesterol and serum cholesterol or clinical cardiovascular events in general populations.” This change to guidelines was made 10 years ago, and I still don’t think that many people are aware of it. The cholesterol caution was in effect for decades, and habits don’t change overnight.
Many foods high in cholesterol are also high in saturated fat, and that’s where the problem lies. Let’s be really clear here: Dietary cholesterol isn’t important, but blood levels of cholesterol, particularly LDL or “bad” cholesterol, absolutely is. Daghlas and Gill found higher LDL cholesterol associated with shorter life spans (Daghlas, Gill 2021). Another study from 2024 found dose-dependent associations of higher LDL cholesterol with coronary artery disease and all cause mortality(Yang et al 2024). Finally, a 2020 study found significant associations with long term exposure to higher LDL cholesterol levels (Domanski et al 2020). All of this is to say that increased cholesterol levels result in significantly higher risk of heart disease. That’s important in this discussion because saturated fat intake raises LDL cholesterol, directly driving increased risk of heart disease (Maki et al 2021). Yet, the people trying to scare you about seed oils want you to believe that cooking with tallow and butter—both high in saturated fat—is good for you. We’ve known better for a long time.
Let’s get into seed oils and how health outcome data compares with that of saturated fat. It’s worth nothing that in the previously referenced study by Maki et al specifically mentions PUFA as a beneficial swap for SFA in improving risk of heart disease. That, when the seed oil fear mongers insist that PUFA are bad for us and these oils—even in a study—are a standard source in a typical diet.
A 2009 analysis of 11 cohort studies with over 344,000 participants found a “significant inverse relationship” with PUFA intake instead of saturated fat in regard to risk of “coronary events.” This was with a 5% swap of energy intake from saturated fat in favor of PUFA (Jakobsen et al 2009). This is significant with the entire premise of seed oil fear-mongering being predicated on PUFA being bad and saturated fat heavy oils being advocated in their place. A 2015 study with over 126,000 participants found similar results (Li et al 2015); a 5% reduction in SFA in favor of PUFA was associated with a 25% reduction in risk of CHD. Interestingly, MUFA were associated with a 15% reduction. Also worth mentioning is that “carbohydrates from whole grains” was associated with a 9% risk reduction. I mention that because many in the anti-seed-oil crowd also claim carbohydrates of being evil. While these were cohort studies, a 2010 systematic review of randomized controlled trials also came to similar conclusions “with 10% reduced CHD risk … for each 5% energy of increased PUFA”(Mozaffarian et al 2010).
We see that PUFA are heart-healthy, but there’s more than just PUFA in seed oils, right? Sure, there are many different fatty acids, and PUFA, MUFA, and SFA are just the large classifications and don’t include some of the specific fatty acids that may be of concern. Let’s talk about a couple of those because the anti-seed-oil people make bold claims about some of them. Linoleic acid is frequently cited by our tallow and butter loving friends as being of major concern for causing inflammation. A 2014 systematic review of 15 clinical trials “failed to find any support for the ‘diet LA causes inflammation hypothesis’” (Fritsche 2014). Also worth noting, the study also mentions “Since the 1970s researchers have known that linoleic acid (LA) reduces blood cholesterol levels and lowers the risk of heart disease.” A 2012 systematic review of 15 studies concluded that “virtually no evidence is available from randomized, controlled intervention studies among healthy, non-infant human beings to show that addition of LA to the diet increases the concentration of inflammatory markers” (Johnson, Fritsche 2012). One of the reasons cited for concerns with linoleic acid is that it can convert to arachidonic acid, which they claim to be inflammatory and of major concern. Interestingly enough, one group vilifying seed oils is the carnivore diet crowd that claims that fruits and vegetables are bad for you and that humans are better off only eating meat and eggs, which happen to be the richest food sources of arachidonic acid. I don’t think I can overstate how illogical it is to claim arachidonic acid conversion is bad but then advocate a diet high in arachidonic acid. It’s also worth mentioning that one of the most popular natural bodybuilding supplements of the past 20 years is an arachidonic acid supplement from Molecular Nutrition sold by the name X-Factor. In a 2007 double-blind study with 31 resistance trained males, the study authors found that those on the arachidonic acid supplement had a lessened inflammatory response to training (Roberts et al 2007).
What about data specifically about seed oils, though? A 2019 meta-analysis of RCT’s on subjects older than 50 found that supplementation of canola oil “significantly reduced total cholesterol and LDL”(Ghobadi et al 2019). A 2015 study investigated the effects of dietary canola oil supplementation. The study authors found that canola oil reduced total and LDL cholesterol. Additionally, canola oil also improved liver enzymes and inflammation markers, even compared to olive oil (Kruse et al 2019). A 2012 randomized controlled trial involving vegetable oil compared with butter resulted in lower liver fat, lower insulin, lower total and LDL cholesterol with vegetable oil. This RCT was conducted with 67 obese test subjects with an isocaloric diet, meaning there’s no ‘gotcha’ moment where the vegetable oil group had better values simply because they were being fed less than the butter group. Additionally, the vegetable oil did not “cause any signs of inflammation or oxidative stress” (Bjermo et al 2012).
With all this outcome data in humans, where is this narrative coming from? The answer is probably from studies done on lab rats being misconstrued. To be clear, rodent studies are an important step in research; we test on rats before we do humans because it’s more ethical to kill rats than it is to put humans in danger for the sake of science. That said, studies done on rats provoke questions and not final answers. Humans are not rats, and there are too many instances of outcomes in rats not translating to humans to count. Dosages don’t usually equate because rats and humans don’t metabolize things the same way. Pathways that might be demonstrated in a rat study don’t always translate to outcomes in human studies. That said, let’s talk about a couple such examples because it is incredibly important to understand that mechanistic data in lab rats doesn’t necessarily translate to human outcomes.
Artificial sweeteners are an area where much fear mongering has been done with results from rodent studies. A 2014 study done in rats found that the consumption of artificial sweeteners disrupted gut microbiome(Suez et al 2014). However, a 2020 study in humans concluded that the use of artificial sweeteners “did not cause measurable changes in the gut microbiota … after 14 days of a realistic daily intake in healthy participants” (Ahmad et al 2020). This jives with similar fear-mongering about the safety of aspartame in general, again with rat studies being taken out of context. Since we’re dealing with conspiracy theorists that like to insist the FDA is lying to them, I’ll reference the European Food Safety Authority. In the ESFA’s 2013 risk assessment of aspartame, they concluded that aspartame is safe (EFSA explains the Safety of Aspartame).
One thing about seed oils that gets mentioned is chemical solvents used to extract the oil from the seed. With that, let’s talk about hexane. Hexane is derived from crude oil, and besides being used in the production of seed oils, it’s also used in adhesives. Sounds like something nasty we don’t want in our food, right? Dosage and context matters. We add fluoride to our drinking water for improved oral health, but fluoride is also used in insecticides. Water, even while essential for life and survival, can be toxic at a high enough dosage. With hexane in seed oils, the residual hexane in the oil after processing is no more than .8 parts per million- often lower. That means if you drank 2.2 lbs. of canola oil, you would get .8 mg. of hexane. Most toxicology data, such as from the EPA, referenced for hexane specifically discusses risk of inhalation. Hexane evaporates at room temperature because of how volatile it is, so this is the most likely way you would be subjected to a toxic dose, and that would most likely be in an industrial workplace using large amounts of concentrated hexane. That said, I was able to find a toxicology summary for hexane from the Minnesota Department of Health from 2022. Based on rodent data, they gave a human equivalent toxic dose of 188 mg. per kilogram of bodyweight per day (Toxicological Summary for: n-Hexane). That means a 150 lb. human being would need to ingest 12,760 mg. of hexane per day to reach levels of toxicity. In the interest of safety, let’s adjust that dose by a factor of 100, which is to say that we don’t want to exceed 1% of that risk. That puts it at 127 mg. of hexane per day for a 150 lb. human being. Remember when I said that a kilogram of seed oil would contain .8 mg.? This hexane toxicity fear mongering is a complete non-starter.
So, saturated fat isn’t good for you, polyunsaturated fat is associated with improved health outcomes, human data doesn’t show cause for concern with inflammation from seed oils, and the amount of hexane in seed oils is not nearly high enough to be of any concern. Where is all the madness coming from, then? From what I can tell, it’s largely driven by conspiracy theorists. A pair of 2021 studies conducted by faculty at Paris Nanterre University found that the lower a participant’s critical thinking ability was, the more likely they were to believe in conspiracy theories (Lantian et al 2021).
One of the claims is that we’re all victim to seed oil industry propaganda. Let’s investigate that. Follow the money, wherever it may lead. Statistics released by the Observatory of Economic Complexity for 2023 listed seed oils as having a world trade value of $19 billion. Seed oils are clearly a large industry but what about the beef industry? Per Grand View Research, in 2023 the beef industry was valued at $526.5 billion. The beef industry is over 27 times larger than the seed oil industry.
I don’t think there ever was a real debate here. As I said in the beginning, this was all built on pretty faulty logic. That said, I’m pretty sure I just wrecked the anti-seed oil argument in devastating fashion. Now that we’re beyond data and logical arguments, the big question is, what is my actual position and recommendation?
First of all, let me say that nobody pro-seed-oil is recommending you just drink the stuff. Also, I wouldn’t necessarily say I, or anyone, is pro-seed-oil per se, just that there isn’t any human data we can point to right now to claim seed oil is dangerous, and the evidence we do have seems to show the opposite. All anyone with any credibility has really said is that, within the confines of a reasonably balanced diet, cooking like an averagely mature and intelligent adult, sensible consumption of seed oils as a part of that diet are of little or no concern.
My personal recommendation is that, whenever possible, olive oil is probably most preferable from a health standpoint. A 2024 study found improved fat loss with a higher MUFA diet, despite not reducing calories vs the control group (Lopes et al 2024). Additionally, a 2023 narrative review of the Mediterranean Diet suggested olive oil as being the primary driver of the diet’s benefits, specifically mentioning the polyphenol content of olive oil and not MUFA content (Flynn et al 2023). When the strong flavor of olive oil isn’t desired or you need a higher smoke point, I find canola oil to be a great option. Butter tastes the best for certain things. Overall, though, I say it’s advisable to minimize added oil in your cooking to the extent you can. It’s generally extra calories you don’t really need.
