Seed oils aren’t proven harmful, though two recovered trials keep the debate alive

Key takeaways

• Linoleic acid, the omega-6 fat in seed oils, supplied about 7.6% of Americans’ calories in 2021–2023, several times what traditional diets supplied, the federal guidelines’ scientific report says².
• In recovered data from a 1966–73 Australian trial, 17.6% of men with heart disease who were told to use safflower oil died, compared with 11.8% of men given no diet advice⁵.
• A 2018 Cochrane review of 19 trials found that eating more omega-6 fat lowered cholesterol but made little or no difference to death rates⁹; the federal report, after setting aside trials with other diet changes, also found no effect on deaths².
• In a 2025 study of 221,054 U.S. adults, people who used the most plant oils had a 16% lower risk of dying over more than 30 years¹².
• In lab frying tests, oils richest in polyunsaturated fat, such as sunflower and corn, formed the most toxic aldehydes¹³¹⁴.

Between 1966 and 1973, doctors at a Sydney heart clinic gave 221 men a clear instruction. Each had recently had a coronary event, such as a heart attack. They were to cut back on animal fats and ordinary margarine and use safflower oil and safflower margarine instead⁵.

The trial’s full data then went missing⁶. When researchers recovered them and published the results in 2013, 17.6% of the men in the safflower group had died, compared with 11.8% of men who got no special diet advice⁵.

In 2025, a team from Mass General Brigham and Harvard reported a very different finding. Among 221,054 American nurses and other health professionals followed for more than 30 years, those who used the most plant oils had a 16% lower risk of dying¹².

Both studies get quoted loudly in one of natural health’s fiercest fights. Here is each side’s strongest case, and how they weigh up.

The short answer: Not proven harmful, but not fully settled

Are seed oils harmful? In the amounts most people eat, the evidence doesn’t show it. Trials show these oils lower cholesterol⁹, and large studies that follow people for years link them to fewer heart deaths¹¹.

But the critics have real points. Today’s intake is unusual by historical standards and poorly studied, the guidelines’ scientific report says². A few old trials point the wrong way⁵⁷. And heating these oils hard and often creates breakdown compounds no one should want on their fries¹³¹⁴.

Many natural-health readers distrust seed oils for a reason that has nothing to do with heart trials. They’re factory products, made with solvents, from crops like soybeans that Americans barely used as oil a century ago³. A preference for less-processed food is a sound general rule.

The question here is narrower: does the evidence show that these particular oils cause harm? Mostly no, with exceptions worth knowing.

Americans get more than twice the linoleic acid of a century ago

“Seed oils” usually means oils from soybeans, corn, canola, sunflower and safflower seeds. Their signature fat is linoleic acid, the main omega-6 fat in food. It makes up 85% to 90% of the omega-6 fat people eat¹⁰ and up to 60% of soybean oil⁴.

Researchers at the National Institutes of Health (NIH) used U.S. food supply records. Between 1909 and 1999, linoleic acid rose from about 2.8% to 7.2% of the calories available in the food supply³. Soybean oil use per person grew more than a thousandfold³⁴.

The scientific report behind the 2025–2030 Dietary Guidelines puts U.S. intake at 7.6% of calories in 2021–2023². It says traditional diets got most of their linoleic acid from whole nuts and seeds. Today soybean and corn oils alone make up about half of the added fats and oils Americans use².

The report calls modern intake “a recent and atypical nutritional phenomenon” whose effects in people are poorly understood².

Both sides accept that this change is real. They split over whether it’s a problem.

Two recovered trials gave the critics their strongest evidence

The Sydney diet heart study. The trial enrolled 458 men aged 30 to 59⁵. The safflower group aimed for 15% of calories from linoleic acid and under 10% from saturated fat, and the men were tracked for an average of 39 months⁶.

Compared with the control group, the safflower group was 62% more likely to die from any cause and 74% more likely to die of coronary heart disease⁵. When the team added Sydney to similar trials, the pooled results leaned toward more heart deaths, though the trend fell just short of statistical significance⁵.

The U.S. and Australian team that recovered the data listed funding from NIH’s alcohol research institute and the Life Insurance Medical Research Fund of Australia and New Zealand⁶.

Two caveats come from the researchers themselves. The control group got no specific diet advice, and safflower oil contains no omega-3 fat⁶. So the trial tested one high-linoleic oil, not every seed oil.

The Minnesota coronary experiment. This 1968–73 trial, also covered in our fats article, enrolled 9,423 patients in six state mental hospitals and a nursing home, where corn oil replaced butter and other saturated fats⁷. For decades its raw records sat in storage, some on magnetic tape⁷. An NIH team led by Christopher Ramsden recovered them, with help from the late lead investigator’s son and a graduate student’s old thesis⁷.

The recovered analysis covered 2,355 of those patients⁸. Cholesterol fell about 13.8% on the corn oil diet, versus 1.0% on the control diet⁸. In people over 65, each 30 mg/dL drop in cholesterol tracked with a 22% higher risk of death⁸. Autopsies found almost twice as many heart attacks in the corn oil group⁷.

The researchers urged caution. With only partial data recovered, co-first author Daisy Zamora said, it would be premature to call corn oil harmful⁷. Coverage in the cardiology news site TCTMD added that a trial run where meals were controlled and monitored may not apply to people eating at home⁸.

Pooling Minnesota with four similar trials, 10,808 patients in all, the team found no significant difference in heart deaths⁸. To critics, that means the old advice was never proven. To defenders, it means the trials don’t prove harm either.

Heat and solvents: where the chemistry worries hold up

Two lab studies show how oils rich in polyunsaturated fat break down in the fryer¹³¹⁴.

A team at the University of the Basque Country in Spain ran olive, sunflower and flaxseed oils through an industrial fryer at 190°C (374°F), for up to 40 hours¹⁴. Sunflower and flaxseed oils produced the most toxic aldehydes, and did so soonest¹⁴. Aldehydes are reactive compounds that earlier lab research had tied to cancer and brain disease. Olive oil formed less of them, and only later¹⁴.

At De Montfort University in England, researchers compared five oils, including a commercial algae oil, in pan-frying and repeated deep-frying tests¹³. Sunflower and corn oils made the most aldehydes, canola and olive oil fell in between, and the algae oil made the least¹³. When the same sunflower oil fried batch after batch of chips, the chips soaked up rising levels of aldehydes; chips fried in olive or algae oil had little or none¹³. Fast-food chips from seven restaurants held aldehydes too¹³.

Neither study measured health in people. Guillén, the Spanish study’s first author, stressed that the amount matters and that clear safety limits still need to be set¹⁴.

Hexane is the other chemistry worry. Oil mills use this solvent to pull the last oil out of crushed seeds, because pressing alone recovers only about two-thirds of it¹⁵. A 2025 review in the journal OCL found hexane is virtually the only solvent the industry uses¹⁵.

Refining can cut leftover hexane to below 1 milligram per kilogram of oil, the review says. Canada caps it at 10 parts per million in edible oils; the only U.S. limit the review found applies to fish protein products¹⁵.

In September 2024, the European Food Safety Authority said hexane’s safety as a food-processing solvent needs a fresh review, citing unevaluated studies on the liver, kidneys and hormones¹⁵. That’s a reason for regulators to look harder, not proof that trace residues harm people who eat the oil.

The mainstream case: cholesterol, cohorts and a heart association review

In 2009, an American Heart Association (AHA) science advisory led by William Harris backed getting at least 5% to 10% of calories from omega-6 fats¹⁰. Cutting omega-6 from current levels, it concluded, would more likely raise heart risk than lower it¹⁰.

The advisory also took on the inflammation claim. In studies that followed people, higher omega-6 intake went with inflammation markers that were unchanged or lower¹⁰. The body tightly limits how much linoleic acid becomes arachidonic acid, a building block for both inflammatory and anti-inflammatory signals¹⁰. Volunteers who ate about seven times the usual arachidonic acid for seven weeks showed no change in clotting, blood fats or immune response¹⁰.

The advisory cited a pooled analysis of six trials that found 24% fewer heart events when polyunsaturated fat replaced saturated fat. It noted that the trials weren’t double-blinded and some also advised fish or cod liver oil¹⁰.

A 2018 Cochrane review pooled 19 randomized trials with 6,461 people⁹. More omega-6 fat lowered total cholesterol, and the reviewers rated that finding high certainty⁹. It made little or no difference to deaths or to overall cardiovascular events⁹.

Heart attacks may have been slightly less common, but that finding was low certainty⁹. By the reviewers’ estimate, about 53 people would need to eat more omega-6 fat to prevent one heart attack⁹.

Blood studies point the same way. A 2019 pooled analysis of 30 cohorts in 13 countries, covered in our fats article, tied higher blood linoleic acid to fewer cardiovascular deaths¹¹. People near the top of the range also had 12% fewer ischemic strokes, the kind caused by a blocked artery¹¹. Higher arachidonic acid wasn’t linked to more heart disease¹¹.

Then came the 2025 Boston study¹². People who ate the most butter had a 15% higher risk of dying than those who ate the least; those who ate the most plant oil, such as soybean, canola and olive, had a 16% lower risk¹².

In the researchers’ model, swapping 10 grams of butter a day for plant oil tracked with 17% lower death rates¹². NIH paid for the study. Its participants were mostly health professionals, and it observed people rather than assigning diets¹².

What the new federal guidelines and their science report say

The 2025–2030 Dietary Guidelines never use the words “seed oils,” “vegetable oils” or “linoleic acid”¹. They tell cooks to put oils with essential fatty acids, such as olive oil, first, and they list butter and beef tallow as other options¹. They keep saturated fat at no more than 10% of calories and suggest replacing deep-fried cooking with baking, broiling, roasting, stir-frying or grilling¹.

The scientific report goes further. It sets aside trials where other big diet changes ran alongside the oil, such as Norway’s Oslo Diet Heart Study². There, the oil group also cut hydrogenated fats, ate sardines canned in cod liver oil and was told to eat more produce and less sugar².

With such trials removed, the report finds that replacing saturated fat with linoleic acid-rich oils doesn’t reduce heart events or deaths². It rates the evidence of no effect on deaths as moderate certainty, and the evidence on heart events as very low².

The report cites the recovered Sydney data, calls cohort studies that model food swaps descriptive rather than experimental, and lists limits of blood-level studies often read as proof of protection².

Why do the AHA and the new report disagree? Largely over which trials count. The six-trial analysis the AHA relied on kept in trials that also advised more fish or cod liver oil, a limit the advisory acknowledged¹⁰. The new report leaves such trials out².

The report’s acknowledgments say its data were compiled by Christopher Ramsden, an NIH physician-researcher, with input and revisions from HHS and USDA². Ramsden was also first author of the Sydney reanalysis and led the Minnesota data recovery⁵⁷.

The omega-6 to omega-3 ratio is a weaker tool than it sounds

Many seed-oil critics focus on a ratio. In the NIH food-supply analysis, the ratio of linoleic acid to the plant omega-3 called ALA rose from about 6 to 1 in 1909 to 10 to 1 in 1999³. The authors concluded that the linoleic acid surge has likely lowered the long-chain omega-3s EPA and DHA in people’s tissues³.

The AHA advisory argued that the ratio itself is a poor target¹⁰. More omega-3 lowers heart risk, it said, but cutting omega-6 doesn’t automatically do the same¹⁰.

The practical answer serves both camps: raise your omega-3s. Americans get only about 110 milligrams a day of EPA and DHA combined, the scientific report estimates². Our fats article covers the best food sources and amounts.

Animal research adds fuel to the inflammation worry. In a 2023 University of California study, mice fed lots of soybean oil became more prone to colitis, a type of gut inflammation; an outside expert said the finding still needs testing in people⁴.

Weigh the claims

Claim	Best evidence for	Best evidence against	Our rating
Seed oils in normal amounts raise your risk of dying	Recovered Sydney data: 62% more deaths on safflower oil5; Minnesota autopsies7	No change in deaths across 19 trials9; the federal report also finds no effect on deaths2; blood-level and diet studies link the oils to lower risk1112	Evidence says no, though two recovered trials dissent
Seed oils lower cholesterol	19 trials, high certainty9	None strong; critics note lower cholesterol didn’t mean fewer deaths in Minnesota8	Strong evidence
Swapping saturated fat for seed oils prevents heart attacks	Six-trial analysis cited by the AHA, 24% fewer heart events10; a possible drop in heart attacks, low certainty9; large cohorts12	The report finds no benefit after removing trials with other diet changes2	Promising but early (mixed trial results)
Omega-6 fats inflame the human body	A mouse study4	Human markers unchanged or lower; arachidonic acid not linked to heart disease1011	Not enough evidence (human data lean against it)
Food fried in heavily reused seed oil harms the people who eat it	Aldehydes form in lab frying and turn up in fast-food chips1314	No human health outcomes measured; safe doses unknown14	Promising but early (lab work only)
Hexane residues in cooking oil harm consumers	European regulators want a new safety review15	Refining cuts residues below 1 mg/kg15	Not enough evidence
Fixing your omega-6 to omega-3 ratio is the key to health	The ratio rose over the 20th century3	Ratios hide amounts; more omega-3 matters more10	Not enough evidence

An oil-use guide that works whichever way you lean

Cooking job	Both camps can agree	If you trust the mainstream evidence	If you lean toward the critics
Dressings and no-heat uses	Use what you enjoy, fresh	Canola, soybean or olive oil	Extra-virgin olive oil, or whole nuts and seeds
Everyday sautéing and roasting	Moderate heat; stop before the oil smokes	Olive or canola oil	Olive oil, with butter for flavor inside the 10% saturated fat limit1
Deep-frying	Do it rarely and don’t keep reusing the oil	Choose oils that held up better in heat tests, such as olive oil1314	Olive oil, or skip frying for roasting or baking1

Three moves that help either way

1. Fry less, as the guidelines suggest¹, and retire frying oil early. Our cooking-methods article covers when.
2. Eat more omega-3-rich foods, such as fatty fish.
3. Cook more meals from whole foods, so you choose the fat, the amount and the heat. Restaurant fried food is the hardest to control.

Questions readers ask

The bottom line

The case against seed oils rests on real history: a sharp rise in linoleic acid, two trials whose full data stayed out of sight for decades, and solid lab chemistry about overheated oil. The case for them rests on consistent cholesterol trials and large studies that link these oils to longer lives. The pooled trials that count deaths show no clear effect either way, a point on which the Cochrane review and the federal report agree.

For the kitchen, the answer is simpler than the argument. Fry less, don’t reuse frying oil again and again, eat more omega-3s, cook from whole foods and use the oils that fit your values and budget. On that much, both camps can agree.

Related reading

• Butter made the new food pyramid, but the evidence still leans toward plant oils
• How you cook changes what you eat, so steam more and char less
• Inflammation can harm your heart, but an “anti-inflammatory” label proves little
• Why observational studies got smoking right and hormone therapy wrong
• Antidepressant trials succeeded 94% of the time in journals and 51% of the time in FDA files

Sources

1. U.S. Department of Agriculture and U.S. Department of Health and Human Services. “Dietary Guidelines for Americans, 2025–2030.” January 2026. cdn.realfood.gov
2. U.S. Department of Health and Human Services and U.S. Department of Agriculture. “The Scientific Foundation for the Dietary Guidelines for Americans, 2025–2030,” Chapter 5 (fats). 2026. cdn.realfood.gov
3. Blasbalg TL, Hibbeln JR, Ramsden CE, Majchrzak SF, Rawlings RR. “Changes in consumption of omega-3 and omega-6 fatty acids in the United States during the 20th century.” American Journal of Clinical Nutrition, 2011;93:950–962. digitalcommons.unl.edu
4. The Hospitalist. “How soybean oil could lead to gut inflammation” (on a University of California, Riverside, and UC Davis study in Gut Microbes). 2023. blogs.the-hospitalist.org
5. Ramsden CE, Zamora D, Leelarthaepin B, et al. “Use of dietary linoleic acid for secondary prevention of coronary heart disease and death: evaluation of recovered data from the Sydney Diet Heart Study and updated meta-analysis.” BMJ, 2013;346:e8707. cpc.unc.edu
6. UNC Gillings School of Global Public Health. “New look at old study casts doubt on value of vegetable fats for heart health.” February 2013. sph.unc.edu
7. University of North Carolina Health Care. News release on Ramsden CE, et al., BMJ, 2016, re-evaluation of the Minnesota Coronary Experiment. ScienceDaily, April 12, 2016. sciencedaily.com
8. Maxwell YL. “Diet-Heart Debate Turned on Its Head: New Interpretation of Old Data Questions Harm of Saturated Fat.” TCTMD, April 14, 2016. tctmd.com
9. Hooper L, Al-Khudairy L, Abdelhamid AS, et al. “Omega-6 fats for the primary and secondary prevention of cardiovascular disease.” Cochrane Database of Systematic Reviews, 2018, Issue 11, CD011094. cochrane.org
10. Harris WS, Mozaffarian D, Rimm E, et al. “Omega-6 Fatty Acids and Risk for Cardiovascular Disease: A Science Advisory From the American Heart Association.” Circulation, 2009 (AHA summary, advisory text and news release). ahajournals.org ; natap.org ; sciencedaily.com
11. Marklund M, Wu JHY, Imamura F, et al. “Biomarkers of Dietary Omega-6 Fatty Acids and Incident Cardiovascular Disease and Mortality: An Individual-Level Pooled Analysis of 30 Cohort Studies.” Circulation, 2019;139(21):2422–2436. research.wur.nl
12. Mass General Brigham. “Dietary habits study suggests more plant oils, less butter lead to better health” (on Zhang Y, et al., “Butter and plant-based oils intake and mortality in US women and men,” JAMA Internal Medicine, 2025). March 6, 2025. news.massgeneralbrigham.org
13. Moumtaz S, Percival BC, Parmar D, Grootveld KL, Jansson P, Grootveld M. “Toxic aldehyde generation in and food uptake from culinary oils during frying practices: peroxidative resistance of a monounsaturate-rich algae oil.” Scientific Reports, 2019;9:4125. nature.com
14. University of the Basque Country. “Fried food risks: Toxic aldehydes detected in reheated oil” (on Guillén MD, Uriarte PS, Food Chemistry, 2012). ScienceDaily, February 22, 2012. sciencedaily.com
15. Carré P, Berthold S, Piofczyk T, Bothe S, Hadjiali S. “Solvent solutions: comparing extraction methods for edible oils and proteins in a changing regulatory landscape. General introduction.” OCL, 2025;32:3. doi.org