Seed Oils: Health Hazard or Internet Hysteria?

Not every concern about cooking oils is nonsense

That does not mean every concern about cooking oils is nonsense. Polyunsaturated oils can oxidise when repeatedly exposed to very high temperatures. Reusing fryer oil over long periods is not the same thing as adding some rapeseed oil to a frying pan at home.

Oils are also extremely energy dense, so consuming large amounts can contribute to excess calorie intake regardless of whether the oil comes from a sunflower, an olive or a cow. And many foods containing seed oils are ultra-processed foods that are perfectly reasonable to limit.

” So what does the evidence actually show?

Are seed oils bad for you?

Randomised controlled trials examining linoleic acid — the main omega-6 fatty acid in many seed oils — generally do not show increases in inflammatory markers (Su et al., 2017; Johnson and Fritsche, 2012).

Large prospective analyses have associated higher linoleic-acid exposure with lower, rather than higher, rates of cardiovascular disease, type 2 diabetes and premature death (Li et al., 2020. Marklund et al., 2019; Wu et al., 2017).

Replacing foods high in saturated fat with polyunsaturated fats also tends to reduce LDL cholesterol and can reduce cardiovascular events (Mozaffarian, Micha and Wallace, 2010; Hooper et al., 2020).

That does not mean every use of every oil is equally desirable. Repeated or prolonged heating at high temperatures accelerates oxidation and increases degradation products (Bhat et al., 2022).

” It is: there is little evidence that seed oils are inherently toxic when consumed normally, while repeatedly overheating cooking oils is a separate issue (Nagra et al., 2026). A 2024 perspective on commonly consumed plant oils reached a similar view (Petersen et al., 2024).

What actually counts as a seed oil?

Common seed oils include:

• rapeseed oil, also called canola oil
• sunflower oil
• soybean oil
• corn oil
• safflower oil
• grapeseed oil
• cottonseed oil
• rice bran oil

Olive oil and avocado oil are generally described as fruit oils. But even within the seed-oil category, fatty-acid composition differs considerably.

Traditional sunflower oil is relatively high in the omega-6 polyunsaturated fatty acid linoleic acid. High-oleic sunflower oil contains much more monounsaturated oleic acid. Rapeseed oil has a different profile again and also provides some alpha-linolenic acid, an omega-3 fatty acid.

Treating all seed oils as metabolically identical is therefore an oversimplification.

Why have seed oils become so controversial?

Most of the argument centres around a few ideas. Many seed oils contain a lot of linoleic acid. Linoleic acid is an omega-6 polyunsaturated fatty acid. Polyunsaturated fats contain multiple double bonds, making them more susceptible to oxidation than highly saturated fats under certain conditions.

Linoleic acid can also participate in metabolic pathways involving arachidonic acid, from which numerous signalling compounds can be produced.

From those facts comes a much bigger claim: seed oils → omega-6 → arachidonic acid → inflammation → chronic disease.

The first parts describe genuine biochemistry. The final conclusion does not automatically follow. That is why human evidence matters.

What is linoleic acid?

Linoleic acid is an omega-6 polyunsaturated fatty acid. It is also an essential fatty acid, meaning humans must obtain it from food. It contributes to normal cell membranes and can be metabolised into a range of biologically active compounds. Many seed oils are major dietary sources.

The controversy largely comes from the fact that linoleic acid can contribute to arachidonic-acid metabolism. Arachidonic acid can then be used to make signalling molecules known as eicosanoids. Some participate in inflammatory processes.

That sounds like a convincing mechanism for saying more linoleic acid must mean more inflammation. But human metabolism is more tightly regulated than that simple pathway suggests.

Do seed oils cause inflammation?

This is probably the most common claim made against seed oils. It is also one of the easiest to test. If increasing linoleic acid causes meaningful systemic inflammation, people consuming more linoleic acid should show higher inflammatory markers.

A 2017 systematic review and meta-analysis included 30 randomised controlled studies involving 1,377 people (Su et al., 2017). Increasing dietary linoleic acid did not significantly increase:

• tumour necrosis factor
• interleukin-6
• C-reactive protein
• adiponectin
• monocyte chemoattractant protein-1
• fibrinogen
• several adhesion molecules

The authors concluded that increasing dietary linoleic acid did not significantly affect blood concentrations of inflammatory markers overall. An earlier systematic review of 15 randomised controlled trials reached much the same conclusion (Johnson and Fritsche, 2012).

That does not prove every possible intake under every circumstance is harmless. But it directly contradicts the stronger claim that ordinary linoleic-acid-rich seed oils automatically cause chronic inflammation.

But doesn't linoleic acid turn into inflammatory arachidonic acid?

Some linoleic acid can contribute to arachidonic-acid metabolism. The mistake is assuming that conversion occurs without biological regulation and that all arachidonic-acid metabolites are harmful.

Arachidonic acid contributes to inflammatory signalling, but it also contributes to compounds involved in normal physiological regulation and resolution processes.

Human physiology therefore does not divide neatly into “omega-3 = good” and “omega-6 = inflammatory”. That is far too simplistic.

What about the omega-6 to omega-3 ratio?

The omega-6:omega-3 ratio is often blamed for modern inflammation. But ratios can be misleading. A high ratio might mean someone eats a lot of omega-6. Or it might mean they consume very little omega-3. Those are not the same thing.

For most people, ensuring adequate omega-3 intake is more useful than chasing a specific ratio. A 2024 review of linoleic acid and cardiometabolic health similarly argued that absolute intakes of the individual fatty acids are generally more informative than the ratio alone (Jackson et al., 2024).

Are seed oils linked to heart disease?

If seed oils were major drivers of cardiovascular disease, that relationship should appear fairly consistently in long-term human evidence. It does not.

A large pooled analysis examined omega-6 fatty-acid biomarkers across 30 prospective studies from 13 countries (Marklund et al., 2019). It included 68,659 people and 15,198 cardiovascular events. Higher linoleic-acid levels were associated with:

• lower total cardiovascular disease
• lower cardiovascular mortality
• lower risk of ischaemic stroke

Arachidonic acid was not associated with increased cardiovascular risk. Because these were observational data, they cannot prove that linoleic acid caused the lower risk. But they are difficult to reconcile with the claim that ordinary linoleic-acid exposure is strongly cardiotoxic.

Other meta-analyses of prospective cohort studies point the same way (Farvid et al., 2014), again as associations rather than proof (Sadeghi et al., 2025).

What happens when researchers examine dietary linoleic acid?

A 2020 systematic review and meta-analysis examined 38 studies covering 44 prospective cohorts (Li et al., 2020). The dietary-intake analysis included more than 811,000 people. Higher versus lower linoleic-acid intake was associated with approximately:

• 13% lower all-cause mortality
• 13% lower cardiovascular mortality
• 11% lower cancer mortality

Again, observational associations do not establish causation. But the findings provide little support for the claim that linoleic acid is inherently toxic.

What matters is what the oil replaces

One of the most important questions in nutrition is: compared with what? Adding more oil and more calories to a diet is not the same as replacing another fat source.

If someone replaces a food high in saturated fat with a polyunsaturated oil, LDL cholesterol generally falls. That matters because LDL-containing particles play a causal role in atherosclerotic cardiovascular disease.

A meta-analysis of randomised trials included 13,614 participants and 1,042 coronary events (Mozaffarian, Micha and Wallace, 2010). Increasing polyunsaturated fat in place of saturated fat was associated with a 19% reduction in coronary heart-disease events.

A later Cochrane review also found that reducing saturated fat for at least two years reduced combined cardiovascular events (Hooper et al., 2020).

This is why dietary guidelines generally recommend replacing excess saturated fat with unsaturated fats (World Health Organization, 2023), including the UK Eatwell Guide (NHS, 2022).

But haven't some trials found harm?

Yes. A balanced discussion needs to include them. Two older diet-heart trials are repeatedly cited in the seed-oil debate.

The Sydney Diet Heart Study

The Sydney Diet Heart Study involved 458 men who had already experienced a coronary event (Ramsden et al., 2013). The intervention replaced substantial amounts of saturated fat with linoleic-acid-rich safflower oil and safflower-oil margarine.

When previously unavailable data were reanalysed decades later, mortality was higher in the intervention group. All-cause mortality was approximately 17.6% versus 11.8% in controls. Cardiovascular and coronary mortality were also higher. That deserves to be taken seriously.

However, the trial was small, involved men with established cardiovascular disease, used a very large dietary change and included products whose exact fat composition complicates interpretation. Possible trans-fat exposure from margarine has also been debated. So it should neither be ignored nor treated as the final word.

The Minnesota Coronary Experiment

The Minnesota Coronary Experiment also replaced saturated fat with linoleic-acid-rich corn oil in institutionalised participants (Ramsden et al., 2016). The intervention lowered serum cholesterol. But recovered data did not show the expected reduction in mortality. Some analyses also raised the possibility that greater reductions in cholesterol were associated with worse outcomes in older participants.

Again, this is relevant evidence. But the trial had major limitations. Many participants did not remain on the controlled diet long enough for strong long-term outcome assessment. The recovered dataset was incomplete. And the institutional diet does not closely resemble normal modern dietary use of cooking oil.

The cardiovascular evidence is therefore not perfectly tidy. Nutrition research rarely is.

What do more recent trial meta-analyses say?

A systematic review and meta-analysis examining omega-6 supplementation and cardiovascular outcomes did not find a statistically significant effect on total cardiovascular events, cardiovascular mortality, coronary events or stroke (Mazidi et al., 2022).

That means we should not oversell omega-6 as a cardiovascular treatment. But it also does not support claims of major cardiovascular harm.

The fairest summary is: clinical-event trials are mixed and imperfect. The effects on LDL and other risk factors generally favour unsaturated fats when they replace saturated fat. And large prospective datasets do not show the major harm claimed online.

Do seed oils cause type 2 diabetes?

Seed oils are also frequently blamed for insulin resistance and diabetes. Again, the human evidence does not fit that simple story.

A pooled analysis combined individual-level data from 20 prospective cohort studies involving 39,740 adults (Wu et al., 2017). During more than 366,000 person-years of follow-up, 4,347 people developed type 2 diabetes. Higher biomarker levels of linoleic acid were associated with substantially lower incidence of type 2 diabetes. Arachidonic-acid biomarkers were not associated with increased diabetes risk overall.

This cannot prove that eating more linoleic acid prevents diabetes. But it gives little support to the claim that omega-6 itself is a major cause of type 2 diabetes.

Do seed oils make you fat?

There is no convincing evidence that seed oils have some unique calorie-independent ability to cause obesity. Oil can absolutely contribute to weight gain.

Dietary fat provides approximately 9 kcal per gram. That makes all oils energy dense.

If you add a large amount of oil to your diet without reducing calories elsewhere, energy intake goes up. That applies to:

• sunflower oil
• rapeseed oil
• soybean oil
• olive oil
• avocado oil
• butter
• beef tallow

So excess seed-oil intake can contribute to weight gain. That does not demonstrate a unique obesogenic property of seed oils.

Seed oils and ultra-processed foods are not the same thing

Crisps may contain sunflower oil. Biscuits may contain rapeseed oil. Takeaway food may be fried in soybean, sunflower or rapeseed oil. That does not prove seed oils are the reason those foods may be poor nutritional choices. Those foods can also contain:

• lots of calories
• refined starch
• added sugar
• large amounts of salt
• little fibre
• highly palatable combinations of fat and carbohydrate
• large portion sizes

If someone stops eating crisps, pastries, takeaway food and biscuits because they are “avoiding seed oils”, their diet may improve dramatically. But that does not prove the seed oil was the harmful component.

It may simply mean they stopped eating a lot of highly processed, energy-dense food.

What about hexane and industrial processing?

Some refined oils are extracted using solvents such as hexane. Hexane itself can certainly be toxic at sufficient exposure. But toxicology depends on dose. The use of a chemical during manufacturing does not mean dangerous quantities remain in the finished product.

Refining is designed to remove extraction solvents, and food-production standards limit residues. People who prefer mechanically pressed or cold-pressed oils can choose them. That preference is reasonable. It is not evidence that ordinary refined seed oils are poisonous.

What about oxidation?

This is one area where concern has a genuine biochemical basis. Polyunsaturated fatty acids have multiple double bonds. That makes them more susceptible to oxidation than saturated fats. Heat, oxygen, light and time can accelerate the process. Oxidation can produce compounds including aldehydes and other degradation products.

So “polyunsaturated oils can oxidise” is true. But “therefore a teaspoon of sunflower oil used once for cooking is toxic” does not follow automatically. Exposure conditions matter.

Are seed oils safe for cooking?

Normal home cooking and repeatedly running the same oil through a deep-fat fryer are very different situations.

A 2022 systematic review and meta-analysis examined 33 studies involving 21 edible oils (Bhat et al., 2022). Heating oils below approximately 200°C produced little appreciable increase in trans-fat concentrations. Between roughly 200°C and 240°C, trans fats increased progressively, particularly with prolonged heating. Other oxidation products can also increase with prolonged high-temperature exposure.

The practical message is simple: avoid repeatedly overheating and reusing cooking oils.

Repeated fryer oil deserves its own category

When oil is repeatedly heated, cooled, exposed to oxygen and contaminated with food particles, it progressively degrades. That situation is very different from briefly frying food at home with fresh oil. Repeated heating can increase lipid oxidation products and reduce oil quality.

There are legitimate mechanistic concerns about heavily degraded oils. However, much of the long-term health evidence comes from laboratory and animal research rather than strong human outcome trials.

So it is reasonable to be cautious about repeatedly used fryer oil. It is not reasonable to treat every bottle of fresh seed oil as equivalent to heavily degraded frying oil.

Do seed oils become trans fats when heated?

Small amounts of trans-fatty acids can form under sufficiently extreme heating conditions. But ordinary home frying is not equivalent to industrial hydrogenation.

The 2022 systematic review found little appreciable increase in trans fats below about 200°C. Higher temperatures and longer heating produced more. Again, the dose and exposure conditions matter.

Are seed oils worse than butter?

For cardiovascular risk, replacing unsaturated plant oils with butter is not supported by current evidence. Butter contains much more saturated fat. Plant oils such as rapeseed, sunflower and soybean oil contain much more unsaturated fat. When unsaturated fat replaces saturated fat, LDL cholesterol generally falls.

A large prospective analysis published in 2025 followed more than 221,000 adults (Zhang et al., 2025). Higher plant-oil intake was associated with lower mortality, while higher butter intake was associated with higher mortality. Replacing 10 g per day of butter with an equivalent amount of plant oil was associated with an estimated 17% lower total mortality.

Because this was observational research, it cannot prove the substitution caused the entire difference. But it certainly does not support claims that butter is clearly safer.

Is olive oil better than seed oils?

This does not need to become a competition. Extra-virgin olive oil has an excellent evidence base. It is rich in monounsaturated fat and contains polyphenols. For someone who enjoys olive oil, it is a very good choice. That does not require sunflower or rapeseed oil to be poisonous.

Different oils have different fatty-acid profiles and culinary uses. Rapeseed oil, for example, is relatively low in saturated fat, high in monounsaturated fat and supplies both omega-6 linoleic acid and some omega-3 alpha-linolenic acid. A healthy diet can contain more than one source of dietary fat.

Should you avoid seed oils?

For most healthy adults, current evidence provides little reason to eliminate seed oils completely. A more useful approach is to look at the whole diet.

• Use oils in sensible amounts.
• Get most dietary fat from unsaturated sources.
• Include omega-3-rich foods.
• Eat nuts and seeds.
• Use olive oil if you enjoy it.
• Use rapeseed or other plant oils where appropriate.
• Avoid repeatedly overheating and reusing cooking oil.

And pay more attention to the overall quality of the food than to whether one ingredient happens to come from a seed.

What the evidence does not prove

The evidence does not prove unlimited seed-oil intake is beneficial. It does not prove every seed oil is nutritionally identical. It does not remove concerns about badly degraded repeatedly heated oil. It does not make an ultra-processed food healthy simply because its fat is unsaturated. And observational associations do not prove causation.

But the evidence also does not support the much stronger claims that ordinary seed-oil consumption:

• inherently causes chronic inflammation
• causes obesity independently of calories
• causes type 2 diabetes
• is a proven major cause of cardiovascular disease
• is inherently toxic because it contains omega-6
• must be eliminated from a healthy diet

Why the seed-oil argument sounds convincing

There is a common pattern in nutrition arguments. Start with a genuine biochemical mechanism. Make that mechanism sound frightening. Then assume the mechanism proves the food causes disease in humans. Seed oils are a good example.

• Yes, linoleic acid participates in arachidonic-acid metabolism.
• Yes, some arachidonic-acid metabolites participate in inflammation.
• Yes, polyunsaturated fats can oxidise.
• Yes, repeated high-temperature heating can produce degradation products.

Every one of those statements can be true while “seed oils are toxic” is still unsupported. Mechanisms tell us what could happen. Human trials and long-term outcome studies help tell us what actually happens. We need both.

The bottom line: are seed oils bad for you?

Current human evidence does not support treating seed oils as toxic foods that everybody needs to remove from their diet.

Increasing linoleic acid has not consistently increased inflammatory markers in randomised controlled trials. Higher linoleic-acid exposure is generally associated with lower cardiovascular disease, type 2 diabetes and mortality in large prospective datasets. Replacing saturated fat with polyunsaturated fat lowers LDL cholesterol and can reduce cardiovascular events.

Older trials such as Sydney and Minnesota complicate the cardiovascular story and should not be ignored. Repeated high-temperature heating also genuinely increases oil degradation. But none of that justifies calling ordinary use of rapeseed, sunflower, soybean or similar oils toxic.

If somebody improves their diet by eating fewer takeaways, crisps, pastries and packaged snacks while “avoiding seed oils”, that may be a good thing. Just be careful about assuming the seed oil itself was the problem.

The less dramatic advice is probably the more useful advice:

• Eat mostly minimally processed foods.
• Use added fats sensibly.
• Favour unsaturated fats over large amounts of saturated fat.
• Get enough omega-3.
• And avoid repeatedly abusing the same cooking oil.