ClearFactAll topics

Do high-omega-6 seed oils (vegetable oils) increase the risk of chronic inflammatory diseases compared to traditional fats?

4 peer-reviewed studies for, 3 against

Key takeaways

Stronger evidence: Against

Verdict

Whether seed oils are a major cause of chronic disease has become one of the most polarising nutrition questions of the 2020s, but the contemporary peer-reviewed evidence weighs decisively against the "seed oils cause chronic disease" thesis. The "for" side has real, careful work behind it. Ramsden and colleagues recovered participant-level data from the 1966–73 Sydney Diet Heart Study (BMJ, 2013) and showed that the intervention group given linoleic-acid-rich vegetable oil experienced higher all-cause and cardiovascular mortality than controls. Ramsden et al. (BMJ, 2016) similarly recovered data from the Minnesota Coronary Experiment and found that lowering cholesterol via linoleic acid did not translate into a mortality benefit. DiNicolantonio and O'Keefe (Open Heart, 2018) argued mechanistically that high seed-oil intake raises oxidised linoleic acid metabolites and promotes the oxidative-stress and inflammation pathways relevant to coronary heart disease.

The "against" side, however, draws on substantially larger and methodologically stronger evidence bases. Farvid and colleagues (Circulation, 2014) meta-analysed 13 prospective cohort studies covering 310,602 individuals and 12,479 CHD events and found that higher dietary linoleic acid intake was inversely associated with both CHD events and CHD mortality in a dose-response manner. Marklund and colleagues (Circulation, 2019) used objective biomarkers — circulating and tissue linoleic acid measured directly rather than estimated from food-frequency questionnaires — across 30 cohorts and 68,659 participants and found that higher in-vivo linoleic acid was associated with lower cardiovascular events and mortality. Hooper and colleagues (Cochrane, 2020) reviewed the randomised-trial evidence and found that replacing saturated fat with polyunsaturated fat reduced combined cardiovascular events by roughly 21%, the substitution that "switching from animal fats to seed oils" effects in the diet.

The synthesis: the "for" side's strongest evidence is two recovered-data RCTs from the late 1960s and early 1970s and a mechanistic critical review. The recovered-data trials are valuable historical correctives but limited by their age, single populations, the use of dietary patterns rather than seed oils per se, and the now-known peculiarities of the trans-fat-laden margarine used in the Sydney intervention arm. The mechanistic argument relies on biomarkers of oxidised LA but does not establish that population-level seed-oil intake at current levels causes the chronic-disease outcomes attributed to it. The "against" side draws on larger pooled cohort data, biomarker-based analyses that bypass dietary-recall errors, and the Cochrane synthesis of randomised trials — exactly the evidence types that the evidence-based-medicine framework treats as most reliable. The "against" side prevails because the question — are seed oils a major cause of chronic disease — is currently answered no by the bulk of contemporary peer-reviewed evidence. The honest framing is that the for-side concerns about industrial seed-oil oxidation, omega-6/omega-3 ratios, and processing are biologically plausible and worth continued research, but they do not currently overturn the consistent finding that linoleic acid intake at population levels is associated with neutral or favourable cardiovascular outcomes.

Strongest study on each side

Ranked by citation count.

Strongest study for

Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968–73) (BMJ, 2016)

Christopher E Ramsden, Daisy Zamora et al.

Ramsden and colleagues recovered raw participant-level data from the Minnesota Coronary Experiment, a large double-blind RCT of replacing dietary saturated fat with linoleic-acid-rich corn oil. The intervention group did achieve a 13.8% reduction in serum cholesterol, but Kaplan–Meier analyses showed no mortality benefit and a paradoxical increased risk of death among those with the largest cholesterol reductions (in particular, 22% higher risk per 30 mg/dL drop). The recovered data did not support the hypothesis that lowering serum cholesterol via linoleic acid translates into lower cardiovascular mortality, and the authors argued that systematic underreporting of negative trial data has biased the published evidence base toward seed oils.

Strongest study against

Reduction in saturated fat intake for cardiovascular disease (Cochrane Database of Systematic Reviews, 2020)

Lee Hooper, Nicole Martin et al. · 41 citations

Hooper, Martin, Jimoh, Kirk, Foster, and Abdelhamid produced the Cochrane systematic review of randomised controlled trials of reducing saturated fat intake, the gold-standard evidence-synthesis method in evidence-based medicine. Across 15 RCTs and roughly 56,000 participants, replacing saturated fat with polyunsaturated fat — the macronutrient substitution effected by replacing animal fats with seed oils — was associated with a 21% reduction in combined cardiovascular events (RR 0.79, 95% CI 0.66–0.93). The Cochrane authors concluded that lifestyle advice to reduce saturated fat and replace it with polyunsaturated fat reduces cardiovascular events, directly supporting seed oils as a net-beneficial substitute for saturated animal fats in the diet.

Open the interactive page →

The evidence

For (4)

Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968–73) (BMJ, 2016)

Christopher E Ramsden, Daisy Zamora et al.

Ramsden and colleagues recovered raw participant-level data from the Minnesota Coronary Experiment, a large double-blind RCT of replacing dietary saturated fat with linoleic-acid-rich corn oil. The intervention group did achieve a 13.8% reduction in serum cholesterol, but Kaplan–Meier analyses showed no mortality benefit and a paradoxical increased risk of death among those with the largest cholesterol reductions (in particular, 22% higher risk per 30 mg/dL drop). The recovered data did not support the hypothesis that lowering serum cholesterol via linoleic acid translates into lower cardiovascular mortality, and the authors argued that systematic underreporting of negative trial data has biased the published evidence base toward seed oils.

Omega-6 vegetable oils as a driver of coronary heart disease: the oxidized linoleic acid hypothesis (Open Heart, 2018)

James J DiNicolantonio and James H O’Keefe

DiNicolantonio and O'Keefe synthesised mechanistic and clinical evidence that high intake of industrial seed oils — by raising tissue concentrations of linoleic acid and its oxidation products — promotes oxidative stress, oxidised LDL, chronic low-grade inflammation, and atherosclerosis. They argued that the displacement of fats with low LA content by seed oils over the 20th century parallels the rise in coronary heart disease, and that the oxidation products of LA (OXLAMs) are detectable in atherosclerotic plaques. The review consolidated the case that seed oils are not biochemically inert substitutes for animal fats but a plausible contributor to the chronic-disease environment.

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)

C. E. Ramsden, D. Zamora et al.

Ramsden, Zamora, Leelarthaepin and colleagues recovered original participant-level data from the 1966–73 Sydney Diet Heart Study, a single-blinded RCT of replacing saturated fat with omega-6 linoleic acid (largely from safflower oil) for secondary prevention. In the recovered data, the intervention group experienced higher all-cause mortality (17.6% vs 11.8%, HR 1.62, 95% CI 1.00–2.64), higher cardiovascular mortality (HR 1.70), and higher coronary heart disease mortality (HR 1.74) than controls. An updated meta-analysis incorporating these previously suppressed data shifted pooled estimates toward harm. The authors argued that the early enthusiasm for replacing saturated fat with linoleic acid as a CHD-prevention strategy was based on incomplete evidence.

Against (3)

Reduction in saturated fat intake for cardiovascular disease (Cochrane Database of Systematic Reviews, 2020)

Lee Hooper, Nicole Martin et al. · 41 citations

Hooper, Martin, Jimoh, Kirk, Foster, and Abdelhamid produced the Cochrane systematic review of randomised controlled trials of reducing saturated fat intake, the gold-standard evidence-synthesis method in evidence-based medicine. Across 15 RCTs and roughly 56,000 participants, replacing saturated fat with polyunsaturated fat — the macronutrient substitution effected by replacing animal fats with seed oils — was associated with a 21% reduction in combined cardiovascular events (RR 0.79, 95% CI 0.66–0.93). The Cochrane authors concluded that lifestyle advice to reduce saturated fat and replace it with polyunsaturated fat reduces cardiovascular events, directly supporting seed oils as a net-beneficial substitute for saturated animal fats in the diet.

Dietary Linoleic Acid and Risk of Coronary Heart Disease: A Systematic Review and Meta-Analysis of Prospective Cohort Studies (Circulation, 2014)

Maryam S. Farvid, Ming Ding et al.

Farvid, Ding, Pan, Sun, Chiuve, Steffen, Willett, and Hu performed a systematic review and meta-analysis of 13 prospective cohort studies covering 310,602 individuals and 12,479 CHD events including 5,882 CHD deaths. They found that higher dietary linoleic acid intake was inversely associated with both total CHD events (pooled RR 0.85 highest vs lowest, 95% CI 0.78–0.92) and CHD mortality (RR 0.79), with a dose-response relationship. The findings support, rather than undermine, the public-health recommendation to replace saturated fat with polyunsaturated fat from sources including seed oils, and they directly contradict the claim that seed oil consumption is a major contributor to coronary heart disease.

Biomarkers of Dietary Omega-6 Fatty Acids and Incident Cardiovascular Disease and Mortality: An Individual-Level Pooled Analysis of 30 Cohort Studies (Circulation, 2019)

Matti Marklund, Jason H.Y. Wu et al.

Marklund, Wu, Imamura and colleagues conducted a harmonised individual-level pooled analysis of 30 prospective cohort studies across 13 countries, totalling 68,659 participants and 15,198 incident cardiovascular events. Using objective biomarkers of linoleic acid (LA) and arachidonic acid (AA) in blood and tissue rather than self-reported intake, they found that higher in-vivo LA levels were associated with lower risks of total cardiovascular events, CVD mortality, and all-cause mortality, with similar but weaker associations for AA. Because biomarkers are not subject to the recall bias of food-frequency questionnaires, this is among the strongest available evidence that omega-6 fatty acids — the dominant component of most seed oils — are protective rather than harmful at the population level.

The debate

Maya: I want to start with the largest-scale contemporary evidence. Marklund and colleagues in Circulation in 2019 ran an individual-level pooled analysis of 30 prospective cohort studies in 13 countries — 68,659 participants, over 15,000 cardiovascular events — using blood and tissue biomarkers of linoleic acid rather than self-reported diet. Higher LA was associated with lower cardiovascular events and lower mortality. Farvid and colleagues in Circulation in 2014 meta-analysed 13 cohort studies covering over 310,000 people and found dietary LA inversely associated with coronary heart disease in a dose-response manner. The signal at population scale points the same direction.

Daniel: Cohort studies are vulnerable to confounding. People who eat more seed oils also tend to follow other health-conscious behaviours — exercise more, smoke less, eat more vegetables. Even biomarker analyses inherit the confounding because the biomarker reflects diet, and diet correlates with everything. The randomised-controlled-trial evidence is the only thing that breaks confounding, and there the picture changes. Ramsden and colleagues recovered data from the Sydney Diet Heart Study and published it in the BMJ in 2013. The intervention arm given linoleic-acid-rich vegetable oil had higher all-cause mortality than the control. Their 2016 BMJ paper on the Minnesota Coronary Experiment showed the same — cholesterol fell, but mortality did not.

Maya: The recovered-data trials are valuable but they are old. The Sydney Diet Heart intervention used a margarine that was high in trans fats — that confounds the linoleic-acid finding badly. The Minnesota trial had short follow-up that did not capture later cardiovascular events. The Cochrane systematic review by Hooper and colleagues in 2020 is the contemporary RCT synthesis. They reviewed 15 randomised controlled trials and roughly 56,000 participants and found that replacing saturated fat with polyunsaturated fat reduced combined cardiovascular events by 21%. That is the gold-standard evidence-based medicine answer to the substitution question, and it is the substitution that switching from animal fats to seed oils represents.

Daniel: The Cochrane review pools heterogeneous trials and applies a substitution pattern that is theoretical rather than always achieved in the trials. And the Hooper review focuses on saturated-fat reduction, not specifically on seed oils — those are correlated but distinct questions. The fact that PUFA can replace saturated fat with apparent benefit does not establish that seed oils, as industrially processed and consumed at modern intake levels, are net beneficial. DiNicolantonio and O'Keefe in Open Heart in 2018 argued that the oxidation products of linoleic acid — OXLAMs — are atherogenic and accumulate with high seed-oil intake. That mechanism does not apply to PUFA from whole-food sources like nuts and fish.

Maya: The OXLAM hypothesis is biologically interesting but it has not been validated in the kinds of large prospective studies that test whether the mechanism actually plays out at the population level. If high LA intake meaningfully drove atherosclerosis through OXLAMs, you would expect the cohort and biomarker evidence to show harm. It does not. Farvid found inverse association with CHD mortality. Marklund found inverse association with CVD mortality and total mortality. The mechanism may be real but it is dominated by other effects in real people eating real diets.

Daniel: That is one possible interpretation. Another is that the cohort studies are picking up the broader healthy-user pattern — people who follow dietary guidelines exercise more and have healthier lifestyles overall, which masks any seed-oil specific harm. The fact that two RCTs that did test the substitution head-to-head failed to show a mortality benefit is concerning evidence in any case, even after adjusting for the trans-fat confounding in the Sydney trial.

Maya: Concerning enough to put a footnote on the population recommendation, perhaps, but not enough to overturn the dominant finding. The American Heart Association's 2017 Presidential Advisory, Cochrane 2020, the Marklund pooled biomarker analysis, and the Farvid meta-analysis all point the same way. The "seed oils cause chronic disease" framing in popular discourse does not match the contemporary peer-reviewed evidence base. It draws disproportionately on two recovered-data trials, a mechanistic critical review, and a great deal of social media.

Daniel: I will accept that the strong public-discourse claim — "seed oils are toxic and a major cause of chronic disease" — overstates what the literature supports. But I will not accept that the question is closed. There are real signals from RCT data and there are real mechanistic concerns about industrial processing, oxidation, and the modern omega-6/omega-3 ratio that deserve continued study, not dismissal. The honest position is that the bulk of evidence does not currently support the for-side claim at strength, while the for-side concerns are not nothing.

Maya: That is a position I can fully agree with. The bulk of contemporary evidence does not support seed oils as a major cause of chronic disease, but the questions raised by the recovered-data trials and the mechanistic literature are scientifically legitimate and worth continued investigation rather than reflexive dismissal. The popular framing that elevates seed oils into a primary chronic-disease driver is going beyond what the data support.

Daniel: Then we are agreeing on both: the strong popular-discourse for-side claim is not supported by the data, and there is enough residual signal in the literature that the question deserves ongoing research rather than declarative settlement. That is a reasonable place to leave it.

Frequently asked questions

Are seed oils inflammatory?

Some reviews argue that excess omega-6 intake and oxidised linoleic acid products may promote inflammation and oxidative stress. However, large cohort studies, including one using blood and tissue biomarkers, link higher linoleic acid levels to lower cardiovascular events and mortality.

Is butter or animal fat healthier than vegetable oil?

A Cochrane review of randomised trials found that replacing saturated fat with polyunsaturated fat, the swap made when animal fats are replaced with seed oils, reduced combined cardiovascular events by roughly 21%. The current evidence favours seed oils as a substitute for saturated fat.

What did the Sydney Diet Heart Study and Minnesota Coronary Experiment show?

Recovered data from these older trials found no mortality benefit from replacing saturated fat with linoleic-acid-rich oil, and the Sydney trial showed higher death rates in the intervention group. Their limitations include their age, single populations, and the trans-fat-laden margarine used in the Sydney intervention arm.

Does the omega-6 to omega-3 ratio matter?

A 2025 review suggests omega-6 fats are essential but may worsen inflammation when consumed in excess, and that the ratio to omega-3 influences chronic disease outcomes. This concern is considered plausible and worth further research, but it does not overturn the broader evidence on linoleic acid and heart health.

Related questions