Key Takeaways

Human evidence does not support treating ordinary seed-oil intake as a unique toxin. The clearer dietary question is what the oil replaces and whether it arrives in a mostly minimally processed or ultra-processed eating pattern.

Linoleic acid is stored in body fat, but storage and a long tissue half-life do not establish harm. [1]
A Cochrane review found that increasing omega-6 fat may have little or no effect on death or cardiovascular events and may reduce myocardial infarction risk. [2]
Two recovered diet trials raised legitimate questions, though their populations, food products and linoleic-acid doses limit application to ordinary modern diets. [3] [4]
A 2025 cohort analysis associated replacing 10 grams of butter per day with plant oil with 17% lower total mortality. The substitution was modeled, not assigned. [5]
Coconut oil raised LDL cholesterol by about 10 mg/dL compared with non-tropical vegetable oils in short trials that did not measure heart attacks or deaths. [6]

What is stable

Replacing saturated fat with polyunsaturated plant oils lowers LDL cholesterol. [2]
The human evidence does not support using cell storage or oxidation mechanisms alone to infer clinical harm from usual seed-oil intake. [1] [2]
Butter, coconut oil and ghee supply more saturated fat than common liquid plant oils. [6]

What remains uncertain

Randomized evidence for long-term mortality effects of increasing omega-6 intake remains limited. [2] [3] [4]
Cohort substitution estimates may still reflect differences between people who choose butter and those who choose plant oils. [5]
The health effect of a cooking oil cannot be separated cleanly from dose, heating, the food carrying it and the rest of the diet. [2] [5]

Common questions

Are seed oils toxic at ordinary dietary amounts?

Current human evidence does not establish a unique toxic effect at usual intake. Mechanistic concerns deserve testing, but they do not replace trials or clinical outcomes. [1] [2]

Do seed oils cause inflammation?

The clinical evidence reviewed here does not show that usual linoleic-acid intake creates a consistent inflammatory disease signal. Effects depend on the comparison food and the outcome measured. [2]

What did the Minnesota Coronary Experiment show?

Corn-oil substitution lowered cholesterol but did not reduce mortality in the recovered analysis. Short exposure for many participants, institutional turnover and an unusually high linoleic-acid target limit the inference. [3]

Why does the Sydney trial remain controversial?

Men with recent coronary disease assigned to safflower oil and margarine had more deaths. The intervention changed several foods, the original protocol was unavailable and the margarine composition was uncertain. [4]

Is coconut oil healthier than other plant oils?

Short feeding trials found higher LDL cholesterol with coconut oil than with non-tropical vegetable oils. They did not test whether that difference changed cardiovascular events or survival. [6]

Does olive oil outperform every other plant oil?

Olive oil has favorable cohort and trial evidence, but the 2025 butter-and-oils cohort did not establish that it is superior to every other plant oil. Choice still depends on use, taste and the food pattern around it. [5]

Seed Oils, Butter, and Plant Oils: What Human Evidence Supports