The seed oil swap
Beef tallow sales are up 275% in three years. The trials that fed people more linoleic acid and measured inflammation directly found nothing to fix.
By Max van Strydonck
Founder of VERTEXPublished
Sunflower oil has become the thing serious people remove from their kitchen. Restaurant chains advertise tallow fryers, and US retail sales of products made with beef tallow reached about $1.1bn in the year to March 2026 — up roughly 275 per cent on three years earlier, according to SPINS retail data.
The claim underneath the swap is specific, and it is testable. Seed oils are high in linoleic acid. Linoleic acid is the raw material for arachidonic acid. Arachidonic acid feeds inflammation, and inflammation is what stands between you and recovering from the session you just did.
The middle of that chain has been measured directly, twice over. It does not behave the way the claim needs it to.
12–17 g
Linoleic acid the average US adult eats a day — around 6% of energy
None
Effect on blood inflammatory markers across 30 randomised trials
Where the fear comes from
The biochemistry is not invented. Linoleic acid is an omega-6 fatty acid, and it is the precursor your body elongates into arachidonic acid. Arachidonic acid is the substrate for the eicosanoids — prostaglandins, thromboxanes, leukotrienes — a family that includes some of the most powerfully pro-inflammatory signalling molecules you make. Sunflower, soybean, corn and rapeseed oil are the richest ordinary sources of linoleic acid in the food supply, and the amount of them in the diet has risen sharply over the last century.
Every link in that chain is real. The argument fails on quantity rather than on chemistry.
That intake sits inside the 5 to 10 per cent of energy reference range used in dietary guidance rather than beyond it. The exposure is ordinary. The question is what it does.
The step that does not happen
For dietary linoleic acid to raise inflammation by that route, eating more of it has to put more arachidonic acid in your tissues. That has been checked on its own, and it does not.
Across the human trials, changing linoleic acid intake in adults eating Western-type diets leaves arachidonic acid content in plasma, serum and red blood cells where it was. The conversion is tightly rate-limited. Feeding more raw material into the front of the pathway does not push more product out of the back.
"Linoleic acid is the precursor of arachidonic acid, so eating more of it makes you more inflamed."
Not in humans
What happens when you measure it directly
You do not have to settle the mechanism, because the outcome has been measured without it. A 2012 review of fifteen randomised controlled trials in healthy adults looked across C-reactive protein, fibrinogen, plasminogen activator inhibitor-1, the cytokines, the soluble adhesion molecules and tumour necrosis factor-α. Not one of the trials returned a significant effect of added linoleic acid on any of them.
A later meta-analysis of thirty trials and 1,377 participants got the same answer with numbers attached to it.
Higher versus lower linoleic acid intake
Every pooled effect crosses zero.
Standardised mean differences from the pooled trials, for the cytokine and adipokine outcomes reported. An interval spanning zero means the trials could not separate the effect from nothing at all.
−0.01
Tumour necrosis factor-α
95% CI −0.19 to 0.17
0.11
Interleukin-6
95% CI −0.07 to 0.29
0.14
Monocyte chemoattractant protein-1
95% CI −0.33 to 0.60
0.17
Adiponectin
95% CI −0.17 to 0.50
The literature was reviewed again this January, in Nutrition Today, by Petersen, Messina and Flickinger. It lands in the same place: the clinical trial evidence does not show linoleic acid promoting inflammation or oxidative stress.
What the swap actually changes
That does not make the swap neutral. It changes something — just not the thing it is sold on.
Replacing saturated fat with linoleic acid lowers LDL cholesterol. Across forty randomised controlled trials and 2,175 participants, the average reduction is 3.26 mg/dL, or about 0.08 mmol/L, and it is clearest where linoleic acid displaces saturated fat specifically. On its own that is a modest number. It points in a direction, though, and that direction has been tested at the level of events rather than markers.
17%
The reduction in combined cardiovascular events when saturated fat intake is cut for two years or more — across fifteen trials and more than 56,000 people. The benefit tracked what replaced it, polyunsaturated fat or starch. Sunflower oil to tallow runs that exchange backwards.
Where the sceptics have a point
The case is not airtight, and the honest version says so out loud.
Two of the trials that tested this most directly were lost for decades, then recovered from old records and published in full. The Minnesota Coronary Experiment ran from 1968 to 1973 and made the swap at real scale — saturated fat down from 18.5 to 9.2 per cent of calories, linoleic acid up from 3.4 to 13.2. Serum cholesterol fell as predicted. Deaths from coronary heart disease and from all causes did not. The Sydney Diet Heart Study found higher coronary and all-cause mortality in the group that made the swap, though its intervention used safflower margarine of the era, and whether that carried meaningful trans fat is still argued over.
So hold both of these at once. Higher circulating linoleic acid tracks lower risk of major cardiovascular events across thirty pooled cohorts, and modestly lower all-cause, cardiovascular and cancer mortality across forty-four cohorts and 811,069 people. And the randomised mortality evidence is thinner and messier than that picture implies. Anyone certain in either direction is out ahead of the data.
The argument, sorted by evidence
Not every claim in this fight is equally supported.
Strength of evidence ↑ · Claim →
Well supported
- Linoleic acid lowers LDL when it replaces saturated fat
- Higher blood linoleic acid tracks lower cardiovascular risk
- Cutting saturated fat cuts cardiovascular events
Not supported
- Eating linoleic acid raises tissue arachidonic acid
- Linoleic acid raises inflammatory markers
- Your cooking oil is why you are not recovering
Genuinely unsettled
- Whether the swap lowers death rates
- What the recovered 1960s trials mean
- What a century of rising intake has done
Where the attention belongs
- Total energy intake
- Protein and food volume
- How much of it arrives pre-made
The variable that is actually moving
Here is what the swap obscures. Linoleic acid in an ordinary diet does not mostly arrive by the spoonful from a bottle at home. It arrives inside crisps, biscuits, ready meals, takeaways and fried food — which is precisely why intake of it tracks intake of ultra-processed food so closely. When a study finds that people eating more seed oil are less healthy, the oil and the packet it came in cannot be pulled apart.
The packet has been tested on its own. Twenty adults were admitted to a metabolic ward and fed, in turn, an ultra-processed diet and an unprocessed one, matched for the calories presented and for sugar, fat, sodium, fibre and macronutrients. Allowed to eat as much as they wanted, they ate about 500 calories a day more on the ultra-processed diet. They gained weight on it and lost weight on the other.
+500 kcal
The daily difference in what people chose to eat on an ultra-processed diet versus an unprocessed one, with both matched for calories presented, sugar, fat, sodium, fibre and macronutrients. Nobody in that ward changed a cooking oil.
For anyone training, that is the whole game and the oil is not in it. Five hundred calories a day is the difference between a lean gaining phase and a sloppy one, or between a deficit that works and one that quietly does not. Frying in tallow does not recover any of it — tallow is close to half saturated fat and carries the same nine calories per gram as the oil it replaced.
The short version
The mechanism behind the seed oil panic needs dietary linoleic acid to raise tissue arachidonic acid, and in adults eating Western diets it does not. The trials that ignored the mechanism and measured inflammation directly found nothing to report, across thirty trials and the markers they pooled.
What the swap does change is blood lipids, in the direction with the most trial evidence pointing against it. Whether that translates into fewer deaths is the part the literature genuinely has not settled, and it is worth saying so rather than picking a side.
If you want to change how you eat, change the food rather than the fat you fry it in. The ultra-processed version of a meal costs around five hundred calories a day more than the version you cooked. The oil in the pan was never the variable.