Are Fatty Acids Bad for You, or Just Unreplaced?

Fatty acids are not bad for you as a category. What determines the health effect is what you swap them for. Industrial trans fat is the one fatty acid with no safe floor; saturated fat does not belong in that category. The randomized trials built to test this question answer it more clearly than the “saturated fat versus unsaturated fat” debate online usually lets on, and the practical answer is about substitution. Simply cutting fat without a plan for what replaces it is beside the point.

What the trials that swap one fat for another show

The strongest human evidence here comes from randomized controlled trials that fed people different fats and tracked what happened. That is a different kind of evidence than comparing what people already happened to eat.

• The Cochrane review. A 2020 update pooling 15 randomized trials in 56,675 people found that reducing saturated fat intake for at least two years lowered combined cardiovascular events by 17 percent (risk ratio 0.83).^[1] That update did not find a significant difference between replacing the lost saturated fat with polyunsaturated fat or with carbohydrate, and the data on monounsaturated fat and protein were too thin to judge. An earlier edition of this same review had reported the benefit concentrated in the polyunsaturated-fat trials specifically. The 2020 update revised that, which is worth knowing if you have seen the older claim repeated.
• The polyunsaturated-fat estimate. A separate pooled analysis of 8 randomized feeding trials in 13,614 participants found that each 5 percent of daily energy shifted from saturated fat to polyunsaturated fat was linked to a 10 percent lower risk of coronary heart disease events (risk ratio 0.90).^[2]

17% lower combined cardiovascular events after cutting saturated fat for 2+ years, across 15 RCTs PMID 32827219

Both of these are interventional trials. Researchers randomized people to different fats and then measured what happened to them, which lets the results support a causal claim about the swap itself. A correlation between what people already eat and what disease they get cannot do that on its own.

Why the observational studies on saturated fat alone look so different

If you have seen a headline claiming saturated fat was cleared, it was probably built on cohort data, and cohort data asks a narrower question than it sounds like.

A widely cited 2010 meta-analysis of 21 prospective cohort studies, covering 347,747 people, found no statistically significant association between saturated fat intake and coronary heart disease (relative risk 1.07), stroke (0.81), or cardiovascular disease overall (1.00).^[3]

That result measures saturated fat intake in isolation. It does not account for what calories replaced the saturated fat when someone ate less of it, which is exactly the variable the replacement trials above are built to test. The two bodies of evidence are not contradicting each other. They are answering different questions, and only one of them can tell you what to eat instead.

The point people miss: “saturated fat isn’t linked to heart disease” and “replacing saturated fat with polyunsaturated fat lowers heart disease risk” can both be true at once, because the first statement never specifies a replacement.

Not every fatty acid behaves the same way, either. A meta-analysis of 60 controlled feeding trials, in which researchers precisely varied the fat content of what participants ate, found that individual fatty acids move LDL and HDL cholesterol by different, reproducible amounts.^[4] Lauric acid raised total cholesterol, but mostly through HDL. Myristic and palmitic acid, the dominant saturated fats in butter and palm oil (our palm oil breakdown covers palmitic acid specifically), barely moved the total-to-HDL ratio. Stearic acid, the saturated fat that makes up much of the fat in dark chocolate and cocoa butter, reduced that ratio slightly, essentially acting as the neutral fatty acid in the saturated fat family.^[4] Lumping “saturated fat” into one bucket hides that spread.

What a high-saturated-fat meal does in the hours after you eat it

The online debate over whether a saturated-fat meal and an unsaturated-fat meal provoke different short-term inflammation points at a real and separate question from the trials above: what does a fatty meal do to your body in the hours right after you eat it, on a timescale of hours rather than years.

A randomized crossover trial gave 20 healthy adults a 760-calorie high-fat meal on one visit and an isoenergetic high-carbohydrate meal (pizza) on another, measuring blood markers including TNF-alpha, IL-6, ICAM-1 and VCAM-1. The high-fat meal raised several markers of endothelial activation in these healthy adults, while the isoenergetic high-carbohydrate meal did not, an effect blunted by 800 IU of vitamin E and 1,000 mg of vitamin C given alongside either meal. Whether the fatty acid type within a fat-matched meal, rather than the total fat load, drives this response in healthy adults is less consistent across trials, and some trials matching meals exactly for fat content have found no difference by fatty acid type at all.^[5]

A separate crossover trial in 20 healthy adults tested four meals differing only in fat type: a saturated-fat (coconut oil) meal, an omega-3 (fish oil) meal, an omega-6 (grapeseed oil) meal, and an olive oil control. The saturated-fat meal raised circulating endotoxin, a bacterial cell-wall fragment linked to low-grade inflammation. The omega-3 meal lowered it. The omega-6 meal was no different from the control.^[6] Even so, the endotoxin changes in that trial did not track with the inflammatory markers researchers measured in the same blood samples.

These are postprandial studies. They measure what a blood marker does for a few hours after one meal. What happens to your arteries over a decade is a separate question these studies cannot answer, which makes them useful for a mechanism and useless as proof that any single meal or snack changed your long-term risk.

The one fatty acid category with no safe replacement logic

Everything above is a debate about degree and substitution. Industrial trans fat is not that kind of debate.

A systematic review comparing trans fat with saturated and unsaturated fatty acids found that industrially produced trans fat tends to raise LDL cholesterol and lower HDL cholesterol at the same time, relative to unsaturated fat, which is why the review’s authors concluded intake should be reduced as much as possible.^[7] Saturated fat, by contrast, mostly just raises LDL, and which saturated fatty acid you are talking about changes even that, as the section above shows.

Regulation backs this up with measurable results. A systematic review of trans-fat reduction interventions found that outright legislative bans were the most effective policy tool tested, and Denmark’s 2004 ban drove population trans fat intake down from 4.5 grams a day in 1976 to near elimination.^[8] Country-level reformulation mandates cut intake by around 2.4 grams a day elsewhere; worksite programs and individual counseling managed roughly 1.2 and 0.8 grams a day respectively, both far weaker than simply removing the ingredient from the food supply.

Trans fat reduction strategies compared

Average daily reduction in trans fat intake achieved by intervention type.

Bottom line on trans fat: this is the one fatty acid category where “less of it” is unambiguously the right answer, which is not the same conclusion the rest of this article reaches about saturated fat.

Where the omega-6 inflammation claim and E570 fit in

Two more pieces of the “fats are scary” story deserve a direct answer, because both get repeated more confidently than the evidence supports.

The first is the claim that omega-6 fat itself drives inflammation. A systematic review of 15 randomized controlled trials testing increased dietary linoleic acid, the main omega-6 fatty acid, found no significant effect on inflammatory markers in healthy adults, including CRP, fibrinogen, cytokines, and adhesion molecules.^[9] The popular version of this claim, that eating more seed oil raises your baseline inflammation, is not what the controlled trials show.

The second is a labeling mix-up rather than a nutrition question. E570 is a food additive. It is not a nutrition claim. It shows up on ingredient lists as an emulsifier, anti-foaming agent, or glazing agent, the technical job fatty acids do in processed food, and it has nothing to do with whether the fat profile of that food is saturated, unsaturated, or anything else. Seeing “fatty acids (E570)” on a label tells you about processing, not about the fatty acid composition the rest of this article covers. The fat a packaged food is built on shows up further down the ingredient list, which is the kind of detail EatMatrix reads for you from a photo of the label.

What this changes about how you eat

Pull the trials together and the practical takeaway is narrower than “eat less fat” or “fat is fine.” It is about substitution.

The evidence that follows from the trials is about what replaces the calories you cut. It is not a verdict on fat as a nutrient category. A large prospective cohort study of 127,536 participants tested exactly this: replacing 5 percent of energy from saturated fat with polyunsaturated fat, monounsaturated fat, or whole grain carbohydrate was tied to 25 percent, 15 percent, and 9 percent lower coronary heart disease risk respectively. Replacing it with refined starches or added sugar instead showed no such benefit.^[10]

That last line matters because it is the swap people reach for without thinking: trading saturated fat for refined carbohydrate or added sugar is not the swap the trials reward, and neither is trading it for nothing in particular. If you are choosing between butter and a seed oil blend, or between a sunflower oil product and a palm oil one, the trials say the swap toward unsaturated fat is the one with evidence behind it. Swapping toward sugar is not, whatever the “fat-free” label implies.

Industrial trans fat is the one fat worth avoiding outright rather than swapping carefully. Everything else in this article is a question of what you put on the other side of the scale.

This article summarises publicly available regulatory assessments and peer-reviewed research about food ingredients. It is general information and is not intended to diagnose, treat, cure, or prevent any disease, and it is not personalised dietary or medical advice. Regulatory positions change; the EatMatrix app reflects the current assessments in our database. If you have an allergy, a medical condition, or are pregnant, talk to a qualified clinician or dietitian before changing what you eat.