The Ingredients Nobody Explains to You.
Refined oil has been extracted from the seed or fruit and then deodorized at high heat so that it tastes neutral and keeps. That last step has a named problem. The European Food Safety Authority found that heating oil to about 200°C during refining forms contaminants called glycidyl esters and 3-MCPD esters, and that most refined vegetable oils carry them, palm oil most of all. This entry is about those refining by-products. It makes no claim about omega-6 fat: the American Heart Association (2017) recommends unsaturated fat in place of saturated fat, and the World Health Organization (2023) says the fat people eat should be primarily unsaturated.
These are the foods the European Food Safety Authority measured. A label names the plant, not the process. European limits for these contaminants cover vegetable oils “except virgin olive oils.”
What forms in the refinery. When vegetable oil is deodorized at about 200°C, two families of contaminants form: glycidyl esters and 3-MCPD esters. The European Food Safety Authority (EFSA) assessed both in 2016. Digestion releases glycidol from glycidyl esters, and the chair of EFSA’s panel said: “There is sufficient evidence that glycidol is genotoxic and carcinogenic, therefore the CONTAM Panel did not set a safe level for GE.” For 3-MCPD the evidence is kidney damage and reduced male fertility in rats. EFSA’s 2018 update set a tolerable daily intake of 2 micrograms per kilogram of body weight.
How much, and in which oils. EFSA’s 2016 averages, in micrograms per kilogram of oil:
A US Food and Drug Administration survey of 116 oils bought in 2011 to 2013 (MacMahon, Begley and Diachenko, Food Additives & Contaminants: Part A, 2013) drew the line between refined and unrefined: bound glycidol reached 10.5 milligrams per kilogram in refined oils and no more than 0.03 in unrefined ones. The highest readings were in refined palm oil.
Who is over the line. EFSA’s 2018 update found that adults do not exceed the tolerable intake for 3-MCPD, and reported “a slight exceedance” among high consumers in the younger age groups, above all infants fed only formula. For glycidyl esters, EFSA in 2016 called exposure a potential health concern for all younger age groups at average intake and for high consumers at every age. It also recorded that levels in palm oils and fats had halved between 2010 and 2015 after producers changed their process.
What regulators did. The European Union set legal maximums in Commission Regulation (EU) 2023/915: 1,000 micrograms per kilogram for glycidyl esters in vegetable oils, and for 3-MCPD 1,250 in soybean, corn, rapeseed, sunflower, coconut and palm kernel oils and 2,500 in other vegetable oils, with lower limits for oils going into baby food. In the United States the FDA lists these contaminants among chemicals under review, at the step it calls “Risk Management Action,” and reports working with infant formula makers to bring levels down. Its page on them names no legal limit.
Two smaller refining questions. Hexane, a solvent used to extract oil: EU law caps the residue at 1 milligram per kilogram of oil (Directive 2009/32/EC), and EFSA concluded in 2024 that the solvent’s safety needs re-evaluating. That work was still open at EFSA’s last update in December 2025, so there is no finding yet. Trans fat: the FDA wrote in 2015 that refined oils “may contain trans fatty acids as a result of high-temperature processing, at levels typically below 2 percent.”
What this entry does not claim. It does not claim that seed oils inflame the body or cause heart disease. Su and colleagues pooled 30 randomized trials (Food & Function, 2017) and found that eating more linoleic acid, the main omega-6 fat, did not significantly change inflammatory markers in blood. The American Heart Association’s 2017 advisory (Sacks and colleagues, Circulation) concluded that replacing saturated fat with unsaturated fat, “especially polyunsaturated fats, will lower the incidence of CVD.” The best-known dissent is the 2016 re-analysis of the Minnesota Coronary Experiment by Ramsden and colleagues in the BMJ, which found no mortality benefit from replacing saturated fat with corn oil. That argument is about the type of fat. This entry is about what refining adds.
Correction, October 12, 2026. Until this date the entry argued that the omega-6 fat in seed oils drives inflammation and disease, and it recommended butter, ghee, tallow and lard in their place. The sources it cited do not show that, and no source on the page compared those fats with the oils. It was rewritten around the one refining concern that has a named assessment behind it. The full account is on the corrections page.
Where Diet Discipline stands
Diet Discipline’s advice is to avoid refined cooking oils wherever an unrefined choice does the same job. The reason is not omega-6 fat. It is what refining adds and what it takes away.
What it adds: the European Food Safety Authority found that glycidol, released in digestion from a contaminant that forms when oil is deodorized, is genotoxic and carcinogenic, and it set no safe level. The European Union now caps these contaminants by law. The US Food and Drug Administration names no legal limit. The groups EFSA flagged first are infants and children, and among adults the highest consumers.
What it takes away: a refined oil is a food stripped down to its fat. It carries the calories of the seed, olive or nut it came from and almost none of the fiber, protein or minerals.
Extra virgin olive oil avoids the first. Fat eaten in whole foods, such as olives, avocado, nuts and seeds, avoids both. This preference is a position Diet Discipline holds. It is not a finding that refined oil harms an adult at ordinary intake. The frame to keep: “vegetable oil” on a label names a plant, not a process.
Learn more: Why Fat Is Essential for Your Brain and Body
Or jump straight to:
