Emissions from High-Temperature Frying and Cancer

In 2010, an IARC working group looked at frying for the first time. IARC is the cancer research arm of the World Health Organization. The result is Volume 95 of the IARC Monographs.

The wording matters. IARC did not judge fried food. It did not judge cooking in general. It judged emissions from high-temperature frying. That means the air that rises off hot oil.

Three methods were in scope: stir-frying, deep-frying, and pan-frying. Baking was left out on purpose. The group judged that baking gives off something too different.

The finding came in two parts:

• Limited evidence in humans.
• Sufficient evidence in animals, for fumes from hot unrefined rapeseed oil.

Together those gave the overall label. Emissions from high-temperature frying are probably carcinogenic to humans, Group 2A.

The visible cloud over a frying pan is not one thing. Much of it is steam from water in the food. The rest is chemistry.

Heated oil breaks down. Some products drift off as gases. Others cool and turn into tiny particles. IARC lists the main gases from frying as aldehydes, alcohols, ketones, alkanes, phenols, and acids.

Frying also drives up ultrafine particles and fine ones called PM2.5. Those particles carry other compounds stuck to their surface. Two named groups are polycyclic aromatic hydrocarbons and heterocyclic amines. IARC also names four gases in cooking air: formaldehyde, acetaldehyde, acrylamide, and acrolein.

The mix is not fixed. It shifts with the oil, the heat, the food, and the style of cooking.

This is the most useful part of the whole volume. IARC reports two heat levels. Below 100 degrees Celsius, cooking oil has little mutagenic potential. Above 230 degrees Celsius, it has high mutagenic potential. Mutagenic means able to damage DNA.

That upper figure is 446 degrees Fahrenheit. Most oils start to smoke below it. Smoking oil is the warning sign you can see.

IARC adds one more note. Unsaturated fats seem to be more mutagenic than other fats when heated.

Since the 1970s, 17 case-control studies have looked at cooking fumes and lung cancer. All were done in Chinese populations. The sites were mainland China, Taiwan, Hong Kong, and Singapore.

The reason is specific. Smoking is a major cause of lung cancer in Chinese men and women. But a high share of lung cancer in Chinese women is not explained by smoking. Many of those women never smoked. So researchers went looking for what else was at work.

IARC judged only four of the 17 studies most informative. The test was strict. A study had to split the effect of the cooking oil from the effect of the stove fuel.

The Hong Kong study built a single score for frying. It combined how often and how long a person fried, across all three methods. The unit was dish-years.

• More than 150 total dish-years: a significant threefold increase in lung cancer risk.
• More than 200 total dish-years: an eightfold increase.

Two studies in Shanghai and one in Gansu, China, showed risk rising with frying frequency. The highest-frequency group had a nearly twofold rise. In Gansu, the rise was significant for stir-frying but not for deep-frying.

IARC is candid about the weak spot. In three of those four studies, the solid fuel under the pan could not be ruled out as the real cause.

IARC's overall reading is simple. No clear difference showed up between the three methods when each was studied on its own.

The Hong Kong study did compare them head to head, per 10 dish-years. Deep-frying carried the highest risk. Pan-frying was in the middle. Stir-frying was lowest. All three were significantly raised. The Shanghai and Gansu studies did not adjust the methods against each other, so they cannot separate them.

The mechanism data point mainly at lipid peroxidation. Heat drives oxygen reactions inside the fats, and those reactions make harmful products. IARC calls this an important mechanism for these mixtures. The polycyclic aromatic hydrocarbons may add harm through their own known routes.

Positive mutagenicity results turned up in nearly every kind of live-animal test. IARC says those data alone would have supported Group 2A. That holds even if the animal or human evidence had been weaker.

The group drew a line here, and it is worth respecting. The human data cannot pin the risk on one chemical. They cannot even pin it on the cooking oil alone. Some risk could come from the food, from the hot stove or pan, or from the fuel below.

The animal evidence covers only unrefined rapeseed oil heated to high heat. No other oil has that much evidence behind it.

This volume sets no safe number of dish-years. It sets no top frying temperature for home use. It sets no required kitchen air-change rate. Beyond the 100 and 230 degree figures, any stated threshold is not from IARC.

What the evidence base does support:

• Keeping oil below its smoke point, since mutagenic potential climbs steeply with heat.
• Moving the fumes out of the room. These studies used fume extractors, kitchen windows, and having a separate kitchen as measures of exposure.
• Treating years of professional cooking as the heaviest exposure, since risk rose with total dish-years.

To see how one label fits into overall risk, read our overview of cancer prevention and our explainer on hazard versus risk.

Emissions from high-temperature frying sit in IARC Group 2A. That rests on lung cancer studies in four populations, strong mutagenicity data, and one animal model. The exposure that mattered was decades of frequent frying, often in kitchens with poor airflow. The two numbers to carry away are 100 and 230 degrees Celsius. Our page on lung cancer covers symptoms and screening eligibility.

• IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, Household Use of Solid Fuels and High-temperature Frying: Evaluation and Rationale, IARC Monographs No. 95, 2010, accessed August 6, 2026
• IARC Working Group, Household Use of Solid Fuels and High-temperature Frying: Summary of Data Reported, 2010, accessed August 6, 2026
• IARC Working Group, Household Use of Solid Fuels and High-temperature Frying: Exposure Data, 2010, accessed August 6, 2026
• IARC Working Group, Household Use of Solid Fuels and High-temperature Frying: Studies of Cancer in Humans, 2010, accessed August 6, 2026