55

About a year ago I heard that burning your food can lead to an increased risk of getting cancer. The explanation was that the burned portion of the food was carcinogenic. The Carbon that would be produced in burning of the food was also involved in this somehow although its relevance wasn't clearly explained. I was dubious of the claim until I posed the question to some chemist friends of mine who, to my surprise, were not as dismissive of the idea as I was.

Is burning your food something that can increase the likelihood of getting cancer and, if so, it the increase enough to be concerned about?

Update

This question has hit the headlines again (at least in the UK early in 2017). For example, The Independent reports:

Over-cooked potatoes and burnt toast could cause cancer, new research suggests

The headlines have emerged because the Food Standards Agency has issued new advice:

The Food Standards Agency (FSA) has issued a public warning over the risks of acrylamide - a chemical compound that forms in some foods when they are cooked at high temperatures (above 120C).

A new campaign tells people how they can cut their risk, including opting for a gold colour - rather than darker brown - when frying, roasting, baking, grilling, or toasting.

The warnings are not just about burnt food but about well-cooked food and specifically mention acrylamide as the guilty party.

update 2018

The story has reappeared (again) in the UK where a major supermarket has been criticised by several bodies for selling "well fired" bread: The Sun, for example, reports:

Experts claim that the company should warn people of the blackened edges of the bread, as they may contain a cancer-causing chemical.

The story appears to be based on the same Food standards Agency advice that triggered some of the previous stories.

  • 4
    @Thomas As far as I remember, burning meat produces some carcinogenic compounds. The carbon part is probably a misunderstanding, probably carbon-containing chemicals are meant instead of pure carbon. – Mad Scientist Jun 18 '11 at 22:15
  • 9
    I think you are talking about Acrylamide – Oliver_C Jun 18 '11 at 22:23
  • 1
    @Thomas You're correct in that I wasn't trying to suggest that it was Carbon itself that is carcinogenic. That's obviously not true. I've edited my question to reflect that. – jon_brockman Jun 18 '11 at 22:27
  • 3
    @Oliver_C Likely also N-nitrosamines. – Konrad Rudolph Jun 21 '11 at 10:45
  • 2
    @ThomasO Hydrocarbons like benzene and benzopyrenes produced by incomplete combustion are definitely carcinogenic. – endolith Jun 7 '17 at 18:15
28

Acrylamide (C3H5NO) is a chemical compound produced when starchy foods are burnt. It is also in coffee, prunes and olives amongst other foods, and is inhaled from cigarette smoking.

Ingestion of acrylamide has been linked to a number of health concerns, including

(Direct exposure to acrylamide causes problems too, but that's out of scope here.)

For balance, acrylamide has been ruled out of causing several other concerns, such breast cancer and cancer through exposure at work.

The Wikipedia page on acrylamide provides a more comprehensive list of the research and various government's strategies for dealing with it.

Note: None of this answer addresses the seriousness of the dosages found on a typical piece of burnt beef. The issue may be too small for serious concern, when placed in the context of other dietary concerns. @Matt Black's answer addresses this shortcoming - please consider it for an upvote.

  • 12
    Note: I am not happy with my answer. It does not provide any indication of the seriousness of the dosages found on a bit of burnt beef. I use the C word, without explaining the size of the relative risks, which I suspect are so small as to be not worth even considering. (e.g. there are other, bigger, reasons to cut back on cigarettes, meat and coffee, and that prune juice is a net gain, healthwise) I would appreciate the input of others here. – Oddthinking Jun 18 '11 at 23:29
  • 5
    I remember a friend telling me about this as a prime example of the press picking up the "dangerous" claim and then losing interest (as disconfirmatory studies and explanatory notes like the WHOs effect size refresher trickled in), so maybe public fear is somewhat exaggerated. It would be helpful if claims like this were quantified against something like lung cancer risk from smoking, otherwise it may have the effect of "Oh everything's cancerous, but it's okay for me to smoke because I avoid charred toast". – Ruben Jul 19 '12 at 9:09
  • 1
    Newbie question - what's the C word? (I would guess correlate or cause. Could be cancer or coffee or chemical.) – djechlin May 14 '13 at 14:17
  • @djechlin I assume cancer. – Wertilq Jun 3 '13 at 13:31
  • 1
    @Oddthinking I've included some specifics in a new answer that might address some of the dosage issues you were worried about which have emerged again as the UK food safety agency have specifically warned about acrylamide from overcooked food. – matt_black Jan 23 '17 at 15:32
28

There is no evidence that the acrylamide produced by browned food causes measurable harm to people

The big worry about browning or burning food is the presence of acrylamide which is produced by the reaction of sugars and amino acids (via the Maillard reaction which also produces many of the attractive flavours of cooked food).

The UK Food Standards Agency has launched a campaign to encourage people to reduce their exposure to acrylamide. This seems reasonable given that the chemical is pretty nasty in large quantities and is known to cause cancer in laboratory animals.

But the dose makes the poison. David Spiegelhalter, an expert on the statistics of risk and the communication of science, has criticised the FSA campaign. The text below summarises his conclusions.

Firstly, epidemiological evidence of any association of acrylamide intake from food and human cancer is missing despite extensive studies. As a large european literature analysis concluded:

In the epidemiological studies available to date, AA [acrylamide] intake was not associated with an increased risk of most common cancers, including those of the GI or respiratory tract, breast, prostate, and bladder. A few studies suggested an increased risk for renal cell, and endometrial (in particular in never-smokers) and ovarian cancer, but the evidence is limited and inconsistent.

Spiegelhalter points out that:

Remember that each study is testing an association with a long list of cancers, so using the standard criteria for statistical significance, we would expect 1 in 20 of these associations to be positive by chance alone.

Which, given the lack of consistent results in a large number of studies, tends to suggest there is no significant effect to be discovered.

He also analyses the numbers associated with the animal studies. He points out that the doses that caused measurable cancers in animals were far higher than those that people are exposed to in their typical diet.

Simplifying slightly, the lowest dose known to cause significant tumours in mice is about 170 microgrammes per kg of body weight per day (and the mice used are often more susceptible to cancer than people). In real dietary surveys of people, the high exposures were more like 1.1 microgrammes per kg of body weight per day. Almost everyone in dietary surveys consumes less.

Spiegelhalter concludes this analysis:

So, for example, adults with the highest consumption of acrylamide could consume 160 times as much and still only be at a level that toxicologists think unlikely to cause increased tumours in mice (that’s essentially what the ‘margin of exposure’ means).

There are plenty of things in our diets we should worry about if we want to reduce our risk of cancer (drinking too much alcohol and being fat both matter). The best that reducing acrylamide can do is to give a bigger margin of safety between an unmeasurably small risk and an extremely unmeasurably small risk.

Spiegelhalter argues that eliminating risks that are practically meaningless is a bad way to create trust in dietary advice.

5

Yes it does.

Other answers are about acrylamide, but there are many other carcinogens, too:

Benzo[a]pyrene (BaP) is ... one of the benzopyrenes ... and is the result of incomplete combustion at temperatures between 300 °C (572 °F) and 600 °C (1,112 °F).

It is also found in ... all smoke resulting from the combustion of organic material ... and in charbroiled food. A 2001 National Cancer Institute study found levels of BaP to be significantly higher in foods that were cooked well-done on the barbecue, particularly steaks, chicken with skin, and hamburgers: Cooked meat products have been shown to contain up to 4 ng/g of BaP, and up to 5.5 ng/g in fried chicken and 62.6 ng/g in overcooked charcoal barbecued beef.

BaP's metabolites are mutagenic and highly carcinogenic, and it is listed as a Group 1 carcinogen by the IARC.

which is a specific type of benzopyrene:

Benzopyrene is ... naturally emitted by forest fires and volcanic eruptions and can also be found in coal tar, cigarette smoke, wood smoke, and burnt foods such as coffee. Fumes that develop from fat dripping on blistering charcoal are rich in benzopyrene, which can condense on grilled goods.

Benzopyrenes are harmful because they form carcinogenic and mutagenic metabolites (such as (+)-benzo[a]pyrene-7,8-dihydrodiol-9,10-epoxide from benzo[a]pyrene) which intercalate into DNA, interfering with transcription. They are considered pollutants and carcinogens.

which is a specific type of PAH:

Heterocyclic amines (HCAs) and polycyclic aromatic hydrocarbons (PAHs) are chemicals formed when muscle meat, including beef, pork, fish, or poultry, is cooked using high-temperature methods, such as pan frying or grilling directly over an open flame. In laboratory experiments, HCAs and PAHs have been found to be mutagenic—that is, they cause changes in DNA that may increase the risk of cancer.

HCAs are formed when amino acids (the building blocks of proteins), sugars, and creatine (a substance found in muscle) react at high temperatures. PAHs are formed when fat and juices from meat grilled directly over an open fire drip onto the fire, causing flames. These flames contain PAHs that then adhere to the surface of the meat. PAHs can also be formed during other food preparation processes, such as smoking of meats.

HCAs are not found in significant amounts in foods other than meat cooked at high temperatures. PAHs can be found in other charred foods, as well as in cigarette smoke and car exhaust fumes.

 

Some heterocyclic amines (HCAs) found in cooked and especially burned meat are known carcinogens. Research has shown that heterocyclic amine formation in meat occurs at high cooking temperatures. For example, heterocyclic amines are the carcinogenic chemicals formed from the cooking of muscle meats such as beef, pork, fowl, and fish. HCAs form when amino acids and creatine (a chemical found in muscles) react at high cooking temperatures. Researchers have identified 17 different HCAs resulting from the cooking of muscle meats that may pose human cancer risk.

The United States National Cancer Institute's Division of Cancer Epidemiology and Genetics found a link between individuals with stomach cancer and the consumption of cooked meat, and other studies for colorectal, pancreatic, and breast cancer is associated with high intakes of well-done, fried, or barbecued meats. People who eat medium-well or well-done beef were more than three times as likely to suffer stomach cancer as those who ate rare or medium-rare beef. Other sources of protein (milk, eggs, tofu, and organ meats such as liver) have very little or no HCA content naturally or when cooked.

  • "yes it does" as an answer is not supported by the evidence you supply. Unless you judge not just the presence of carcinogens but the dose in context and support that with strong epidemiological evidence, you cannot conclude that. – matt_black Apr 5 '18 at 15:27
  • @matt_black Did you read the entire answer? – endolith Apr 6 '18 at 14:04

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .