tim's experiment

Nothing makes a good headline like a new cancer scare. So the Consumer Council should have known what the resultant headlines would look like when they sounded the alarm over … coffee.

Coffee? Is nothing sacred? William Blake complained that “priests in black gowns were walking their rounds, and binding with briars my joys and desires.” Now that we no longer take much notice of priests, it seems the scientists have taken over the important task of telling us that anything we enjoy is bad for us.

The Consumer Council’s case against coffee is that 45 of the 47 samples it tested contained acrylamide, which it described as a “genotoxic carcinogen”. Consumers should “pay attention to how much coffee they consume,” said the chair of the council’s research and testing committee.

The first thing that puzzled me about this was the absence of a standard. In this sort of investigation you generally look for some figure issued somewhere by a regulator or scientist, above which the danger is supposed to lurk. But the Consumer Council’s figures were all, without exception, inside the limit recommended in the European Union. Hong Kong does not regulate acrylamide at all.

And so to Google. Acrylamide is not a toxic chemical introduced to coffee, or anything else, as a piece of industrial food modification. It is a natural result of food, especially foods containing some kinds of starch or sugar, being cooked at high temperatures. High concentrations are typically found in chips, crisps, other versions of cooked potato, and toasted bread. Roasting and deep-fat frying reach the required temperature; boiling and steaming do not. So rice is presumably OK. Nobody disputes that acrylamide is also found in coffee, because the beans are roasted.

Indeed this has been assumed for so long that the Consumer Council’s report that two of its samples did not have acrylamide looks a bit suspicious.

It is also universally accepted that acrylamide has been shown to produce cancer in laboratory animals. As the US Food and Drug Administration puts it: “In laboratory studies, acrylamide caused cancer in animals, but at acrylamide levels much higher than those seen in foods. FDA is now conducting research to determine whether the much lower levels of acrylamide in food pose a health risk to people.”

This brings us to the second generally accepted point, which is that it has not been shown that acrylamide causes cancer in humans. The European Food Safety Authority says that “Currently, studies on human subjects have provided limited and inconsistent evidence of increased risk of developing cancer.”

The corresponding body in the UK has “The assessment confirms that acrylamide levels found in food have the potential to increase the risk of cancer for people of all ages. However, it’s not possible to estimate how much the risk is increased.”

The Food Standards Authority Australia and New Zealand states that “There’s no direct evidence that acrylamide can cause cancer in humans. There is evidence that it can cause cancer in laboratory animals. Therefore, FSANZ believes that it is prudent to reduce our exposure … in food.”

Other authorities are less impressed by the reports on mice and rats, which notoriously do not always transfer to larger animals, like us, consuming smaller quantities of the studied substance.

Cancer UK says that “Good quality studies have not shown that acrylamide from food causes cancer in humans.” The US Cancer Society also believes that “dietary acrylamide is unlikely to increase the risk of people developing cancer”.

Hovering over all these carefully chosen words is the thought that science these days is not particularly scientific. The ferocious competition for research grants and publication produces a strong preference for dramatic, or at least substantial, results. Finding that something does not cause cancer is a disappointment.

Readers who wish to go into this sort of thing in more detail need “Science Fictions”, a book by Stuart Ritchie. To which I am indebted for this hair-raising example of what the system can now produce: “Biotechnology company Amgen attempted to replicate 53 landmark preclinical cancer studies that had been published in top scientific journals … A mere six of the replication attempts were successful.” And so on.

For coffee, though, we can point to one specific study, conducted for legal purposes. This arose after a California judge had decided that Starbucks should, in conformity with a state law which requires purveyors of carcinogens to display a warning to consumers, tell its customers that coffee might be giving them cancer. This would have applied also to other coffee sellers.

There were, of course, appeals. The matter was submitted for a decision to California’s Office of Environmental Health Hazards, which looked at more than 1,000 studies and found no substantial evidence linking coffee to cancer.

USA Today quoted Sam Delson, a spokesperson for the Office, as saying that coffee was a mix of hundreds of chemicals – carcinogens and anti-carcinogens. “The overall effect of coffee consumption is not associated with any significant cancer risk,” Delson said.

So it seems that coffee consumers (of whom I am one) can relax, and the Consumer Council’s research was a little overstated, or possibly out of date. The California court case was decided in 2019.

Someone who was eager to beat up the council, which I am not, might even dub its coffee comments “false news”. A more charitable conclusion, perhaps, is that this is an example of scientists being seduced by the desire to make a news splash. Mr Ritchie offers plenty of others. Cancer is a frightening word. It should not be brandished lightly.