The risks of scaremongering Calls for a bacon ban are a hysterical distraction from deadlier things

Smoking, though, is a special case. There’s nothing else that a fraction of the population uses lots, most people use not at all, and which is unambiguously poisonous. In the years after Richard Doll et al showed that smoking caused lung cancer, epidemiologists were hopeful that there’d be lots of similar things: that they’d be able to show, say, that people who ate pomegranate all died of pleurisy, or drinking tea prevented gout, or something.

But it’s messier than that. Research into nutrition is really, really hard – you can’t reasonably do a randomised controlled trial on broccoli, and make 5,000 people go on a broccoli-free diet for 20 years and another 5,000 eat broccoli every day. If something were as bad for us as smoking, then you’d be able to see the effects with cruder, more observational research. But there isn’t, really. There’s a semi-jokey piece of dietary advice from the writer Michael Pollan from a few years ago: “Eat food. Not too much. Mostly plants.” Despite all the claims about superfoods and so on, and despite many decades of research into the impacts of food on our health, that remains about as precise as scientifically well-supported dietary advice can get.

But wait! Weren’t we just talking about how bacon causes cancer? Well: yes. The International Agency for Research on Cancer, a subset of the World Health Organisation, issued a report in 2015 declaring that processed meats including bacon and sausages are carcinogenic. It was a careful assessment of years of prior research, and I’m sure it’s as good as it can be.

That said, there’s carcinogenic and carcinogenic. Smoking all your life raises your risk of lung cancer from about a one-in-1,000 chance of diagnosis in any given year to about a one-in-75 chance, according to this study; as best I can work out, your lifetime lung cancer risk goes up from about 0.2% to about 10%. (It’s kind of unclear because it all depends on how much you smoke and for how long.)

For comparison, if you eat two sausages or two rashers of bacon every day for your entire life, your risk of colorectal cancer goes up, roughly speaking, from about a one-in-1,500 chance per year to about a one-in-1,400, or about 5% in your lifetime to 6% in your lifetime.

To put it another way, if your anti-smoking measures stopped 10 people smoking, you’d probably prevent one case of lung cancer. But you’d have to stop 100 people eating bacon to prevent one case of colorectal cancer. That’s before you think of all the other health effects of smoking (various other cancers, vascular damage, heart disease), and the fact that lung cancer is more deadly than colorectal cancer is.

Sugar is more complicated: people who consume lots of sugar seem to be at greater risk of diabetes, over and above the risk caused by eating too many calories and becoming obese, which sugar, a very energy-intensive foodstuff, obviously contributes to.

But I can’t get a good sense of the scale of the risk; the meta-analysis I linked to earlier seemed to think that about 8,000 cases of type 2 diabetes in Britain per year could be attributed to sugary drinks, which sounds bad, but there are about 230,000 cases a year in total so it’s not a huge proportion. That is, however, only a small part of the overall problem; sugar also contributes to the obesity crisis, which is at the root of a much larger proportion of our health problems. A fifth of 10- to 11-year-olds are classified as obese, a direct result of poor, sugar-heavy diet.

So should we let government step in? My instinct is that a tax on processed meats for health reasons would be overkill – although there might be good environmental and animal-welfare reasons – while I’d be happy to take PHE’s lead on a sugar tax, because of the impact on obesity more than for its own sake.

The most important thing, though, is to express these risks realistically. The science of all this stuff is, as I have said, messy. In the case of processed meat, the cancer risk is relatively small. But the risks are constantly being compared with those of tobacco, which are literally of a different order of magnitude.

This stuff matters. Last week, another study was published about public perceptions of vaping. Vaping is hugely less damaging for you than smoking; it’s probably not harmless, but so far the evidence suggests that the risk is near-negligible compared to tobacco. But the study finds that nearly half of people think that vaping is about as dangerous as smoking.

That’s a direct result of scaremongering, by the media, but also by health campaigners and others who go around hyping every tiny study that finds that vape smoke has some irritant effect on lung cells in petri dishes or whatever, without putting in the context. It will cost lives, because people who might otherwise have switched from fags to e-cigs now will not, and a large number of those people will, therefore, die.

I’m not saying that scaremongering about bacon is as bad. But we do need to recognise that “bad for you” is not a binary condition, and not all carcinogens are born equal. If we rampage around insisting the government tax everything that might possibly give us cancer if we eat too much of it for 40 years, then the really important stuff – the message on smoking, obesity, drinking to excess – is going to get lost in the noise.

---

Tom Chivers is a science writer. His second book, How to Read Numbers, is out now.

TomChivers