Beware the lethal sausages of doom! Or, you know, don’t.

Ok, so hands up if you read the terrifying news that people who eat lots of processed meat are 44% more likely to die prematurely than their counterparts who eat a small amount of processed meat.

44%? That’s, like, nearly half! And I’ve eaten loads of bacon sandwiches! OMG! OMG! I’m going to die!! I can feel the evil, sausagey death gremlins wrapping their hands around my vital organs, and – oh, no! – they’re starting to SQUEEZE …

So, yeah, anyway I did my thing that I do, which is to try and offset my panic by taking my uncomprehending, non-specialist head off to have a look at the scientific paper that is the ultimate source (open access) of the news stories to see what I could puzzle out for myself, and whether the !!BEHOLD YOUR DOOM!! reporting was justified. As always when I write about these kinds of topics (yawn), I must point out that I’m not a scientist, doctor, dietician or anyone else with relevant expertise and experience, that my highest scientific qualification is a grade D in A level biology, and that in fact I’m a complete layman whose opinion is worth precisely nothing. But from what I can tell, the reality is a lot less terrifying than the headlines.

That big, scary number

I’ve scanned the entire paper – including the statistical analysis which goes way, way over my head, and the tabulated results that go a little way over my head – and I can’t for the life of me see where the “44% more likely to die young” figure comes from. (Though if you pinned me up against the wall and forced me to speculate, I’d guess that someone has pulled a couple of raw numbers from the tabulated results and calculated the difference between them.) The main statistic that the authors themselves report in both their abstract and main conclusions is that 3.3% of the deaths in the group of people they were studying would have been avoided if everyone had eaten less than 20g of processed meat a day. 3.3% is, you won’t need me to tell you, a much smaller number than 44%.

That smaller, less scary, number in context

The researchers followed 448,568 people from several European countries – who began the study in good health, and varied in age from 35 to 69 – for an average of a little under 13 years. In total, 26,344 of those people died. If 3.3% of those deaths were avoidable with a low-processed-meat diet, that translates to 869 avoidable deaths – amongst almost half a million people, and across a period of 13 years. In percentage terms, 0.19% of the entire group died because they ate too much processed meat. You won’t need me to tell you that 0.19% (which is the actual percentage of people who died because they ate too much processed meat) is a very small number indeed.

That thing the journalists didn’t tell you

The processed-meat-related deaths were the result of cardiovascular disease and cancer, but mainly cardiovascular disease. In other words, this study has rediscovered the well known fact that too much fat and salt lead to an increased risk of heart disease, high blood pressure and stroke, and those diseases can kill you young. In this case the too-much-salt-and-fat were found in processed meat, but they’re harmful whatever they’re eaten in. The increase in cancer deaths is also a concern, but a secondary one: bacon and sausages are mainly bad for you because they’re fatty and salty, and too much salt and fat are bad for your heart and circulation. We already knew this.

That other thing the journalists didn’t tell you

Previous, smaller studies in America had concluded that processed meat was harmful, but that red meat in general was even more harmful. This study found that fresh red meat had no statistically significant effect on death rates, but processed meat had a larger effect than the US studies suggested. This is good news if you like sirloin steaks, and bad news if you like gammon steaks.

That little niggling doubt that just popped up

It’s also quite odd news, since fresh red meat is high in saturated fat, and the association between early death from cardiovascular disease and a high intake of saturated fat is well established. The authors’ speculation that eating lots of red meat didn’t cause early deaths in their study because it is ‘often consumed after removing the visible fat tissue’ doesn’t seem persuasive to me: what’s the basis for assuming that the participants in this study cut off more fat from their red meat than the participants in other studies who did die young? So far as I can see, the authors have an unexplained anomaly in their findings when it comes to red meat and early deaths from cardiovascular disease.

That reason for treating the whole study with caution

At the head of their results section, the study authors note that

Men and women in the top categories of red or processed meat intake in general consumed fewer fruits and vegetables than those with low intake. They were more likely to be current smokers and less likely to have a university degree.

This tells us that people who eat a lot of processed meat are likely to have a less healthy lifestyle overall. That in turn creates a problem when it comes to assessing the effects of processed meat consumption – since smoking and low fruit and veg intake also correlate with early death, how to be sure that the early deaths are the result of the processed meat and not one of those other causes? The solution is detailed analysis of the results. So, for example, the authors of this study checked to make sure that the relationship between high processed meat consumption and early death held true for smokers and non-smokers, and for people who eat lots of fruit and veg and people who don’t, and for people who are overweight and people who aren’t. The size of the effect was different in the different sub-groups, but because it was detectable in all the sub-groups it’s possible to conclude that the effect is real – or, at least, is not purely the result of one of the other factors.

So, that’s it, then, case closed – eating lots of processed meat increases the risk of dying young? Well, yes and no.

The problem with this kind of analysis is that it looks at variables like smoking, diet and education (that relative lack of university degrees) as though they were fully independent: you either smoke or you don’t; you either eat lots of fruit and veg or you don’t; and so on. But in reality, all the variables cluster together, they’re partially interdependent, and their effects are synergistic – this study found, for example, that the effect of processed meat consumption on early death is much more pronounced among smokers than non-smokers. That means that the averaged results are slightly misleading – they suggest that eating lots of processed meat has a consistently harmful effect for everyone when it doesn’t – and that in turn leads to widespread misinterpretation.

That reason not to buy into the media’s anti-sausage hysteria

The take away message from this study shouldn’t be, as the media have interpreted it, “bacon and sausages kill”, and it certainly shouldn’t be heralded by strident claims that cutting out this one particular foodstuff will make a substantial difference to your chances of dying young. What this study actually suggests is that 0.19% of people may die early as a result of eating too much processed meat – and that a disproportionate number of those extra deaths will be found amongst people who lead otherwise unhealthy lives, and were already known to be at greater risk of dying young. It suggests that processed meat is mainly harmful for reasons we already knew about, and are widely understood – that foods high in saturated fat and salt are bad for cardiovascular health, and poor cardiovascular health can lead to early death. And it suggests that the sensible thing to do if you want to avoid an early death is to give up smoking, and to pursue a balanced diet that’s rich in fresh fruit and veg, low in fat, and includes a variety of different sources of protein. Not to run round in a headless panic about the amount of processed meat you eat while ignoring the rest of your diet and lifestyle.

And, finally, that reason just to live your life without stressing out too much

You can’t part die, so anything that carries a greater than 0% and less than 100% chance of killing you early effectively carries a 50% chance – it’ll either kill you young, or it won’t. If you’re careful to eat no more than 20g of processed meat a day a statistician will be slightly more surprised if you die early than she would be if you spent your every waking moment cramming sausages wrapped in bacon down your gullet, but her very slight surprise won’t matter much to you, because you’ll be dead. You can take every conceivable precaution, follow every piece of sensible healthy-living advice, and still drop dead of a heart attack or cancer aged 45 (although you almost certainly won’t). Conversely, you can do everything you shouldn’t and live to a ripe old age. You’d be unlucky in the first case, and lucky in the second, but that doesn’t mean it won’t happen – in fact, the statistics prove it does happen, sometimes. It’s sensible to reduce your chances of having horrible things happen to you – you’ll either be killed crossing the road or you won’t, but it still makes sense to look both ways before you step off the kerb – but there’s no point obsessing about it. It makes sense to try and eat a broadly healthy diet, but weighing every rasher of bacon before you put it in your mouth is just silly.

This entry was posted in Media commentary, Stuff I've read and tagged , , , , , , , . Bookmark the permalink.

2 Responses to Beware the lethal sausages of doom! Or, you know, don’t.

  1. J. Wibble says:

    One of my favourite Tim Minchin quotes is, “The day they find out yoga mats are carcinogenic will be the happiest day of my life.”

  2. Kapitano says:

    37.52% of journalists are full of shit.

    So are the other 54%.

    11% can’t do percentages.

Comments are closed.