There’s a story that we’ve all been told for decades. It varies, but it goes something like this:
“If you want to be healthy, eat natural. Processing is awful, all those nasty chemicals, you don’t know what you’re eating!”
It’s usually followed by an in-depth description of someone’s newest paleo-keto-raw-juicing diet, because what’s a PKRJ diet without sharing it around to every single person you meet?
You see, we all know that processing is bad. It’s obvious. Just look at processed versus unprocessed food! A pizza is basically the embodiment of Satan’s temptation, and apples are a virtuous delight reserved for angels and people who work at soup kitchens.
Which brings us nicely to the newest study to grace the headlines of virtually every news source from around the world.
Because, apparently, not only is processed food a bit high in fat and sugar, it is giving you cancer.
But before you run screaming from MacDonalds and dive headfirst into a life of subsistence farming it’s worth taking a second look at the science behind all these stories.
Processed food isn’t that scary after all.
The study the media is talking about is a new piece of research that looks at the association between the amount of processed foods that people eat and how much cancer they get. It was done in France, and looked at the rates of three cancers: breast, prostate, and colorectal. They conducted some really quite amazing statistical analyses that showed that overall, increases in the amount of ultra-processed foods that people at were associated with increases in all cancers.
Basically, processed food is bad and gives you cancer.
Except it’s not quite that simple.
Firstly, this was observational research. I’ve said this time and again, but observational research isn’t there to prove that one thing causes another. With a number of large studies such as this, in different areas, combined with animal models and some small RCTs you could draw a reasonable conclusion, but from a single large study it’s a bit of a reach.
More importantly, something that everyone missed is that the increase in cancer that this study found was tiny. It sounds terrible when you say “a 12% increase in cancer”, but this was based on a total number of 2,228 cancers in over 100,000 people. The base rate of cancer was ~0.5% per year, and the absolute difference between people who ate the most processed food and those who ate the least was roughly 0.1%.
These were also proportional hazard ratios. What this means is that they reflect an increase in risk for one thing happening when another thing increases. In this case, the 12% increase was based on a 10% increase in the proportion of ultra-processed food that people were eating. In other words, you have a 12% increased risk of developing cancer for every 10% of your diet that you replace with ultra-processed food. Which is a huge amount. In real terms, that’s replacing your breakfast with hash browns every day, or having a french fry weekend every week.
There are other issues with making statements based on this study. The increase in risk didn’t actually hold true for all types of cancer. The study was entirely based on two self-report surveys, meaning that it’s possible that these results don’t actually reflect the amount of processed food that people were eating. People who ate more ultra-processed food were also unhealthy in many other ways — more likely to smoke, less likely to exercise, higher total fat/sugar intake — which means that it’s hard to make any definitive conclusions. It might just be that poor people, who are less able to access health services and are less well-off in many ways, also eat more processed food.
With such a small increase in risk, it’s far more likely that these results can be explained by social factors — say, poor people being less able to access healthy food — than by something to do with the food itself.
Which is not really that surprising. You see, ‘processed’ food is a bit of a meaningless category.
And here we come to the crux of the matter. This study may be scary, but it’s also based on something of a myth: the idea that processing food makes it significantly less healthy.
You see, the guidelines that designate foods into ‘processed’ versus ‘unprocessed’ don’t really differentiate based on how good the food is for you, but focus on things like packaging, preservatives, marketing, and profitability. This is a bit weird, because none of these things actually has an impact on your health.
There are even many examples of unprocessed foods that are far less good for you than processed ones. Fresh orange juice, for example, is considered a minimally processed or unprocessed food, despite being extremely high in sugar, whereas diet Coke is an ultra-processed food despite not having any sugar in it at all. As I found when I started trying to lose weight, the number of calories in fresh orange juice makes it a terrible thing to drink if you’re trying to be healthy. Similarly, ‘raw’ food bars can be both minimally processed and truly awful for your health (hint: if it’s soaked in honey, it’s probably not great to eat).
Ultimately, this is just another Big Scary Study that has been hyped to terrifying proportions. Is processed food bad for your health? Maybe. It depends on the food. Is unprocessed food good for your health? Again, maybe.
The study itself is a fascinating piece of work that will likely be built on in years to come, but to you? For your life? These results don’t really mean all that much. If you want advice on your diet, see a registered dietitian. They spend years in university learning how to give you the best possible advice on how to eat healthily.
But don’t worry about the odd chicken nugget.
The evidence suggests it isn’t going to harm you much, if at all.