How harmful are processed foods?

    The average American eats around 130 pounds of wheat flour per year, and around 65 pounds of added sugar. This begs the question: how bad are processed foods for your health?

    Now this is a trick question, we’ll just say that up front. But it’s also a very important question.

    Humans have processed their foods for thousands of years, in order to save the time and tedium of chewing tough meats and overly fibrous plants. Basically, we’re not cows, and we don’t really have the stomach or jaw to be chewing all day long.

    And while our ancestors ate a lot more unprocessed food than we do, some also supplemented their diets with flour on occasion. Flour!

    Around ten million years ago, our ape ancestors developed an enzyme to metabolize alcohol, which could be naturally found in rotten fruit. A few million years later, these occasionally drunk apes gave rise to humans who purposely processed (via fermentation) plants into beer, wine, and other alcohols. Fermentation of milk allowed for yogurt production stretching at least as far back as ancient India 6,000 years ago. Overall, fermentation provided not just a tangy new flavor and a bit of a buzz in some cases, but also a great way to process foods into a more well-preserved form. Other processing methods are also an integral part of human history, such as the widespread practice of curing meats.

    Processed foods are not new, they just make up a much larger part of our diet. Modern methods of processing differ from our ancestors as well - with less fermentation, more flour or other ultra-processing, and more monoculture (hello, corn and soy domination!).

    The overall health impact of processed foods is really hard to pin down, because there are so many different kinds of foods processed in different ways. So let’s start with an easily quantified harm of processing … allergens.

    Everyone knows that allergies on the rise, which is terrible for those affected (especially kids) who at risk of dying from even tiny amounts of allergens, especially peanuts.


    Unfortunately, highly processed food made in factories is at higher risk for contamination with allergens. For example, over 80% of oat samples in one study were found to be contaminated with gluten, and factories are not always careful about controlling peanut exposure.

    Outside of allergens, other risks are much harder to pin down. Processed red meat has been labeled a carcinogen by the World Health Organization, and various processed foods have been linked to a bunch of conditions … from instant noodles linked to eczema in Korea, to ultra-processed foods linked to worse cardiovascular profiles in Brazilian children.

    But what does that really mean? Most of that evidence is observational, since you can’t do a randomized trial that assigns people to eat junk food for years, and then check how bad their health gets. So we’re limited to using that evidence to generate hypotheses, and then trying to understand the mechanisms behind what happens in our bodies.

    For example, acellular carbohydrates like flour and processed sugar may predispose people to chronic disease (if eaten in large amounts) due to energy density and possible gut impacts. Large amounts of meat cooked at high temperatures can increase cancer risk through compounds like heterocyclic amines. And so on, and so on.

    The bottom line is that processed food is not inherently harmful, for two distinct reasons. First is that “processed food” isn’t one monolithic thing. Frozen mashed sweet potatoes are technically processed, but their health effect will be much different than deep fried Oreos. Second, chronic disease is nearly always dependent on dose: having cake on yours and your friends’ birthdays is different than eating donuts every day.

    The most prudent way to assess the risk of processed food is combining observational evidence with some some mechanistic evidence. In other words, do humans who eat varying amounts of this processed food develop disease? If so, do the mechanisms make sense for the processed food impacting physiological mechanisms that cause disease?

    To rephrase the takeaway: sugar isn’t evil, nor is flour or cured meat or other processed food. But we humans are not robots, and some people can’t stop themselves from eating too much processed food, in effect eating their way into shorter lifespans caused by chronic conditions. It’s wise to be aware of both the evidence and your personal food habits and triggers for overeating junk food.