Ultra-Processed Foods (Part 1)

Ideally, we should try to center our diets around healthy, unprocessed plant foods, but in today’s world, most of what fills our grocery carts isn’t just processed but ultra-processed, which is good for shelf life, but not necessarily our life. Let’s take a closer look.

Modern nutrition science began about a century ago the in context of nutrient deficiency diseases. Editorials in the Journal of the American Medical Association had titles like “Sugar as Food,” heralding sugar as one of the cheapest sources of calories. For six cents, you could buy 3,000 calories.

But the Nutrient Deficiency era gave way to the Dietary Excess era. No longer were we dying of nutrient deficiency diseases like scurvy as much as we were dying from nutrient excess diseases like obesity and heart disease. So, nutrition science became more about avoiding too many calories, too much saturated fat, too much sugar, and too much sodium, while still focused on nutrients. This allowed food companies to get away with whipping out fiber-fortified Froot Loops. But “Food, Not Nutrients, Is the Fundamental Unit in Nutrition.” And to its credit, the field of nutrition started moving towards a more holistic view.

First generation dietary guidelines emphasized individual nutrients, then moved to second generation food-based dietary guidelines, which largely converged on encouraging diets rich in vegetables, fruits, legumes (beans, split peas, chickpeas, and lentils), whole grains, and nuts. But an area of emerging importance is the degree of food processing.

National nutrition guidelines first told us to cut down on saturated fat, alcohol, cholesterol, salt, and sugar. Then they started actually naming names, suggesting citizens might want to cut down on cake, for example. So, from nutrients to foods, and more recently, countries recommending limiting processed foods.

This was the bombshell that really started the momentum: a commentary in the journal Public Health Nutrition in 2009 by Carlos Monteiro, a nutrition professor at the University of Sao Paulo. Maybe the issue with nutrition and health is not so much the food, nor the nutrients, as much as the level of processing. For example, food-based dietary guidelines might tell us to eat more vegetable soup. Great! But there’s vegetable soup and then there’s vegetable soup. Are we talking about cleaning-out-your-fridge vegetable soup? A health-haloed organic quinoa and kale soup with a heart-stopping 1,200 milligrams of sodium per can? How about a vegetable soup that has more salt than it has vegetables? Or a vegetable-flavored soup that contains more artificial flavors and MSG than it has vegetables. Just like there is breakfast cereal and then there is breakfast cereal.

Professor Monteiro and colleagues proposed the “NOVA” food classification system. NOVA is not an acronym. It means new in Portuguese, as in a new way to classify foods, based on the level of process. NOVA classifies all foods and food products into four groups. Group 1 foods are unprocessed or minimally processed, like fresh, dried, frozen, or cooked plants or animal parts, with nothing bad added, like salt or sugar. Group 2 foods are the salt, sugar, and fats used in cooking, and group 3 includes traditionally processed foods, like when you add group 2 ingredients to group 1 foods. But what really put NOVA on the map is group 4, its concept of ultra-processed foods.

These are industrial formulations of several ingredients which, besides salt, sugar, oils and fats, include food substances not used in culinary preparations, like added flavors, colors, artificial sweeteners, emulsifiers, and other additives used to imitate real foods. An example is a frosted grape pop-tart that has more grapes on the front of the box than inside it, with less grapes than salt, but may artificially taste like grapes and look like grapes because of one, two, three, four, five different food dyes.

“Simply put, ultra-processed foods are foods that can’t be made in your home kitchen because they have been chemically or physically transformed using industrial processes.” “They typically contain little or no whole foods, are ready-to-consume or heat up, and are fatty, salty or sugary and depleted in dietary fibre” and other nutrients—like all the sweet, fatty, or salty snacks like potato chips, ice cream, soda, candy, French fries, burgers, hot dogs, chicken nuggets, and fish sticks. Basically, almost everything in a box or a bag. Why not just call them “packaged foods”? Well, they were thinking about it, but were afraid consumers might look at a bag of apples or something and get confused.

Before the NOVA classification system was introduced, the issue of food processing was largely ignored or minimized in education and outreach, as well as in public health policies. But NOVA exploded onto the stage, resulting in a superNOVA in the world of nutrition.

But what exactly is so revolutionary about this concept of ultra-processed foods? I mean, wasn’t fatty, salty, sugary junk always a bad idea? What’s so new about NOVA? Monteiro suggests that there are harmful effects of ultra-processed foods that are independent of their nutrient profile. Is that true? That’s exactly what I’ll cover next.

At present, the United States has the lowest life expectancy of the top dozen richest countries in the world. We also have the greatest share of our diet composed of ultra-processed foods. In fact, most of our diet. Most of what goes into our mouths isn’t considered real food. Could there be a connection?

Ultra-processed foods are said to be troublesome for a wide variety of reasons, like all the packaging, etc., but I’m just going to discuss the health aspects. Based on studies encompassing nearly ten million participants, greater exposure to ultra-processed food was associated with a higher risk of a variety of adverse health outcomes, including all-cause mortality––meaning living a significantly shorter life.

As a reminder, ultra-processed foods are “industrial formulations of processed food substances … that contain little or no whole foods and typically include flavorings, colorings, emulsifiers, and other cosmetic additives.” Think Twinkies. So nutritionally vacuous that you could go blind from nutrient deficiencies if that’s all you ate, some of the health outcomes make sense, like higher ultra-processed food consumption being associated with more dental cavities in kids. That makes sense, since a lot of them are packed with sugar. Higher ultra-processed food consumption is also associated with an increased risk of high blood pressure. That makes sense, since a lot of them are packed with salt. And with high blood pressures, you can see how there can be a higher rate of cardiovascular events like heart attacks and strokes. And since cardiovascular diseases are our leading killers, you can see how that could lead to increased overall mortality. Obesity is also a no-brainer. A lot of ultra-processed products are packed with calories, and furthermore intentionally engineered with hyperpalatable combinations of fat, sugar, salt, and flavors so you can’t just eat one, promoting weight gain through so-called hedonic eating, meaning eating for pleasure even if you aren’t really hungry.

I can see all that, but why would there also be a higher overall cancer risk associated with ultra-processed food consumption? Or a higher risk of dementia? Or a higher risk of inflammatory bowel disease? Or a higher risk of irritable bowel syndrome? And a higher risk of developing chronic kidney disease, too? Overall, direct associations have been found between exposure to ultra-processed foods and 70 percent of the health issues that were investigated. A higher risk of dying from all causes put together, a higher risk of getting cancer, not sleeping well, suffering from anxiety, depression and other common mental disorders, wheezing, cardiovascular disease, Crohn’s disease, abdominal obesity, obesity in general, fatty liver disease, and type 2 diabetes. So, increased risk of a variety of chronic diseases and mental health disorders, and apparently not a single study reported an association between ultra-processed food intake and beneficial health outcomes. There was never a take two Twinkies and call me in the morning.

Given the large body of evidence implicating ultra-processed foods in human diseases, and their ever-increasing consumption around the world, there is a pressing need to recognize their contribution to the global burden of disease.

All right, but wait. These are associations. Just because A is correlated with B doesn’t mean that A causes B. Maybe B causes A. For example, snapshot-in-time studies show that intake of higher ultra-processed foods is associated with increased odds of depression or anxiety symptoms, but individuals might turn to ultra-processed foods in an attempt to mitigate stress-related anxiety. They don’t call it comfort food for nothin’. Now there are longitudinal studies showing that greater intake of ultra-processed foods preceded depression, but maybe the food was just used to fend off bad feelings for years before any official diagnosis. But reverse causation—B causing A—seems less plausible when it comes to non-mental health outcomes. For example, it’s harder to paint a picture of how a heart attack would cause people to then go on to eat more junk food, instead of vice versa.

The other big issue when trying to establish cause-and-effect from observational, correlational data is confounding. Maybe the reason A is associated with B is that some third confounding factor C is linked to both. For example, if people who eat more junk tend to smoke more, no wonder junk eaters would be dying more, but it would have more to do with the cigarettes they put in their mouth rather than the food.

And indeed, people consuming unhealthy diets are also likely to smoke, drink, and not exercise, but we have ways to control for these factors. Indeed, the majority of studies were adjusted for these confounders. The most obvious confounding factor, though, is that highly processed foods just tend to be more likely to be junky crap. The reason cakes, crullers, candy, and cola are bad for us may have less to do with how much they’re processed, and more to do with the fact they are cakes, crullers, candy, and cola—packed with sugar and calories, with very few nutrients. Diets high in ultra-processed foods are diets high in added sugars, saturated fat, sodium, and calories, displacing more nutritious foods, such as fruits, vegetables, legumes, nuts, and seeds. So maybe ultra-processed foods are just markers of poor diet quality. And indeed, the more ultra-processed foods people eat, the more refined carbs they’re eating, the more sugar, saturated fat, sodium, energy, cholesterol, and the less fiber in their diets.

So, where’s the mystery? People who eat more bad foods get more diseases—we already knew that. What does processing per se have to do with it? Because ultra-processed category foods are more likely to be crappy foods, we could use it as a kind of heuristic that they probably won’t be good for us. For example, if you made up some rule that you’re going to avoid breakfast cereals with cartoon characters on them, you’d probably end up healthier––but not because the graven image of Tony the Tiger magically hastens you to the grave, but rather because they’re proxies for high sugar content. But for the developers of the ultra-processed concept, “ultra-processed” was no proxy. They claimed that the harmful effects of ultra-processed foods are not fully captured by their nutrient profile, and even go as far to say we should avoid ultra-processed foods, irrespective of their nutrient profiles.

Whether ultra-processed diets are detrimental to health simply because they are of a poor nutritional quality, or whether the nature and extent of processing itself has health consequences has been an ongoing debate. How might we settle it? Well, if it were the case that the association between ultra-processed foods and poor health was solely because of the excess saturated fat, sugar, and salt, or just from a less healthful dietary pattern in general, then, if we were to adjust for dietary quality, the association between ultra-processed foods and disease should disappear. In other words, if it was all about the nutrients, then people eating crappy diets high in saturated fat, salt, and sugar should have high disease rates, regardless of whether they eat more ultra-processed foods or less of them. And people who eat really healthy diets should have low rates of disease, regardless of how many ultra-processed foods they consume. But that’s not the case.

The majority of the associations between ultra-processed foods and obesity and health problems remain significant and unchanged in magnitude after adjustment for diet quality or pattern. This means even if you balance out dietary quality, those eating more ultra-processed tend to do worse. This suggests that at least some of the adverse consequences of ultra-processed foods are independent of dietary quality, so it’s not as though the ultra-processed food companies can just reformulate their products to be lower in salt or sugar and be done with it. There appears to be something about this class of foods beyond the standard nutrient profile that contributes to their deleterious effects. And that’s just what I’ll be discussing in this extended video series on ultra-processed foods.