The 60+ Year Nutritional Cold War: Looking towards a truce - a three-part miniseries

PART 1: The Food Wars: How We Keep Picking the Wrong Enemies

James McCormack is returning to Sensible Medicine for the third time. He is a Doctor of Pharmacy in the Faculty of Pharmaceutical Sciences at the University of British Columbia. When he approached us with this idea, I was hesitant, given that it was clear it needed to be a multi-day affair. However, after reading it, it became clear that his work on this subject needed to be on Sensible Medicine. So, for the next three days, Sensible Medicine becomes Sensible Nutrition.

Adam Cifu

For more than 60 years, we have been caught in a war, not over oil or land, but over what we eat. The enemy keeps changing: first it was fat, then sugar, now ultra-processed food (UPF). Misinformation and propaganda drive the intensity of the battles, and sadly, there has often been immense personal collateral damage. Have you ever felt guilty after eating a delicious piece of cheesecake?

Given that eating is one of life’s great pleasures, we have to decide how real the threat is from eating certain foods. To understand the threat, you need the evidence – presented clearly, without hype. This three-part series explores what the research actually says, or does not say, about UPFs - and more importantly, what those numbers might mean for you. You might find the evidence reassuring, especially if you like eating.

The war on fat lasted 30 years, from the 1960s to the 1990s. Low-fat diets became the rage.

With time, we learned that the blanket demonization of fat lacked nuance. Not all fats are created equal. It is trans-fats that have been consistently associated with bad outcomes.

Then, in the 2000s, the war shifted, and sugar became our nemesis. For the most part, it still is. But yet again, this focus on a specific nutrient lacked the proper nuance. Not all sugars are the same. As stated by Sievenpiper in 2020, “Most of the evidence supporting public health recommendations to limit sugars derives from sugar-sweetened beverages providing excess energy”. Keep this history in mind as we delve into the latest food bad actor.

Since the 2010s, our foe has shifted from fats and sugar. The new enemy is UPF. In our attempt to define yet another nutritional enemy, we are making very similar mistakes as with did with fats and sugars. So let’s look at UPF.

How Much UPF Do We Eat and What Is UPF

It is essential to be clear and specific about what we know (and don’t know) about UPF, as we consume a lot of it. In the USA and UK, more than 50% of the energy people get comes from these foods. People in half of the countries listed below get 30% or more of their energy intake from UPFs.

Before examining the evidence linking health issues to UPF, we must first define what is considered UPF. We have been down this road before with fat and sugar—the confusion often starts with the definition. Let’s get that straight first.

NOVA is a food classification system that categorizes foods according to the extent and purpose of food processing, rather than in terms of nutrients (such as proteins, fats, and carbohydrates). Despite scientists' concerns that it is broad and vague, NOVA is the system many researchers use when investigating the health aspects of food, and it is undoubtedly the main one used in UPF research.

NOVA divides foods into four categories.

Based on this system, common examples of UPFs are carbonated soft drinks, fatty or salty snacks, candies, pastries, cakes and cake mixes, margarine, sweetened cereals, fruit yogurt, pasta, pizza, poultry or fish nuggets, sausages, burgers, hot dogs, powdered or instant soup, noodles, and desserts.

Due to concerns about the broad UPF definition, some researchers have suggested examining the NOVA UPF category not as a single entity, but instead dividing it into the following subtypes.

Typical UPF subtypes include

1. Sugar-sweetened beverages

2. Artificially sweetened beverages — sometimes combined with sugar-sweetened beverages

3. Processed meat — sometimes called animal-based products

4. Sweet snacks and desserts

5. Ready-to-eat dishes

6. Savory snacks

7. Yogurt and dairy-based desserts

8. Bread and cold cereal

9. Sauces, spreads, and condiments

How Do We Study The Health Impact of UPFs?

Now let’s delve into what the actual evidence says about UPF, versus what the sensationalized headlines and internet pundits seem to be telling us. And most importantly, let’s put all the UPF research findings numbers into context for you, so you, as an informed individual, can find your way through all the guidelines, advice, and rhetoric — and yet still enjoy food.

But first, how do we research the impacts of nutrition on health? For almost all nutrition research, the headlines come from observational studies. The headlines can be worrisome:

“Ultra-processed foods may be linked to early death”

“Trans Fats Double Risk of Heart Disease”

“Ultra-processed foods significantly raise heart disease risk”

Readers of Sensible Medicine are already familiar with the limitations of observational studies in nutrition research. Despite these limitations, these studies often provide the best evidence we have, as conducting large, long-term diet trials is very difficult and expensive.

In a 2024 paper on the nature of the evidence base and approaches to guide nutrition interventions, the authors suggested, “Observational studies, including large and diverse populations, provide evidence of effects at scale and over long periods and may be the only practicable form of assessing the potential impact of a particular diet or nutrient (the exposure) on a health outcome, where a randomized controlled trial [often done to study medications] is not feasible.”

A significant amount of other nutritional research is being published. Much of that is mechanism of action, basic science, or animal research. Notably, authors have suggested that this type of evidence should be considered insufficient or weak.

This weaker research is often cited as the justification for specific nutritional advice, particularly when higher levels of evidence are not available. When this lower-level research is presented to you with unwarranted confidence as to what food is good or bad for you, it creates the illusion of certainty. This lower-level research has some value, but it is fraught with problems if used to provide specific health advice on food.

For that reason, I’m focusing primarily on observational studies as they are considered the highest level of evidence we have for most important nutrition-related questions.

With observational studies, researchers typically use diet questionnaires, dietary records, diaries, and 24-hour recalls to gather information from thousands of people about what types and amounts of food they eat. Study participants are then followed for years - sometimes decades - and then the health outcomes between people who eat a lot or a little of different types of food are compared. This is done by placing people in different categories based on what and how much they report eating. These categories are known as quantiles - the most common being quartiles (25% categories) or quintiles (20% categories). These categories are based on the distribution of a population rather than any specific values, and often vary significantly between studies.

Researchers then compare these different percent categories to see how many in each category die or get heart disease or cancer, etc. When examining the association between food and health outcomes, the results typically presented are comparisons between the highest quantile and the lowest quantile - the lowest 1/4 or 1/5 (referred to as the baseline) versus the highest 1/4 or 1/5. So, if you were to hear “eating more donuts increases the risk of heart attacks” the difference they are talking about is what was seen when comparing the highest versus the lowest intake.

Importantly, when a difference between the highest and lowest values is observed, it is often not evident when less extreme differences are compared, such as baseline versus the 2nd or 3rd quantile. What this means is that the “dose” of nutrition is a crucial issue to assess when it comes to nutrition research - in many cases, as you will see, the dose makes the poison. A great example is that we all need water to live, but drinking too much water can, literally, kill you.

We are now ready to ask the million-dollar question: What does the evidence actually say about UPFs and your health?

Tomorrow, in Part 2, I will review all the numbers with the proper context—the key to unraveling this latest food-fight skirmish.

James McCormack is a professor at UBC in pharmacy and medicine. He works to get useful evidence-based information out to the masses.