The Dangers of Ultra-Processed Foods

This past week, Harvard’s T.H. Chan School of Public Health hosted a webinar on the putative dangers of ultra-processed foods (UPF). Let me share my recap.

For those interested in listening to the entire one-hour broadcast, it is available here. The discussion was moderated by journalist Larissa Zimberoff, and featured

  • Jerry Mande, CEO of Nourish Science, “a new non-governmental organization (NGO) focused on solving the country’s current nutrition crisis,” formerly at the FDA and USDA.
  • Josiemer Mattei, Associate Professor of Nutrition at the T.H. Chan School of Public Health, investigates “the genetic, dietary, and psychosocial risk factors of obesity, type 2 diabetes, cardiovascular disease,” among other issues.
  • Kevil Hall, Senior Investigator at the National Institute of Diabetes & Digestive & Kidney Diseases. He is the author of perhaps the best study on UPFs and their impact on weight. My earlier review of this work can be found here.

The discussion is in bold; comments where I cannot remain silent follow.

The Opening

Ultra-processed foods (UPF) are everywhere and makeup nearly 60% of the calories of the typical American diet. While the stereotypical UPF contains chemicals, is packaged in a box, and is in the center of the supermarket, other foods are considered UPFs, including many cereals, breads, yogurts, and frozen dinners — not to mention sweets and sodas. There’s mounting evidence that ultra-processed foods are potentially addictive and uniquely dangerous, linked to conditions including obesity, Type 2 diabetes, heart disease, and cancer.

Those emphasized words “potentially” and “linked” are important because no smoking gun was present, demonstrating a causal relationship between UPFs and those linked conditions.

What are UPFs

The discussants agreed that UPFs were defined by the NOVA classification, which ignores the nutrient values of food but stresses the extent of processing. For a deeper dive, ACSH discussed the NOVA classification and its limitations here. As a generalization, UPFs involved industrial formulations with inexpensive agricultural inputs, including refined ingredients and additives “not typically used in home kitchens.” Not all UPFs are non-nutritional. Ready-to-eat meals “that appear nutritionally sound” are still UPFs. Dr. Hall mentioned that Cauliflower crust pizza is a UPF. Other UPFs include cereals and breads, “even wholegrain breads, and ketchup” because of adding ingredients that extend a product's shelf life.

It should not be surprising that foods must be reformulated to be produced at the scale required to feed the planet and survive the increasingly long food distribution chains. We could remove all the preservatives today, but then we would immediately lose the economy of scale in production, raising prices, and find that many of the foods we take for granted are less available because they are further away than their natural “shelf life” allows. Other additives are introduced to improve mouth feel or palatability. But, the knowledgeable home cook also uses additives to improve taste or mouthfeel, e.g., adding stale bread to soup to increase its thickness or adding some sugar to highlight a peppery tang to a dish. UPFs, because they are produced at scale, create the same mouthfeel with a gel and enhance the tang with an industrial sugar. UPFs, defined solely by their processing, reflect people's concerns about GMOs or the advent of mRNA vaccines. 

UPFs are convenient; they get us out of the kitchen

UPFs are associated with the risk of Type II diabetes and cardiovascular disease. But this is a generalization; some UPFs may be more risky than others. Processed meats are the UPFs exemplifying strong associations with health risks, but other UPFs may result in greater health benefits.

The Atlantic reporting on a study done ironically at Harvard showed that

“Dietary patterns characterized by higher dairy intake, especially low-fat dairy intake, may lower the risk of type 2 diabetes in men.”

Those dairy products include yogurt and ice cream, by NOVA definitions UPFs

Dr. Hall discussed his study showing that individuals on a UPF diet gained weight and body fat compared to control individuals eating a nutrient-matched, equally palatable non-UPF control. However, he pointed out that his study showed that the nutrients were not the underlying driver. He is currently undertaking studies to determine whether calorie density or particular sugar, fat, and salt formulations drive UPF’s impact. He repeatedly stated that any underlying mechanism for UPFs causing weight loss and other health harms has yet to be identified.

Dr. Hall could not have been clearer; we have no idea of a mechanism by which UPFs are “addictive” or, more importantly, harmful to our health. The common narratives are hypotheses and talking points, not settled science.

There has been reported an association between issues of mental health and consumption of UPFs. Dr. Mattei pointed out that “consumption is a two-way street,” and whether the eating preceded the anxiety or depression or followed is unclear. As she pointed out, we call some UPFs comfort foods for a reason, which relates to their physical properties and the emotional context in which they are eaten.

Why are companies making these “hyper-palatable” UPFs?

The majority of manufacturers are publicly owned companies that achieve growth by selling more calories. Their creation of problematic UPFs is an unintended consequence of their search for greater revenue, and food companies must be given more aligned incentives. Jerry Mande, the policy expert in the discussion, stated that,

“The law requires that food not make us sick, but the government is not using its ability to regulate these foods.”

The FDA has “lost its way,” focusing far more on drugs than foods. You should be able to eat what you want without fear of it making you sick.

Policy Options and What if you had unlimited funding?

  • Food as medicine projects - prescription programs of covered meals are associated with increased diet quality but have not been sufficiently long to measure health impacts.
  • Reformulate foods once we understand the underlying mechanism that causes them to be harmful. UPFs are popular for a reason, and we shouldn’t throw out convenience and taste for health.
  • Invest more in nutritional science and cross-specialty research. For example, Dr. Hall’s study raised questions about palatability, now being addressed three years later but hampered by a lack of facilities to carry out the research (which, in this case, requires a 30-day stay in a clinically controlled setting).
  • School meals have shown improvement in HEI scores based on government guidelines. Do the same with the Supplemental Nutritional Assistance Program (SNAP), limiting what may or may not be purchased. The current #1 purchase by SNAP recipients is soft soft drinks.  
  • Set a National goal for all children to be at a healthy weight at age 18 – That has been an aspirational goal of the federal government since President Kennedy introduced a fitness program in the early '60s.
  • Increase FDA funding to regulate chronic food-related illnesses.
  • Tax sugar-sweetened beverages (SSB). This suggestion was offered in the last moments of the discussion; I was surprised that it occurred so late. The current evidence shows that raising the cost of SSBs does reduce their consumption, but in some cases, it shifts to less “healthful” beverages, and there has yet to be a study demonstrating a health benefit.

My takeaway

Despite the program’s title, little demonstrated a causal relationship making UPFs dangerous. Dr. Hall, who freely admitted eating UPFs for their convenience, seemed to be the most objective. He repeatedly pointed out that we do not know the underlying mechanism by which UPFs act. Without that information, it is hard to affix causality or take action.

Jerry Mande’s comment that people are being made sick from their food seems to point to the next regulatory or legal approach. The think-tank types that believe UPFs are harmful will often describe the situation as analogous to the tobacco playbook, where corporations knowingly produced a defective product that harmed its users and withheld that information from us. Certainly, plaintiff’s lawyers will be able to find a sympathetic court that will rule a category of UPFs responsible for a health harm. Of course, the next play in the tobacco playbook is the significant financial settlements. Can a Master Settlement Agreement over UPFs be in our future?