Ultra-Processed Foods: Fine Points and a Broad Brush

“Yes, not all types of food processing are bad and not all UPF are equally bad,” writes Carlos Monteiro. He’s commenting on a new study of ultra-processed foods in the American Journal of Epidemiology. Because he is the author and a big promoter of the NOVA UPF classification scheme, his comments are notable. But the authors of this study go a bit further. In fact, they tell us that the risk profiles of different processed foods can be quite distinct. It seems that with ultra-processed foods, many people are trying to make some fine points with a very broad brush. In the words of senior author Taulant Muka:

“Our meta-analysis shows that not all ultra processed foods are harmful, and current classification terminology should be revised as it can be misleading and cause confusion for the public.”

A Systematic Review and Meta-Analysis

Muka and colleagues conducted a systematic review and meta-analysis of 40 unique prospective cohort studies. They used random effects models to calculate pooled relative risk ratios. The focus was on ultra-processed food consumption and all-cause mortality.

Overall, they found a 29 percent higher risk of death associated with the high consumption of UPFs. Individual foods groups with an association of  higher risk were sugary beverages (11 percent increase), artificially sweetened beverages (14 percent), and processed or red meat (15 percent). However, breakfast cereals had a 15 percent lower risk.

But here’s the problem. These are all correlations. Invisibly in the background, many other behavioral factors move in parallel with a tendency to eat or not eat certain kinds of foods. Those other factors can easily be as important or more important as the UPFs this study tracks.

In fact, relative risk ratios in the range of 15 to 30 percent are not especially compelling in light of unmeasured confounding.

Smoke and Fire

We see plenty of smoke coming from the direction of ultra-processed foods. The recent RCT by Kevin Hall and colleagues gives us good reason to believe that something is going on within this grouping of foods. However, the NOVA classification system for ultra-processed foods is a very broad brush for making fine points about food and health. So it’s not completely clear that we know where the fire is.

Some people are calling for global changes to food systems based on current concepts about ultra-processed foods. Do we have sufficient insight into the fine points of ultra-processed foods to justify calls for turning global food systems upside down?

Food for thought.

Click here for the new study by Muka et al.

Moscow Food: Breads. Painting by Ilya Mashkov / WikiArt

Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.

March 14, 2022