Canceling Ultra-processed Foods Won't Solve Obesity

F. Perry Wilson, MD, MSCE


June 16, 2021

This transcript has been edited for clarity.

Welcome to Impact Factor, your weekly dose of commentary on a new medical study. I'm Dr F. Perry Wilson of the Yale School of Medicine.

I have a love-hate relationship with ultra-processed foods — those semi-industrial, nutritionally poor, calorie-dense products that line our grocery store shelves, screaming at us with their bright colors and empty promises.


I hate them because, well, don't they just seem emblematic of all that is wrong with our diets today? Easy calories, oversalted, oversugared, and just sort of unnatural.

But I love them because I have kids and, well, they're easy. I know, I'm a bad parent.

So I was struck by this study in JAMA Pediatrics examining the effect of ultra-processed food consumption on body mass index and other metrics in children. And I was particularly struck by the overall effect, which was, well, not as big as I expected.

Researchers in England enrolled 9025 children from 7 to 13 years of age and followed them for around 10 years. At baseline, they completed a 3-day food diary to determine how much of their diet comprised ultra-processed foods. One weirdness here: They categorized this by weight, not calorie content. This was because some ultra-processed foods, like diet sodas, may not have calories. Regardless, the numbers here are sort of staggering. The bottom 20% of kids had 23% of their diet made of ultra-processed foods. For the top 20%, 68% of their diet was ultra-processed.

That's a lot.

Chang K, et al. JAMA Pediatr. doi:10.1001/jamapediatrics.2021.1573

When an exposure varies this much, if it's an important exposure, you expect outcomes to vary pretty significantly. And while statistically significant differences did emerge, they aren't actually that dramatic. At baseline, the kids all had pretty similar BMI.

Chang K, et al. JAMA Pediatr. doi:10.1001/jamapediatrics.2021.1573

Over time, the kids who ate more ultra-processed foods did have a higher body mass index, but not by a lot; the average difference was 0.06 units of BMI per year between the highest quintile and the lowest quintile. Body weight increased faster in the high ultra-processed food group — about 0.2 kg per year faster — but there was no difference in overall body fat percentage.

Chang K, et al. JAMA Pediatr. doi:10.1001/jamapediatrics.2021.1573

I was pretty surprised by this. Why don't we see bigger differences? The authors call for mandatory transnational regulation to reduce ultra-processed foods, but I'm not sure that these results actually support that viewpoint.

Of course, obesity metrics are just one potential problem with high ultra-processed food consumption. Other studies have linked ultra-processed foods to diabetes, cardiovascular disease, and cancer. I don't think any of us would be upset to see less of this stuff, provided we could make healthier options equally cheap and available.

But let's acknowledge a couple of issues with this type of study. For one, a 3-day diet diary is just not the most accurate thing in the world. If you don't appropriately categorize the kids' intake, you'll tend to bias the outcomes toward the null; mismeasurement makes everyone look more similar than they should.

Let's also be honest about this: Obesity is complicated. The interplay between diet, environment, and lifestyle is really complex and it is unlikely that any single type or category of food will explain it all.

We could have gotten a bit more explanation if the authors had shown their work in their multivariable analysis. For example, we know they adjusted for total calorie intake, but we aren't shown how much total calorie intake accounts for changes in the adiposity measures. And the calorie-in/calorie-out hypothesis for adiposity is one that continues to find support. In fact, a nice adult study of ultra-processed foods I reported on earlier found that the main reason ultra-processed foods lead to weight gain is because they make us eat more. They are literally designed to trick us into eating more than we should.

What should parents like me do? Remember that the health of our kids comes from multiple places — what they eat, yes, but also how they live. Of course, cutting down on ultra-processed foods isn't going to hurt anyone, except maybe the Nestles and PepsiCos of the world, but as this study shows us, it's really just a small step in the right direction.

F. Perry Wilson, MD, MSCE, is an associate professor of medicine and director of Yale's Clinical and Translational Research Accelerator. His science communication work can be found in the Huffington Post, on NPR, and here on Medscape. He tweets @fperrywilson and hosts a repository of his communication work at

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.