Ultra-Processed Foods

Ultra‑processed foods are having a moment. Again. I’m pretty sure we’ve been hearing this since maybe the 60s and wonder how is this new?  I have a distinct memory when I was a teenager in the late 1970’s of a comedian basing a whole routine around reading off the ingredients on a loaf of bread.  At the time, we all thought it sounded bad but the medical community didn’t seem too concerned so we just .. ate the bread. 

Fast‑forward to now and UPFs are everywhere — headlines, podcasts, blogs, documentaries. I even started making my own bread (delicious, yes; required, probably not). I don’t think it’s the end of the world if we buy a loaf at the store.

So this week’s rabbit hole – what do we really know about these ultra-processed foods and how worried should we be? 

The NOVA system: where this all started

It turns out the whole “ultra‑processed” conversation is relatively new. Researchers only began commonly categorizing foods by processing level in the late 2000s. The most widely used system – the NOVA system - was first published in 2010 and updated in 2016.

NOVA has four levels:

• Group 1: unprocessed or minimally processed foods

• Group 2: culinary ingredients (oils, butter, sugar, salt)

• Group 3: processed foods

• Group 4: ultra‑processed foods (UPFs)

  
  

Most modern studies use either NOVA (or other similar classification systems) to estimate how much of our diet comes from UPFs and then look for associations with health outcomes. Surprise! Higher UPF intake is consistently linked with worse outcomes: cancer, heart disease, diabetes, obesity, and so on. 

My issue: NOVA lumps everything together

Group 4 includes the obvious villains: packaged cakes, sugary drinks, soda, candy, sugary cereals. But it also includes:

• Store-bought bread

• breakfast cereals

• flavored yogurt

• meatless patties

• fish sticks

• canned fruit

• popcorn

• canned soup

• hot dogs, bacon, salami

And here’s where I get stuck. Are the calories from a can of tomato soup the same as a Twinkie? I don’t think so. NOVA treats them as identical because it measures processing, not nutritional quality. That’s useful for some questions, but not all of them — and it can lead to some head‑scratching conclusions. Food Network had a nice take on this here.

A Revised Approach: look at UPF sub-types

I was very interested in this paper in the Lancet looking specifically at cardiovascular disease.  Using data from over 70,000 people followed for 30 years, they analyzed consumption of various UPFs and looked at cardiovascular outcomes.  The statistical work in this paper is mind‑blowing — they used sophisticated models to account for changes in diet over time, missing data, and a whole list of other complexities. I won’t bore you with the details, but trust me: they were thorough.

Here’s what they found:

UPF categories linked with higher cardiovascular risk:

• sugar‑sweetened beverages

• processed meats (the worst offenders)

• artificially sweetened beverages

  
  

UPF categories linked with lower cardiovascular risk:

• bread and cold cereals

• yogurt and dairy desserts

• savory snacks

  
  

Yes — you read that right. More calories from bread/cereal, yogurt/dairy desserts, and savory snacks were associated with lower cardiovascular risk.

UPF categories with no meaningful association:

• several others, including hard liquor (WTF)

This is why lumping everything into “UPF = bad” doesn’t make sense. The category is too broad to be useful.

Takeaways from the paper

• Avoid processed meats (boo! No bacon or charcuterie boards)

• Avoid sugar‑sweetened beverages

• Avoid artificially sweetened beverages

• Don’t panic about bread, cereal, or flavored yogurt

  
  

This study only looked at cardiovascular outcomes.  Now that the analytic framework is built, I hope there will be more papers on diabetes, cancer, and other outcomes. I really look forward to seeing this work. 

What I do

• I still make my own bread because it’s delicious.

• I stick to plain yogurt but mix in a high‑fiber cereal (Smart Bran) Yes, it’s technically a UPF; yes, it has some added sugar; it also ups my fiber intake so no, I’m not losing sleep over it.

• I cook most meals from scratch because I enjoy it.

• I sometimes make my own loose sausage, which is just ground meat mixed with seasonings.

• I’ll serve a charcuterie board when entertaining — maybe once a month.

• I avoid processed meats otherwise.

• I don’t stress about food when traveling or eating out.

The research will keep evolving, and so will my habits. I’m choosing the patterns that feel reasonable and sustainable for me.