What does the NIH Ultra-processed food study really tell us?

This past month has been a difficult one for defenders of processed food. The recent NIH study is the first to directly implicate ultra-processed food as a contributor to weight gain and a potential cause of obesity, but this research should not be the end of food processing as we know it! Like any good study, this article raises other questions that should be looked at. Unfortunately, these questions are not being discussed in the popular media. News stories like to paint results from science articles as definitive proof to answer all concerns about an issue and point to a clear solution. Not so fast! Science doesn’t work like that. NPR broke the news. Numerous other stories have appeared including one that offers some caution suggesting that the issue is more complicated than what it first seems.

Since then I have had the opportunity to read and carefully study the original article in Cell Metabolism. First, it is an extremely well-planned and well-executed study that represents a randomized, controlled trial—the gold standard for such research. The investigators carefully matched the two diets and provided a very clear overview of the design and the data. They also pointed out the limitations of work, most of which have been ignored by subsequent news articles. The most important limitation is that these participants were locked up in a NIH facility for 4 weeks which is the best way to determine exactly what the subjects ate but not a real-world situation as to how Americans live their lives. There are some implications of these results that the authors don’t seem to have considered.

I usually start out each month with a review of a book on processed food. This month I will review the NIH study using the same style by responding to direct quotes from the article in bold:

“20 inpatient adults received ultra-processed and unprocessed diets for 14 days each” (1). At the heart of the study was a direct comparison between two diets. Ultra-processed refers to food products that “are industrial formulations typically with five or more and usually with many ingredients” as classified by the NOVA group. The unprocessed diet actually relies on many foods classified as processed by the NOVA group which are described as “relatively simple products made by adding sugar, oil, salt or other group 2 substances [processed culinary ingredients] to group 1 foods [unprocessed or minimally processed foods]. Most processed foods under this classification scheme have two or three ingredients.” (2) The study was designed to investigate “whether ultra-processed foods affect energy intake in 20 weight-stable adults” whose average BMI identifies them as overweight. My posts in subsequent weeks will focus on the benefits and limitations of this classification scheme.

To help clarify the situation here are some examples from the study:

Ultra-processed foods: Honey Nut Cheerios (General Mills), Whole Milk (Cloverfield), Oatmeal Raisin Cookies (Otis Spunkmeyer), Blueberry yogurt (Yoplait), Refried beans (Old El Paso), Tempura fried chicken nuggets (Pierce) with ketchup (Heinz), Diet lemonade (Crystal Light) with NutriSource fiber, plain bagel (Lender’s)

Unprocessed foods: Greek yogurt (Fage) parfait with strawberries, bananas, with Walnuts (Diamond), Salt and Olive Oil, Spinach salad with chicken breast, apple slices, bulgur (Bob’s Red Mill), sunflower seeds (Nature’s Promise) and grapes, Scrambled egg (made from fresh eggs), Oatmeal (Quaker) with blueberries and raw almonds, 2% milk (Cloverfield)

Ultra-processed snacks: Baked Potato Chips (Lay’s), Dry Roasted Peanuts (Planters), Cheese & Peanut Butter Sandwich Crackers (Keebler), Goldfish Crackers (Pepperidge Farm), Applesauce (Lucky Leaf)

Unprocessed snacks: Fresh oranges and apples, raisons (Monarch), raw almonds (Giant) chopped walnuts (Diamond)

I have a few quibbles with the diets such as why whole milk and dry roasted peanuts were considered ultra-processed while 2% milk and sunflower seeds were considered unprocessed. On the whole, however, the diets accurately reflect the differences between ultra-processed and all other foods in the NOVA classification.

“Diets were matched for presented calories, sugar, fat, fiber, and macronutrients” Participants of were offered over 5400 calories of food each day on either diet and told they could eat as much or as little as they wished. As shown above in sample foods and snacks for each diet, there were no extreme foods in either diet as it would have been difficult to match calories, sugar, fat and protein otherwise. One problem was matching fiber in the two groups. Nutrisource fiber was added to beverages in the ultra-processed diet while fruits and vegetables provided most of the fiber in the unprocessed diet. It would have made more sense to use whole-grain breads and breakfast cereals, both of which are considered ultra-processed in the NOVA classification. Subjects on the ultra-processed diet consumed significantly more calories at breakfast and lunch than those on unprocessed diet. No significant differences were observed in calorie consumption for dinners or snacks.

boxes of high-fiber breakfast cereals
Breakfast cereals are classified ultra-processed foods

Ad libitum intake was 500 kcal/day more on the ultra-processed versus unprocessed diet” On the ultra-processed diet subjects consumed 2879 of the 5436 calories or 53% of calories offered. On the unprocessed diet subjects consumed 2370 calories or 44% of the calories offered. Much has been made that the ultra-processed diet resulted in a gain 2 pounds in two weeks, while little has been made of the 2 pounds lost when consuming the unprocessed diet over two weeks. News reports herald the benefits of weight loss on the unprocessed, BUT it must be remembered that these participants were offered over 5400 calories daily! Weight maintenance not weight loss is the goal of a diet for a healthy person. Remember, these participants were chosen because they were weight stable. It is not clear if either diet would have resulted in a similar trend over the course of a year. Either way, however, an unintended weight loss of 50 or so pounds would not be considered any healthier than an unintended weight gain of 50 pounds in a typical person starting out at a healthy weight.

“Body weight changes were highly correlated with diet differences in energy intake” This study essentially verifies that a calorie is a calorie. Contrary to numerous stories on the internet, an ultra-processed calorie seems to carry the same weight as an unprocessed calorie. Carbohydrate calories and fat calories increased in the ultra-processed diet relative to the unprocessed diet suggesting that, contrary to the NPR headline, it is about the amount of carbohydrates and fats we consume. It also is clear that the ultra-processed foods in this study are more likely to be finger foods than the corresponding unprocessed foods, and it is so easy to overeat finger foods! The amount of protein consumed in each diet was remarkably the same. Perhaps the most disturbing aspect of the two diets is that they were both very high in sodium with the unprocessed diet resulting in twice the recommended level and 2.5 times the recommendation! Did the participants actually consume that much sodium?

 “Limiting consumption of ultra-processed foods may be an effective strategy for obesity prevention and treatment.” Watching one’s calories is an effective way to prevent gaining weight and thus prevent Americans from becoming overweight or obese. That does not seem to be a novel concept. Likewise, putting overweight or obese people on a restrictive diet is an effective way of helping them lose weight. Whether such weight loss is permanent or temporary after the restrictive diet is finished is another matter. Again, this study has nothing to offer on management of weight control. It would appear that it is easier to over-consume calories on an ultra-processed diet than on an unprocessed diet. Maybe a mix of ultra-processed and unprocessed foods is most likely to result in weight maintenance as neither diet extreme was effective at maintaining weight over a two-week period.

 “policies that discourage consumption of ultra-processed foods should be sensitive to the time, skill, expense, and effort required to prepare meals from minimally processed foods—resources that are often in short supply for those who are not members of upper socioeconomic classes.” A major caveat of this study was noted in the original article. The unprocessed diet was 40% more expensive than the ultra-processed diet and was much more difficult to prepare. Most of these dishes were prepared from scratch by the research staff. As noted in a follow-up to the original story on NPR, it will not be a simple transition from a typical American diet consisting of 57.5% or more ultra-processed foods to one that is predominantly unprocessed.

Bottom line. This NIH study shows that that when consuming a diet of exclusively ultra-processed foods one is likely to consume too many calories. A diet eliminating ultra-processed foods appears to be useful for losing weight, but such an unprocessed diet may not be adequate to maintain weight in healthy individuals. A balanced diet that includes fruits, vegetables, whole grains and lean proteins incorporated into a diet also containing ultra-processed foods could provide an appropriate blend of health and convenience for a modern family.

Next week: What makes a processed food an ultra-processed food?

(1) Quotes in bold type are directly from the NIH study as reported in Cell.

(2) Quotes in regular type are directly from the NOVA classification of foods and culinary ingredients.

16 thoughts on “What does the NIH Ultra-processed food study really tell us?

    1. Journalists may be critical thinkers, but too many fail to review a journal article critically. Too many journalists look for something in an article that can become click-bait without really analyzing the overall findings. They also leave out the caveats and limitations of the study. I addressed this issue earlier in a post titled ‘Merchants of certainty’ See it at https://processedfoodsite.com/2018/01/30/merchants-of-certainty-reading-too-much-into-a-single-study/
      BTW, I will be describing some of these pitfalls from interpretations of stories on the web over the next two weeks. Best wishes.


  1. I have noticed you don’t monetize processedfoodsite.com,
    don’t waste your traffic, you can earn extra cash
    every month with new monetization method. This is the best adsense alternative for
    any type of website (they approve all websites), for more details simply search in gooogle:
    murgrabia’s tools


    1. I understand that blogs can be a source of extra cash. I deliberately do NOT monetize the site because I am more interested in communicating ideas than making money, a weakness endemic to retired academics. I receive no monetary support from any external source including the food industry, even though I have been accused of being a shill for Big Food. I do receive a small amount off every book that is bought on Amazon.com when referred from my site. I do this merely to be able to show the images of the books I review in the text of my posts. In the almost four years I have been an amazon associate I have earned a little less than $20 through this practice, none of which I have cashed in to date. I also use royalties from my books to help support the site and buy the books I review on it. Thank you for helping me clarify my philosophy on monetizing my site.


    2. I am glad that you noticed that I don’t monetize this site. The ideas are mine based on my professional studies and life as a food scientist (12.5 years of graduate and postgraduate education at three colleges and 31.5 as a member of the college faculty at the University of Georgia). I do not receive funding from the food industry or any other source either directly from this site or under the table. Many of my thoughts and posts are controversial but they are my own.


  2. Excellent review! I’m disappointed that this study missed so many obvious opportunities, such as including fiber in the foods themselves, rather than the drinks. You make a great point about including more breads and cereals, since they are apparently ultra-processed.

    I also think that including flavored drinks in the ultra-processed diets added an unnecessary variable and could have potentially contributed to the amount consumed. I know I don’t eat as much pizza if I’m drinking water instead of beer, wine, or soda.

    Liked by 1 person

  3. Well said! I enjoyed reading your comments. This just reinforces that dietary behaviors aren’t simple and there are many factors that influence our eating patterns. I, too, am a little disturbed by the whole/2% milk debacle.


    1. Thank you for your comment. If one is ultra-processed, then so is the other one. My reading of the NOVA classification would be that neither is ultra-processed. As I understand milk processing, the fat is separated initially from the milk during processing and then added back to the desired level of fat. Excess fat is then used primarily to make cheese.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s