This paper’s conclusion isn’t as much of a “duh” as it first seems.

We know, in the broad sense, that processed foods are bad for us. But there are two possible reasons why. One reason might be that food processing is inherently bad; the alternative is that processed foods tend to be bad for other reasons - say, they’re more energy dense, higher in carbohydrate or fat content, higher in glycemic index, lower in fiber, etc. - and that processing is just a proxy for those other factors. To my knowledge, this is the first trial to directly test the first claim, by comparing two diets that are meant to be identical but for the fact that one consists of highly processed foods and the other does not. In other words, is the same amount of fat, carbohydrates, sugars, etc. worse for you when you eat it as processed food?

At a glance, the results seem to suggest the answer is “yes”. A processed-food diet led to higher caloric consumption than an unprocessed diet, when the two diets were as equal as possible in all other factors. That would suggest that processed food leads us to eat more calories, specifically because it is processed. If true, that would have significant implications for the cause, and potential treatments, of obesity. However, there are reasons to hesitate to extrapolate from these results:

First, both diets were a “standard American” diet - moderately high in both carbs and fat. Assuming you COULD create a processed diet that was, say, low-carb or low-fat, would this effect still exist?

Second, despite researchers’ best efforts, there were meaningful differences between diets. If, for example, fiber affects appetite / caloric consumption, fiber consumed as food may have very different effects from fiber consumed as supplements.

Third, and most importantly, this was a short study of a small number of subjects in a highly unusual environment. The nutrition literature is full of short-term studies that do not replicate over longer time periods. I’m sure future research will show if this is one of them.

From a pragmatic standpoint, this may be splitting hairs for many - either way, the average unprocessed diet WILL be healthier than the average processed-food diet, whichever of the two reasons above is correct. But this question of whether processed food is inherently bad may be one of the most important ones to answer as we look toward truly scalable solutions to the epidemics of obesity and metabolic disease in the U.S.