The weight of the food on the plate remained essentially the same. What changed were the calories contained in each gram: after 16 weeks of a low-fat vegan diet, the energy density of the diet dropped by around 30%. It is the result of a new analysis published on JAMA Network Openconducted on data from a randomized trial involving overweight adults. No one had been given a maximum amount of calories to meet.
The difference is far from tiny. In the group assigned to the vegan diet, declared energy intake decreased by approximately 491 kilocalories per day compared to the start of the study; in the control group the decrease was approximately 134. The difference between the two groups was therefore equal to 357 kilocalories per day. The overall weight of food consumed, meanwhile, did not change in a statistically significant way. In practice, a similar amount of food brought with it much less energy.
More low-energy density food, without counting calories
Energy density indicates how many calories are concentrated in one gram of food. One hundred grams of vegetables and one hundred grams of a high-fat food, to give a simple example, can take up comparable space on the scale and have very different calorie content. Water, fiber and fat change this ratio a lot.
In the trial, participants assigned to the intervention were allowed to eat ad libitumtherefore without a prescribed calorie restriction. However, the diet was very precise: vegetables, fruit, cereals and legumes, no animal products and no added fats. In the experiment, about 75% of the energy came from carbohydrates, 15% from proteins and just 10% from fats; a daily supplement of vitamin B12 was also provided. Participants attended weekly meetings with cooking instructions and demonstrations, even if meals were not provided.
And it is useful to remember this, because the word “vegan” alone says little about the energy density of what ends up on the plate. Chips, sweets and many very fatty products can also be free of animal ingredients. Here the intervention had a decidedly different composition.
In fact, the food diaries show where the shift occurred: in the vegan group the average consumption of vegetables went from around 250 to 393 grams per day, that of legumes from 36 to 110 grams and that of fruit from 106 to 185 grams. At the same time, meat, dairy products, eggs and the category that includes dried fruit and fats decreased. The reduction in energy density has been linked above all to the disappearance of animal foods and the increase in vegetables and legumes.
The weight has dropped, but this study tries to understand why
The weight loss was already known. In the original trial, conducted in Washington between 2017 and 2019, the group on the vegan diet lost an average of 6.4 kilograms in 16 weeks, compared to about half a kilo in the group that continued to eat as before. The difference attributed to the surgery was 5.9 kilograms.
Research published in August 2026 returns to the same data to look for one of the possible mechanisms. Of the 244 participants initially randomized, 223 had completed the trial and entered the new analysis. Those who had reduced their energy density the most also tended to lose more weight: the association remained statistically significant after taking into account the change in reported calories.
This allows the authors to indicate energy density as one of the mechanisms that can contribute to weight loss. However, it does not transform that relationship into an independent proof of cause and effect: the diet was randomly assigned, the reduction in energy density was not. Furthermore, this is a secondary analysis ad hoc of data collected several years earlier, not a new experiment conducted in 2026.
Then there is a word that is very easily added to results of this type: satiety. Foods rich in water and fiber can help introduce less energy while maintaining a high volume of food, and the authors point to controlled studies that have observed this effect. In this trial, however, hunger and satiety were not directly measured. To say that participants lost weight “without feeling hungry” would therefore go a step beyond the available data.
The result concerns a specific group and only 16 weeks
The champion also places some stakes. Participants were between 25 and 75 years old and had a body mass index between 28 and 40; the average age was approximately 54 years and 87% were women. People with diabetes were excluded. Furthermore, these were volunteers who had responded to advertisements to participate in the research, a population that the authors themselves recognize as not very representative of the general population.
Diet was reconstructed through diaries filled in by the participants for three days at the beginning and end of the study, therefore with all the limitations of self-reported food records. And the 16 weeks allow us to observe what happens in the short term, much less how sustainable a similar diet is and what effects it produces after years.
Finally, there is financing. The new analysis was supported by the Physicians Committee for Responsible Medicine, with which several authors are affiliated. Neal Barnard is president without compensation and has reported income from books and conferences on nutrition; statistician Richard Holubkov has received honoraria from the organization. In the article, the authors clarify that the funder did not participate in the design, data analysis, or decision to publish. It is information to be placed alongside the results, without making it do any more or less than its job.
That 30%, therefore, tells something quite concrete. During these 16 weeks the participants continued to put more or less the same amount of food on the scale, but mostly filled it with foods that brought fewer calories per gram. No mysterious metabolic switch and no pass for any product with “vegan” written on the packaging: what really mattered was what was actually on the plate.
