New observational study suggests processing itself could partly explain links to diabetes, heart disease, and early death

Concerns about the health effects of ultra-processed foods are growing, as studies increasingly link them to conditions such as heart disease, diabetes, and even early death. But scientists are still debating what’s driving those risks: the nutritional quality of these foods—which are often high in saturated fat, sodium, and added sugars—or the industrial processing and additives used to make them.

A new study from researchers at the Food is Medicine Institute at the Gerald J. and Dorothy R. Friedman School of Nutrition Science and Policy at Tufts University, published in American Journal of Public Health, suggests the processing itself may play an independent role. The researchers found that people who ate more ultra-processed foods had worse health outcomes, even after accounting for the overall nutritional quality of the foods.

“The findings suggest ultra-processed-food factors beyond nutrients—such as changes to foods’ cellular structure, loss of beneficial chemical compounds, additives, and chemicals from packaging—may create health risks not addressed by traditional nutrition metrics or policies,” said the study’s senior author, Dariush Mozaffarian, cardiologist and director of the Food is Medicine Institute.

For the observational study, the researchers analyzed data from 10 consecutive cycles of the National Health and Nutrition Examination Survey (NHANES) from 1999 to 2018, linked to National Death Index through 2018. Study participants had completed one or two 24-hour dietary recalls.

Using a standard classification system, the team grouped foods based on how they were made—from minimally processed food-based ingredients like fruits and vegetables to ultra-processed products made with industrial ingredients and additives not typically used in cooking. The researchers also rated the nutritional quality of foods using a system that scores foods based on their overall healthfulness. Each participant received an overall diet-quality score based on the foods they reported eating. The team then examined how ultra-processed food consumption was linked to current health measures—such as weight, blood sugar, and cholesterol—as well as long-term risk of death.

For every 10% increase in calories from ultra-processed foods, the researchers found worse health markers. People who ate more of these foods tended to have higher body weight, worse blood sugar control, higher blood pressure, and less favorable cholesterol levels. They were also more likely to have conditions such as diabetes, metabolic syndrome, and cancer and had a slightly higher risk of dying during the study period.

These links remained even after researchers accounted for reported foods’ nutrient quality and the amounts of saturated fat, added sugar, or sodium present in the ultra-processed foods. The patterns were largely the same across different subgroups of people.

“Ultra-processed foods make up a substantial portion of the American diet, accounting for more than 50% of adults’ and about 60% of children’s caloric intake,” said Juna Hatta-Langedyk, first author and an undergraduate biology student at Tufts. “Understanding how these foods affect health is a critical public health priority, given the large proportion of the population affected.”

“Addressing structural and policy-related barriers to accessing fresh and minimally processed foods remains critical for promoting dietary changes that improve the health and life span for all Americans,” said Mozaffarian. “Our findings can help inform many current policy efforts, such as a national definition of ultra-processed foods, and multiple states’ endeavors to propose and pass laws addressing ultra-processed foods, including warning labels, bans on certain additives, and limits in school meals.”

Lu Wang, Bingbing Fan, and Peilin Shi from the Friedman School of Nutrition Science and Policy are also co-authors on this study. Research reported in this article was supported by the National Institutes of Health’s National Heart, Lung, and Blood Institute under award number R01HL115189, as well as by an American Diabetes Association’s Pathway to Stop Diabetes award and the Laidlaw Foundation’s Laidlaw Scholars Leadership & Research Programme. Complete information on methodology, limitations, and conflicts of interest is available in the published paper.

The content is solely the responsibility of the authors and does not necessarily represent the official views of the funders.