A 200,000-person study found diet quality matters more for your heart than processing

A large new analysis has landed in the middle of one of the most contested questions in nutrition: whether the harm attributed to “ultra-processed” foods comes from the processing itself or from the poor nutritional quality that often accompanies it. Drawing on more than 200,000 adults tracked over several decades, researchers reported that the overall quality of a person’s diet was strongly linked to the risk of heart disease, type 2 diabetes and death — while the level of food processing, considered on its own, was not.

The distinction matters because public messaging in recent years has increasingly urged people to avoid ultra-processed foods as a category, treating the processing label as a shorthand for unhealthy. The finding suggests that shorthand is unreliable, and that what a food contains nutritionally is a better guide than how heavily it has been manufactured. The work was presented at a major nutrition conference and, importantly, has not yet been through the full peer-review process.

Separating diet quality from processing level

The analysis was led by researchers at Harvard and pooled data from three long-running studies that collectively followed more than 200,000 U.S. adults from the 1980s into the 2020s, according to the announcement carried by EurekAlert. The central challenge the team set out to solve is that nutritional quality and processing level are usually tangled together: many ultra-processed products are high in added sugar, sodium and refined carbohydrates, so people who eat a lot of them tend to have poor-quality diets overall. To pull the two threads apart, the researchers scored each participant’s diet for overall quality using a plant-based diet index, then separately classified the foods people ate by processing level using the Nova system, which sorts foods from minimally processed to ultra-processed regardless of their nutritional content.

Four groups, one clear divide

With quality and processing scored independently, participants were sorted into four combinations: healthy diets built largely from ultra-processed foods, healthy diets built from minimally processed foods, unhealthy diets high in ultra-processed foods, and unhealthy diets made mostly of minimally processed foods. Over the follow-up period, about half of the participants died or developed heart disease or type 2 diabetes, giving the analysis a large number of outcomes to compare. The pattern that emerged was consistent: the risk of those outcomes tracked closely with overall diet quality, with healthier diets carrying lower risk and unhealthier diets carrying higher risk. Within each quality category, however, the risk was almost identical whether the diet leaned ultra-processed or minimally processed. In other words, a healthy diet that happened to include a lot of processed foods looked about as protective as a healthy diet made from scratch, and an unhealthy diet was harmful either way.

Why the Nova system can mislead

The results expose a structural limitation in using processing level as a proxy for healthfulness. Because the Nova classification does not weigh a food’s nutritional value, it can place genuinely nutritious items in the “ultra-processed” bucket while leaving some poor choices in the “minimally processed” one. Fortified breakfast cereals, whole-grain packaged breads, canned beans and plant-based dairy products can all count as ultra-processed even though they deliver fiber, nutrients and other beneficial compounds. Meanwhile, foods such as white rice, fruit juice and homemade desserts made with refined flour and sugar are technically minimally processed yet contribute little to a healthy diet. The lead researcher, Xiaowen Wang, a postdoctoral fellow in nutrition at Harvard, framed the practical takeaway as focusing on what a food actually contains rather than whether it has been processed. Established dietary guidance already leans this way; resources such as the Harvard-affiliated Healthy Eating Plate emphasize the composition of a meal — vegetables, whole grains, healthy proteins and oils — rather than the manufacturing history of its ingredients.

What the researchers recommend

Rather than telling people to avoid an entire processing category, the researchers suggested that dietary advice continue to promote well-established, high-quality eating patterns such as the Mediterranean diet or a healthy plant-based diet, even when those patterns include some processed foods. Wang also noted that encouraging food manufacturers to improve the nutritional content of packaged products could make it easier and more affordable for people to eat well, since convenience foods are a fixed part of how most households actually shop and cook. The framing is a shift in emphasis rather than a green light for junk food: the same analysis found that unhealthy diets raised risk regardless of processing, so ultra-processed products that are also nutritionally poor — the fast food, soda, packaged desserts and refined snacks that dominate many diets — remain foods the researchers say should be limited.

A preliminary finding, not the last word

The findings come with a significant caveat that the researchers themselves flag. The work was presented at NUTRITION 2026, the annual meeting of the American Society for Nutrition held in late July 2026 near Washington, D.C., and abstracts presented at the conference are reviewed by a committee but have not generally undergone the full peer review required for publication in a scientific journal. That means the results should be treated as preliminary until they appear in a peer-reviewed publication, where other scientists can scrutinize the methods and statistics in detail. Observational studies of this kind can also show associations without proving cause and effect, since people who eat higher-quality diets may differ in other ways that affect their health. Even with those limits, the analysis adds a large data set to a debate that has often run on smaller studies and strong opinions, and it points toward nutritional quality — not the processing label alone — as the more useful measure of whether a diet supports long-term heart health.

This article was researched and drafted with the assistance of AI and reviewed before publication.

More from Morning Overview