You have probably seen the warnings. Cut back on ultra-processed food. Stick to whole ingredients when you can. For a few years now, that line has been shorthand for eating well.
A new analysis of more than 200,000 American adults is complicating that advice. According to the American Society for Nutrition, overall diet quality tracked heart disease, type 2 diabetes and death risk far more closely than whether the foods were ultra-processed.
The findings come from a conference abstract presented at the group's annual meeting in July 2026. The work has not yet been peer reviewed, but it draws on an unusually long record of real-world eating.
Harvard researchers pooled data from three long-running cohorts, the Nurses' Health Study, Nurses' Health Study II and the Health Professionals Follow-Up Study. Participants were followed from the 1980s into the 2020s.
Also read: A new epilepsy diagnosis in midlife may carry a six-fold higher risk of early dementia
Across that stretch, about half the group had a heart event, developed diabetes or died. When the team sorted the diets by both nutrition quality and processing level, the pattern lined up with quality, not with how industrial the food was.
What diet quality actually looks like on your plate
The researchers built so-called plant-based diet scores. A healthy pattern leaned on whole grains, fruits, vegetables, nuts, legumes and plant oils. An unhealthy pattern leaned on refined grains, sweets and sugary drinks.
Both patterns had ultra-processed and minimally processed versions. Within each quality group, the risks of heart disease, diabetes and death were close to identical, regardless of how industrial the food was.
Reporting from Healio put numbers on the healthy side. The hazard ratio for cardiovascular disease landed near 0.86, meaning about 14 percent lower risk during follow-up.
Also read: How fast you eat may be one of the most overlooked reasons you gain weight
The risk of type 2 diabetes dropped by roughly a third in the healthiest eaters, and overall death risk fell by about one in ten. These are observational findings. They describe a strong link, not proof that one thing caused the other.
Why the processing label alone can miss the point
The study takes aim at how foods get sorted. The Nova system groups foods by how industrially they are made, with no account of nutrition. As News Medical summed up, that means a healthy food can land in the ultra-processed bucket while an unhealthy one stays out of it.
Under that lens, a fortified whole-grain cereal and a bottle of soda both sit in the "ultra-processed" category. Their effects on the body are very different.
Lead author Xiaowen Wang, a postdoctoral research fellow at the Harvard T.H. Chan School of Public Health, put it plainly in a statement.
Also read: How eating a handful of walnuts every day may protect your memory and brain function
"While many ultra-processed foods tend to be of poor nutritional quality and should generally be limited or avoided, processing itself is not inherently harmful," Wang said.
One caveat matters. The team deliberately zoomed in on the plant-based parts of participants' diets to keep the comparison clean. The findings speak most directly to plant foods, not to overall diet composition.
The practical direction from the researchers is simple. Load the cart with vegetables, fruit, whole grains, nuts, legumes and plant oils. Judge the rest by its nutrition label, not just the ingredient list.
Also read: After a first blood clot, morning coffee was linked to half the risk of a second
Also read: Actress Jennifer Garner shares two protein-rich smoothies you can easily make at home
This article is made and published by Jesper Bengtson, who may have used AI in the preparation.