Foods such as tinned legumes, wholegrain breakfast cereals, and fortified soy-based milk fall into the same processing category as soft drinks, fried potato products, and commercially produced baked goods. The classification framework underpinning that label organises food according to production methods rather than nutritional composition.
A Harvard study tracking over 200,000 adults for 30 years found that dietary patterns were closely associated with rates of cardiovascular disease, type 2 diabetes, and mortality. The origin of the food — whether industrially produced or prepared in a kitchen — made little discernible difference.
The statistical analysis was conducted by Xiaowen Wang, a postdoctoral fellow in Harvard's Department of Nutrition. The findings were delivered at a nutrition conference in July, ahead of formal publication in any academic journal.
The Nova classification framework divides food into four categories based on the degree of industrial processing each product has undergone. The fourth category — ultra-processed foods — accounts for a notably large portion, taking in soft drinks and pre-packaged sweets alongside tinned legumes, wholegrain loaves, and fortified breakfast cereals.
Many processed and packaged food products contain elevated levels of added sugar, sodium, and refined carbohydrates. The two factors — processing and nutritional quality — tend to be intertwined, which complicates efforts to distinguish between them. Those who consume ultra-processed foods in large quantities also tend to have lower overall dietary quality scores.
Research spanning three decades, published in The BMJ in 2024, linked the highest consumption of ultra-processed foods to a modest increase in early mortality risk. When the researchers adjusted their analysis to incorporate broader dietary habits, the association became less pronounced. Wang advised consumers through a media outlet to look beyond ultra-processed classifications and consider a food's nutritional profile alongside their overall eating patterns.
Her dietary guidance centred on increasing intake of whole grains, fresh produce, legumes, nuts, and plant-based oils. Dietary choices she identified as worth reducing include those containing elevated levels of added sugars, excess salt, refined grains, and processed meat products.
The research team led by Wang drew from three longitudinal cohorts: the Nurses' Health Study, tracking female participants since 1976, its follow-on iteration, and the Health Professionals Follow-Up Study, which began recruiting male participants in 1986. The figures broke down to 160,690 female and 42,585 male participants.
At the outset of the study, participants were free of cancer, heart disease, and diabetes, and they submitted dietary records throughout the following 30 years. Each diet underwent two rounds of evaluation by the research team.
One metric assessed the proportion of an individual's plant-based intake coming from nutritionally favourable sources — such as whole grains, vegetables, legumes, nuts, and plant oils — versus less favourable options like refined grains, starchy vegetables, and added sugars. The second assessed the degree to which those plant-based foods had undergone ultra-processing. Participants were divided into four categories based on their combined scores.
By the end of the follow-up period, cardiovascular disease had been recorded in 26,209 participants, type 2 diabetes in 21,006, and a total of 55,493 deaths had occurred. The analysis divided participants into the highest-scoring 20 percent and the lowest-scoring 20 percent for comparison.
Those in the highest-scoring group for the nutritious-packaged-foods dietary pattern faced a 14 percent reduced likelihood of cardiovascular disease, a 33 percent reduced likelihood of type 2 diabetes, and an 11 percent reduced likelihood of mortality. On the healthy-and-barely-processed dietary pattern, the figures registered at 18 percent, 34 percent, and 7 percent respectively.
Wang told a news outlet that the data point to health risks commonly linked to ultra-processed foods potentially stemming from the overall nutritional quality of the diet rather than the degree of processing. She noted that nutritional quality remains the primary consideration, as not all ultra-processed foods meet an acceptable health standard.
Those whose diets ranked highest on the unhealthy-and-barely-processed scale faced a 21 percent greater likelihood of developing heart disease and a 14 percent greater likelihood of developing type 2 diabetes compared with those whose diets ranked lowest. The identical pair of figures — 17 percent and 11 percent — was recorded for the processed-and-unhealthy food category as well. Those who consumed predominantly processed and packaged foods faced a 7 percent higher mortality rate over the study period, while the gap in the minimally processed category fell within the margin of statistical uncertainty.
Wang argued that dietary guidelines ought to continue prioritising food quality over the degree of processing. She also noted that blanket policies treating all ultra-processed foods as uniformly harmful risk oversimplifying the available evidence, while guidance toward dietary patterns with robust research support — such as plant-forward eating or Mediterranean-style nutrition — remains advisable, according to Wang.
Her position also included urging food producers to reformulate packaged goods by boosting dietary fibre and whole grain content while reducing levels of added sugar, sodium, and saturated fats.
The American Society for Nutrition noted that the results remain early-stage and have yet to be evaluated through the peer review process. None of the three studies represent a nationally representative sample of the American population, and the study drew on healthcare workers who tracked and recorded their own eating habits across an extended period.
The research is limited to identifying correlations between dietary habits and health outcomes, rather than establishing causation. Wang noted that food processing exists along a continuum and is not inherently detrimental, and that certain forms of food treatment can enhance microbial safety, prolong freshness, or boost nutritional content through the addition of vitamins and minerals.
The findings were delivered by Wang at the NUTRITION 2026 conference.
