Higher ultra-processed food intake linked to increased mortality risk in older adults

In a new study, older adults who reported consuming higher amounts of ultra-processed foods, as defined by the NOVA classification system, were about 10% more likely to die over a median follow-up of 23 years compared with those who consumed less processed food.

The findings are based on a large study that has tracked over half a million U.S. adults for nearly three decades. According to the results, higher intake of ultra-processed foods was associated with modest increases in death from any cause and from deaths related specifically to heart disease or diabetes, but no association was found for cancer-related deaths.

Our study results support a larger body of literature, including both observational and experimental studies, which indicate that ultra-processed food intake adversely impacts health and longevity. However, there is still a lot that we don't know, including what aspects of ultra-processed foods pose potential health risks."

Erikka Loftfield, PhD, Stadtman Investigator at the National Cancer Institute

Loftfield will present the findings at NUTRITION 2024, the flagship annual meeting of the American Society for Nutrition held June 29–July 2 in Chicago.

The research drew data from more than 540,000 people who provided information about their eating habits and health in the mid-1990s, when they were between 50 and 71 years of age. Over half of the participants have since died. The researchers analyzed overall rates of death among those who were in the 90^th percentile for consumption of ultra-processed foods at baseline versus those in the 10^th percentile, and also looked at associations with specific foods and specific diseases. 

"We observed that highly processed meat and soft drinks were a couple of the subgroups of ultra-processed food most strongly associated with mortality risk and eating a diet low in these foods is already recommended for disease prevention and health promotion," said Loftfield. The Dietary Guidelines for Americans recommend limiting sugar-sweetened beverages and processed meats such as hot dogs, sausages and deli meat.

For this study, researchers used multiple strategies to classify the level of processing for various food items. This included breaking down food frequency questionnaire data into particular food and ingredient types in addition to incorporating expert consensus to categorize dietary components according to a rubric known as the NOVA classification system.

The researchers also accounted for other factors that can increase a person's risk of death, such as smoking and obesity. They found that people who consumed more ultra-processed foods also tended to have higher body mass index and a lower Healthy Eating Index score (a measure of diet quality based on how closely a person's diet aligns with the Dietary Guidelines for Americans). However, the analysis showed that the associations between ultra-processed food consumption and increased mortality were not explained by these variables, as the associations between higher ultra-processed food intake and mortality risk persisted among people categorized as having better or worse diet quality as well as among those classified as normal weight or obese.

One caveat is that the study design did not allow researchers to determine causality. In addition, Loftfield noted that the U.S. food supply and dietary preferences have changed considerably since the study's baseline data was collected in the mid 1990s, underscoring the importance of continued research to further elucidate the relationships between food processing and human health.

Loftfield will present this research at 12:45-1:45 p.m. CDT on Sunday, June 30, during the Nutritional Epidemiology poster session in McCormick Place (abstract; presentation details).