Nine servings of ultra-processed food per day raised cardiac risk 67% - and the toll was steepest for Black Americans

Maria ate the same number of calories as her neighbor. Her overall diet quality score was similar. She did not have diabetes, and her blood pressure and cholesterol were under control. But Maria averaged nine servings of ultra-processed foods each day - frozen dinners, processed deli meats, sweetened cereals, packaged crackers - while her neighbor ate closer to one. Over the years that followed, Maria's risk of suffering a heart attack, stroke, or cardiac death was 67% higher.

That is the central finding of a study being presented at the American College of Cardiology's Annual Scientific Session (ACC.26), and its most unsettling feature is not the size of the risk but its independence. The association persisted after controlling for total calorie consumption, overall diet quality, and the usual suspects of cardiovascular risk: diabetes, hypertension, high cholesterol, and obesity. Something about ultra-processed foods themselves - beyond their caloric content and nutrient profile - appears to contribute to heart disease.

The MESA cohort: diversity matters

The research draws on data from 6,814 adults enrolled in the Multi-Ethnic Study of Atherosclerosis (MESA), a long-running cohort study that deliberately recruited a racially and ethnically diverse population of US adults between ages 45 and 84 who were free of known heart disease at enrollment. Participants completed food questionnaires that researchers used to assess daily ultra-processed food intake based on the NOVA classification system, which categorizes foods into four groups ranging from unprocessed to ultra-processed.

The diversity of the MESA cohort is what gives this study particular value. Most previous research on ultra-processed foods and cardiovascular risk was conducted in European populations. This is among the first large studies to examine the association in a population that reflects the demographic reality of the United States.

The dose-response curve

Participants in the highest quintile of ultra-processed food intake consumed an average of 9.3 servings per day. Those in the lowest quintile averaged 1.1 servings. The 67% increased risk of a major adverse cardiac event in the highest quintile compared with the lowest is a headline number, but the dose-response relationship may be more informative: each additional daily serving of ultra-processed food was associated with a 5.1% increased risk of adverse cardiac events overall.

That per-serving risk increase was not uniform across racial groups. Among Black Americans, each additional daily serving was associated with a 6.1% increase in risk, compared with 3.2% among non-Black participants. The disparity likely reflects structural factors rather than biological ones - minority-targeted food marketing, food desert geography, and neighborhood-level barriers to accessing less-processed alternatives have been extensively documented as drivers of dietary inequality.

Beyond calories and nutrients

The finding that ultra-processed food intake predicts cardiac risk independent of calorie consumption and overall diet quality raises a mechanistic question: what is it about these foods that causes harm beyond their nutritional composition?

Previous research has pointed to several candidates. Ultra-processed foods tend to be energy-dense and engineered for overconsumption, with combinations of sugar, fat, and salt that override normal satiety signals. They often contain additives - emulsifiers, artificial sweeteners, colorings - whose long-term cardiovascular effects are poorly understood. Processing itself can create compounds (such as advanced glycation end products) that promote inflammation. And the high glycemic loads typical of ultra-processed foods may drive metabolic dysfunction through pathways that standard nutrient analysis does not capture.

The current study did not investigate specific mechanisms, but its findings add weight to the argument that food processing methods deserve attention alongside nutrient content in dietary recommendations.

Reading labels, changing access

Lead author Amier Haidar, MD, a cardiology fellow at the University of Texas Health Science Center at Houston, suggests that one practical step is paying more attention to food labels. Ultra-processed foods typically contain higher amounts of added sugar, sodium, fat, and refined carbohydrates per serving compared with less-processed alternatives. Choosing plain oatmeal over sweetened cereal, or fresh produce over frozen convenience meals, can reduce ultra-processed food intake without requiring a complete dietary overhaul.

The ACC has endorsed standardized front-of-package labeling to make healthier choices more visible. Whether labeling alone can shift consumption patterns in communities where ultra-processed foods are the most affordable and accessible option is debatable - but it represents one component of a broader effort to address dietary contributors to cardiovascular disease.

Limitations worth noting

The MESA study was not designed specifically to collect ultra-processed food data. The dietary assessment relied on self-reported food questionnaires and classified intake by number of servings rather than individual food items. Self-reported dietary data are notoriously imprecise, and the NOVA classification system - while widely used - involves judgment calls about where to draw the line between processing categories. The study was also published simultaneously in JACC Advances.

These are observational data, so causation cannot be established. People who eat more ultra-processed food may differ from those who eat less in ways the researchers could not fully measure or control. But the consistency of the findings across multiple cardiac endpoints, combined with the dose-response relationship and the robustness of the association after adjustment for known confounders, makes the signal difficult to dismiss.