A cross-sectional analysis published in the American Journal of Medicine by researchers at Florida Atlantic University contributes to the rapidly expanding literature on ultra-processed food (UPF) consumption and cardiometabolic health. Utilizing data from the National Health and Nutrition Examination Survey (NHANES) collected between 2021 and 2023, the study examined 4,787 American adults aged 18 and older, finding that those in the highest quartile of UPF consumption had a 47% elevated risk of major adverse cardiovascular events (MACE), including myocardial infarction and cerebrovascular accident.
The study's analytical framework employs the NOVA food classification system, which categorizes foods not by nutrient content but by the nature and extent of industrial processing. NOVA Group 4 — ultra-processed foods — encompasses products formulated predominantly or entirely from industrial ingredients and substances derived from foods, with little if any intact Group 1 (unprocessed or minimally processed) food. Characteristic ingredients include hydrogenated fats, modified starches, hydrolyzed proteins, high-fructose corn syrup, emulsifiers, humectants, flavor enhancers, and cosmetic additives. This classification paradigm, developed by Carlos Monteiro's group at the University of São Paulo, has become the dominant framework in nutritional epidemiology for studying processing-related health effects, though it remains controversial.
The epidemiological context is striking: ultra-processed foods now constitute approximately 58% of total energy intake among American adults and 67% among children and adolescents, according to recent NHANES analyses. The U.S. represents an extreme case globally, though UPF consumption is rising rapidly across all income levels worldwide.
A methodological strength of this study is its comprehensive multivariate adjustment model. The observed 47% risk elevation persisted after controlling for age, sex, race/ethnicity, body mass index, smoking status, physical activity level, prevalent diabetes mellitus, diagnosed hypertension, and annual household income. This extensive covariate adjustment is important for addressing confounding — the concern that people who consume more UPF may differ systematically from those who do not in ways that independently affect cardiovascular risk.
Several mechanistic pathways have been proposed to explain the UPF-cardiovascular disease association: (1) sodium load — most UPFs are sodium-dense, contributing to chronic hypertension and endothelial dysfunction; (2) added sugars — particularly fructose — promote hepatic de novo lipogenesis, insulin resistance, and visceral adiposity; (3) chemical additives — emulsifiers such as carboxymethylcellulose and polysorbate-80 have been shown in animal models to disrupt the intestinal mucus barrier and alter gut microbiome composition, potentially promoting systemic inflammation; (4) nutrient displacement — high UPF consumption mechanically reduces intake of cardioprotective micronutrients, fiber, and phytochemicals found in whole foods; (5) advanced glycation end-products (AGEs) — formed during industrial processing at high temperatures, AGEs promote oxidative stress and vascular damage.
However, the study has methodological limitations that warrant critical scrutiny. First, the dietary exposure assessment relied on a single 24-hour dietary recall, which introduces substantial within-person variability and potential misclassification. Habitual diet is more accurately estimated with multiple days of dietary assessment or food frequency questionnaires, though these have their own biases. A single-day recall may be particularly problematic for UPF assessment because eating patterns can vary significantly by day of the week, social context, and seasonal availability.
Second, the cross-sectional design precludes causal inference. The temporal relationship between UPF exposure and cardiovascular outcomes cannot be established — it is conceivable that individuals with existing cardiovascular disease or risk factors modify their diets in ways that affect UPF consumption (reverse causality). Prospective cohort studies with repeated dietary assessments are necessary to establish temporal precedence.
Third, the NOVA classification itself has been subject to substantive criticism. By categorizing foods based on processing degree rather than nutritional composition, it groups nutritionally heterogeneous products: a zero-calorie artificially sweetened beverage, a whole-grain fortified breakfast cereal, and a sugar-laden candy bar all fall into NOVA Group 4. This heterogeneity may obscure differential health effects within the UPF category and could lead to overly broad public health recommendations that fail to distinguish between more and less harmful products.
Fourth, residual confounding remains a possibility despite extensive covariate adjustment. Socioeconomic status is measured imprecisely by household income alone; education level, food environment access, health literacy, and psychosocial stress — all of which correlate with both UPF consumption and cardiovascular risk — were not fully accounted for.
Despite these caveats, the study is consistent with a robust and convergent body of evidence. Recent meta-analyses encompassing prospective cohort studies across multiple countries have demonstrated dose-response relationships between UPF consumption and cardiovascular disease incidence, cardiovascular mortality, type 2 diabetes, obesity, depression, and all-cause mortality. The Bradford Hill criteria for causation — strength of association, consistency across studies, biological plausibility, temporal precedence (in prospective designs), and dose-response — are increasingly satisfied for the UPF-cardiovascular disease relationship, even if individual studies cannot prove causation.
The researchers advocate for multi-level public health interventions: mandatory front-of-pack labeling that communicates processing level, fiscal policies (taxes on UPFs, subsidies for whole foods), regulation of food additives with emerging safety concerns, and reformulation incentives for industry. These policy recommendations, while outside the scope of the study's data, reflect the broader consensus emerging from nutritional epidemiology.