Reviews/Ultra Processed Foods/Ultra-processed food raises cardiovascular disease risk
Cardiometabolic · UPF-CM-01 · v2
Across several separate populations, people who ate the most ultra-processed food tended to have more heart disease, and the direction of this pattern is remarkably consistent [22][23][25]. Supporting studies show that eating more ultra-processed food worsens measurable risk markers such as blood lipids, blood pressure and weight [6][3][26][5]. The weakness is that the studies measuring actual heart attacks and strokes are all observational, so they can show a link but cannot prove ultra-processed food causes the disease rather than simply going along with a generally poorer diet and lifestyle. The strongest direct study is a snapshot in time, which cannot show that the diet came before the disease [22]. Because the direction is well supported but the causal proof is incomplete, the panel treated the claim as probable rather than firmly established.
Three independent datasets link higher ultra-processed food intake to higher cardiovascular or coronary disease risk: a US cross-sectional sample [22] (relative risk 1.47; 95% CI 1.06-2.04), an Iranian prospective cohort of incident coronary heart disease [23] (HR 1.57; 95% CI 1.02-2.43), and a 20-year Whitehall II cohort using a metabolomic signature [25] (HR 1.19; 95% CI 1.02-1.39). Beneath these outcomes sits a coherent mechanistic chain: longitudinal and randomized feeding evidence of worsening cardiometabolic risk factors [6], higher childhood blood lipids [3], higher diastolic blood pressure [26], and less weight loss on an ultra-processed diet in a controlled crossover trial [5]. Every hard-endpoint study is observational; the randomized evidence reaches only intermediate markers. The contested point: The division is over causality versus association: the directional consistency across cohorts is not disputed, but whether ultra-processed food independently causes cardiovascular events or merely marks a poorer overall diet is unresolved. [23] shows high adherence to Mediterranean or DASH patterns confers lower coronary risk even at high ultra-processed intake, suggesting part of the signal reflects overall diet quality.
All 27 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from Ultra Processed Foods, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
No study in the corpus is a randomized trial with a cardiovascular-event endpoint. All hard-outcome evidence ([22] cross-sectional, [23] and [25] observational cohorts) is susceptible to residual confounding; the causal 'increases' rests on surrogate-marker RCTs ([5], [6], [3], [26]) and an unproven surrogate-to-event link.
The single most direct CVD study [22] is cross-sectional, so temporality and reverse causation cannot be excluded; it cannot establish that UPF intake preceded disease.
Where it remains uncertain. No trial in the corpus measures a cardiovascular event, the direct outcome effects are modest with wide confidence intervals [23][25], and diet-quality confounding is evident [23].
Aleatory
Hard-outcome estimates are modest with confidence intervals nearly reaching unity ([23] 1.02-2.43; [25] 1.02-1.39), so sampling variability leaves the magnitude imprecise.
The load-bearing papers from a corpus of 27, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Dose-responsive adverse lipid effect, supporting an atherogenic pathway from early life.
“children in the highest tertile of UPF consumption at age 3 years had higher levels of total cholesterol (TC; β 0·22 mmol/l; 95 % CI 0·04, 0·39) and TAG at age 6 years”
Panel auditInternal validity: medium · scope of inference: Blood lipid levels in 308 young Brazilian children; a CVD risk-factor proxy.Key limitations: children only · lipids not CVD events · 24-h recall · low-income single community
Large longitudinal link between UPF and worsening measured cardiometabolic risk factors (BP, triglycerides, glucose).
“Higher UPF consumption was associated with adverse evolution in objectively measured CMR factors after 12 months of follow-up in adults with metabolic syndrome.”
Panel auditInternal validity: medium · scope of inference: Cardiometabolic risk-factor evolution over 12 months in 5373 older adults with metabolic syndrome.Key limitations: risk factors not CVD events · 12-month follow-up · metabolic syndrome population · FFQ diet measure
Direct hard-endpoint (MI or stroke) association in a US adult sample; 47% higher risk in highest UPF quartile.
“In adjusted analyses, individuals in the highest quartile compared with those in the lowest, had a statistically significant higher risk of cardiovascular disease (relative risk = 1.47; 95% confidence interval = 1.06-2.04; P = 0.02).”
Panel auditInternal validity: low · scope of inference: Prevalent MI/stroke and UPF intake in 4,787 US adults; cannot establish that UPF preceded disease.Key limitations: cross-sectional (no temporality) · prevalent not incident CVD · limited covariates · single diet survey
Randomized feeding trial showing a causal adiposity effect of UPF under matched dietary guidance; supplies causal weight on an intermediate.
“significantly greater %WC on MPF (Δ%WC, -1.01 (95% CI, -1.87, -0.14), P = 0.024; Cohen's d, -0.48 (95% CI, -0.91, -0.06))”
Panel auditInternal validity: medium · scope of inference: Weight change over 8 weeks in 50 adults with high habitual UPF intake; a risk-factor outcome, not CVD.Key limitations: weight not CVD · 8-week exposure · both arms lost weight · n=50
4 of 27 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
Consistent and independent associations with hard CVD outcomes plus causal RCT evidence on surrogates support a probable claim, but the absence of event-level randomized data and the fragility of the observational effect sizes keep it below the Established band.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
No randomized cardiovascular-event trial. All hard-outcome evidence is observational [22][23][25]; the causal claim depends on surrogates that may not translate to events. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A long-duration randomized controlled feeding or dietary-advice trial powered for incident myocardial infarction and stroke.
Open question: Does reducing ultra-processed food intake lower the rate of cardiovascular events, not just risk markers? Why it is still open: Every corpus study reaching hard endpoints is observational [22][23][25]; randomized evidence stops at surrogates [5][6].
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 21 Oct 2026 · protocol v0.19