Ultra-Processed Foods: Consistent Signal, Observational Evidence
Higher intake of ultra-processed foods is associated with worse health outcomes in an umbrella review and a meta-analysis of observational studies (Lane 2024; Leonberg 2025); the sources report association, not cause.
Not for clinical populations without professional guidance
- Signs of disordered eating (food/weight preoccupation, restriction, bingeing, purging, compulsive exercise): see a qualified eating-disorder professional
- Diagnosed medical condition (diabetes, chronic kidney disease, heart failure, GI disease): see a registered dietitian or physician
- Pregnancy or breastfeeding: see a prenatal provider or registered dietitian
- Unintentional weight loss, difficulty swallowing, or persistent GI symptoms: seek medical evaluation
What the evidence says
The sources on this entry are an umbrella review and a meta-analysis of ultra-processed food intake, plus meta-analyses of named dietary patterns and a narrative review. Lane 2024 (umbrella review of 45 meta-analyses, 9,888,373 participants) reports greater ultra-processed food exposure associated with a higher risk of adverse health outcomes, especially cardiometabolic, common mental disorder and mortality outcomes, including 50% higher cardiovascular mortality and 12% higher type 2 diabetes; Leonberg 2025 (meta-analysis of 7 studies, 19,645 participants) reports higher intake associated with a 4–74% higher risk of incident chronic kidney disease. The remaining sources describe the patterns that limit such foods: Bakhshimoghaddam 2025 reports Nordic diet adherence associated with lower mortality and chronic disease incidence; Onvani 2017 reports high Healthy Eating Index adherence associated with lower all-cause, cardiovascular and cancer mortality; Liu 2024 reports EAT-Lancet adherence associated with lower diabetes, cardiovascular and cancer mortality; Liang 2022 reports that Mediterranean, DASH, guideline-based, plant-based and low-fat diets each reduced LDL and total cholesterol compared with a western habitual diet; Ravera 2016 (narrative review) reports a dietary-pattern approach as more effective than a nutrient-based approach for cardiovascular risk. The ultra-processed evidence is observational, and 'ultra-processed' is a broad category; the sources report associations, and this entry states no causal effect.
- 'All processed food is harmful.' Minimal processing (frozen veg, canned beans, plain yogurt) is fine, the concern is ultra-processing.
- 'UPFs have been shown to cause disease.' Lane 2024 and Leonberg 2025 report associations from observational studies, not proof of cause.
- 'Eating one processed snack ruins your diet.' Overall pattern is what matters.
- Do not claim any food, diet, or supplement cures, treats, or prevents any disease
- Do not diagnose nutrient deficiencies or medical conditions
- Do not promise specific weight-loss amounts or timelines
- Do not deliver individualized medical nutrition therapy or replace clinical care
- Agreement
- Direction recorded for 0 of 7 sources; not yet recorded for the rest.
- The claim's outcome
- Outcome not yet recorded
- Certainty
- No source states a certainty rating.
- How the sources were read
- Some sources are based on abstracts, not full texts.
What each source measured and concluded (7)
- What it measured
- Mortality, chronic disease incidence, and cardiometabolic markers
- How it was read
- based on the abstract, not the full text
- What the review says it includes
- GRADE-assessed systematic review and dose-response meta-analysis of prospective cohorts and RCTs
- Studies and participants, as stated
- 32 prospective cohort studies and 15 RCTs
- What it measured
- All-cause, cardiovascular, and cancer mortality
- How it was read
- based on the abstract, not the full text
- What the review says it includes
- Meta-analysis of observational studies
- Studies and participants, as stated
- 12 cohort studies (38 reports) and 1 cross-sectional study (3 reports)
- What it measured
- Diabetes, major CVD, all-cause mortality, and cancer
- How it was read
- based on the abstract, not the full text
- What the review says it includes
- Systematic review and meta-analysis of cohort and case-control studies
- Studies and participants, as stated
- 28 publications, over 2.21 million participants
- What it measured
- CKD incidence.
- How it measured it
- eGFR (objective).
- Its conclusion, in its own words
- “Given the substantial evidence from this systematic review and meta-analysis indicating an association between UPF and CKD, it is recommended for public health policies to address this risk.”
- How it was read
- read in full text
- What the review says it includes
- Meta-analysis of 7 studies; Systematic review and meta-analysis
- Studies and participants, as stated
- N=19,645 participants across 7 studies
- Heterogeneity, as stated
- Heterogeneity in CKD definitions and ascertainment methods
- What it measured
- Effects on NCD biomarkers (e.g., total cholesterol, LDL cholesterol, apoB, insulin resistance).
- How it measured it
- Laboratory-measured markers (objective).
- How it was read
- read in full text
- What it measured
- Adverse health outcomes (mortality, cancer, mental, respiratory, cardiovascular, gastrointestinal, and metabolic health outcomes).
- How it measured it
- Meta-analyses of observational epidemiological studies (cohort, case-control, cross-sectional).
- How it was read
- read in full text
- What the review says it includes
- Umbrella review of meta-analyses
- Studies and participants, as stated
- N=9,888,373 participants across 45 studies
- Heterogeneity, as stated
- Heterogeneity and imprecision due to equivalent treatment of different effect estimates.
- Heterogeneity, as stated
- Synthesis based on few original research articles leading to heterogeneity and imprecision.
- What it measured
- Effects of dietary interventions on prevention and modification of CVD
- How it was read
- screened only, not extracted
Every source below is shown with its study design and links to the original study. This is the audit trail.
“Abstract only” means the review read that paper’s abstract, not its full text; “screened only” means it was included at re-screening and not extracted. Both stay in the trail with their DOI, because how a source was read is part of the audit.