Ultra-Processed Food Consumption and Risk of Type 2 Diabetes: Three Large Prospective U.S. Cohort Studies
Bibliographic record
Abstract
OBJECTIVE: We examined the relationship between ultra-processed food (UPF) intake and type 2 diabetes (T2D) risk among 3 large U.S. cohorts, conducted a meta-analysis of prospective cohort studies, and assessed meta-evidence quality. RESEARCH DESIGN AND METHODS: We included 71,871 women from the Nurses' Health Study, 87,918 women from the Nurses' Health Study II, and 38,847 men from the Health Professional Follow-Up Study. Diet was assessed using food frequency questionnaires and UPF was categorized per the NOVA classification. Associations of total and subgroups of UPF with T2D were assessed using Cox proportional hazards models. We subsequently conducted a meta-analysis of prospective cohort studies on total UPF and T2D risk, and assessed meta-evidence quality using the NutriGrade scoring system. RESULTS: Among the U.S. cohorts (5,187,678 person-years; n = 19,503 T2D cases), the hazard ratio for T2D comparing extreme quintiles of total UPF intake (percentage of grams per day) was 1.46 (95% CI 1.39-1.54). Among subgroups, refined breads; sauces, spreads, and condiments; artificially and sugar-sweetened beverages; animal-based products; and ready-to-eat mixed dishes were associated with higher T2D risk. Cereals; dark and whole-grain breads; packaged sweet and savory snacks; fruit-based products; and yogurt and dairy-based desserts were associated with lower T2D risk. In the meta-analysis (n = 415,554 participants; n = 21,932 T2D cases), each 10% increment in total UPF was associated with a 12% (95% CI 10%-13%) higher risk. Per NutriGrade, high-quality evidence supports this relationship. CONCLUSIONS: High-quality meta-evidence shows that total UPF consumption is associated with higher T2D risk. However, some UPF subgroups were associated with lower risk in the U.S. cohorts.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.009 | 0.013 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.010 |
| Bibliometrics | 0.002 | 0.004 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".