Abstract MP56: Intake of Ultra-Processed Foods is Associated With an Increase in Risk for Type Two Diabetes: Results From Three U.S. Cohort Studies and a Meta-Analysis

Introduction: There is limited evidence on the association between long-term consumption of ultra-processed foods (UPF) and the risk of type 2 diabetes (T2D), among the U.S population. The overall strength of this association has also not been established. Hypothesis: Higher intake of UPF is associated with a higher risk of T2D in U.S. adults. The pooled risk estimates from published literature reinforce the positive relationship between the UPF intakes and T2D. Methods: We first assessed this relationship among 71,871 women from the Nurses’ Health Study (NHS, 1984-2016), 87,918 women from NHSII (1991-2017), and 38,847 men from the Health Professionals Follow-up Study (HPFS, 1986-2016) who were all free of T2D at baseline. Diet was assessed using food frequency questionnaires, every 2-4 years. UPF were categorized according to the Nova classification. Information on incident cases of T2D was obtained through follow-up questionnaires every 2 years. The association between UPF intake and incident T2D was examined using Cox proportional hazards models. Second, after conducting a systematic review of prospective cohort studies, risk estimates from all included cohorts were pooled in a random-effects, dose-response, meta-analysis to assess nonlinearity of the association between total UPF intake and T2D risk. Finally, the strength of the meta-evidence was assessed using NutriGrade. Results: During 5,187,678 person-years of follow-up across the three cohorts, 19,503 T2D cases were documented. The pooled multivariable-adjusted hazard ratios (HRs) for T2D between the extreme quintiles of total UPF intake (% of grams/day), was 1.36 (95% confidence interval (CI): 1.29, 1.44; P trend <0.0001). This relationship was driven by intakes of ultra-processed animal-based products, ready-to-eat mixed dishes and artificially- and sugar-sweetened beverages. Ultra-processed cereals and ultra-processed dark breads and whole-grain breads were inversely associated with T2D risk. In the meta-analysis (7 risk estimates, 415,554 participants and 21,932 T2D cases), a significant positive dose-response association between total UPF intake and T2D was observed (P=0.90 for non-linearity): a 10% increase in total UPF intake (% grams from UPF/day) was associated with a 10% higher risk of T2D (95%CI: 8%, 12%; I 2 =23.1%; P heterogeneity =0.25). Per NutriGrade, the evidence supporting the positive relationship between total UPF intake and T2D was of high quality. Conclusions: High quality evidence shows that total UPF consumption is associated with higher risk of T2D, although not all individual foods classified as ultra-processed were associated with a higher risk in these U.S. cohorts.