Ultra-processed foods and type 2 diabetes mellitus incidence in RaNCD project: A prospective cohort study

Parsa Amirian,Mahsa Zarpoosh,Yahya Pasdar

DOI: https://doi.org/10.1101/2024.05.27.24307997

2024-05-28

Abstract:Background: Following rapid population growth and urbanization, global ultra-processed food consumption levels have increased. Additionally, type 2 diabetes mellitus, a non-communicable disease, is affecting one-tenth of the people worldwide. In this study, we aimed to investigate the association between ultra-processed food consumption and the risk of type 2 diabetes mellitus in different scenarios in a prospective cohort study in the western part of Iran. Methods: The RaNCD cohort includes 10047 participants aged 35 to 65; the main phase commenced in March 2015; we included participants susceptible to diabetes at enrolment with follow-up data. We used the widely accepted NOVA classification to define ultra-processed foods. A multivariable Cox proportional hazards regression model was used as the main model; furthermore, the Cox model with different adjustments and the logistic regression model were used as sensitive analysis to evaluate the association between ultra-processed foods consumption and type 2 diabetes mellitus. Results: A total of 8827 participants with a mean age of 46.92y, a mean follow-up time of 7.1y, and a mean daily ultra-processed food intake of 87.69 grams were included. During the follow-up phases, we included 255 incidences of type 2 diabetes mellitus cases. After adjusting for cofounders in the primary model, including age, gender, residence type, socioeconomic status, physical activity, body mass index, and familial history of diabetes despite the elevated hazard ratio of 1.08 (0.75, 1.55) in the fourth quartile compared to the first quartile, the P-value was insignificant (p-value = 0.665); p for trend in the UPF quartiles was also insignificant. Conclusion: Our study has shed light on the association between ultra-processed food consumption and the risk of type 2 diabetes mellitus in the Middle East region. We have identified significant associations between diabetes incidence and some covariates. However, further investigations are necessary to confirm or refute the UPFs/T2DM association.