Ultra-processed food (UPF) consumption has increased dramatically over the last decades worldwide. Although it has been linked to some cardiometabolic comorbidities, there is limited evidence regarding kidney function. This study aimed to cross-sectionally and longitudinally assess the association between UPF consumption and estimated-glomerular filtration rate (eGFR) based on Cystatin C (CysC).Older adults (mean age 65 ± 5.0 years, 46% women) with overweight/obesity and metabolic syndrome (MetS) who had available data of CysC at baseline (n = 1909), at one-year and at 3-years of follow-up (n = 1700) were analyzed. Food consumption was assessed using a validated 143-item semi-quantitative food frequency questionnaire and UPF consumption (% of g/d) at baseline and changes after one-year of follow-up were estimated according to NOVA classification system. Multivariable-adjusted linear and logistic regression models were performed to evaluate the cross-sectional associations between UPF consumption with eGFR levels and decreased kidney function (eGFR <60 ml/min/1.73 m2) at baseline. Multivariable-adjusted mixed-effects linear regression models were fitted to investigate the associations between one-year changes in UPF and eGFR over 3-years of follow-up.Individuals with the highest baseline UPF consumption showed lower eGFR (β: -3.39 ml/min/1.73 m2; 95% CI: -5.59 to -1.20) and higher odds of decreased kidney function (OR: 1.64; 95% CI: 1.21 to 2.22) at baseline, compared to individuals in the lowest tertile. Participants in the highest tertile of one-year changes in UPF consumption presented a significant decrease in eGFR after one-year of follow-up (β: -1.45 ml/min/1.73 m2; 95% CI: -2.90 to -0.01) as well as after 3-years of follow-up (β: -2.18 ml/min/1.73 m2; 95% CI: -3.71 to -0.65) compared to those in the reference category.In a Mediterranean population of older adults with overweight/obesity and MetS, higher UPF consumption at baseline and one-year changes towards higher consumption of UPF were associated with worse kidney function at baseline and over 3-years of follow-up, respectively.ISRCTN89898870.