Background and aims: Whether coffee consumption is associated with changes in estimated glomerular filtration rate (eGFR) is unknown. We investigated the relationship between coffee consumption and annual eGFR change in a large Dutch population-based study. Methods and results: This study was performed in 78,346 participants without chronic kidney disease (CKD) in the population-based Lifelines Cohort Study. Coffee consumption was assessed at baseline using food frequency questionnaires. Outcomes were annual eGFR change and a composite kidney outcome (defined as eGFR <60 mL/min per 1.73 m2 or >20 % eGFR decline). Multivariable linear and logistic regression analyses were used to evaluate the associations of coffee consumption (categories and cups/day) with kidney outcomes. Overall, 90 % of the participants drank coffee daily and 36 % drank >2–4 cups/day. Unadjusted mean ± SD annual eGFR change ranged from −2.86 ± 2.96 (for non-coffee drinkers) to −2.35 ± 2.62 (for participants consuming >6 cups/day) mL/min per 1.73 m2. During 3.6 ± 0.9 years follow-up, 11.1 % of participants reached the composite kidney outcome. As compared to non-coffee drinkers, higher coffee consumption was associated with less annual eGFR decline in multivariable models (β [95 % CIs] ranged from 0.15 [0.07, 0.22] for >0–2 cups/day to 0.29 [0.20, 0.38] for >6 cups/day, P-trend <0.001). Consumption of one more cup of coffee per day was associated with a 3 % lower risk of the composite kidney outcome (OR [95%CI], 0.97 [0.96, 0.99]). The inverse association was more pronounced in a subgroup of individuals with diabetes. Conclusion: Coffee consumption was inversely associated with annual eGFR change and CKD risk in a large Dutch population-based cohort.