Patients with severely reduced renal function have been excluded from randomized controlled trials of oral anticoagulation in atrial fibrillation (AF). Warfarin treatment in this population is controversial and data on anticoagulation control and the impact on adverse outcomes are needed. By individual-level linkage of nationwide registries, we identified all patients discharged from hospitals with AF in Denmark between 1997 and 2011. Patients with available serum creatinine tests were categorized according to the estimated glomerular filtration rate (eGFR). Time in therapeutic range (TTR) was calculated using the Rosendaal method. The risk of stroke and bleeding was estimated using multivariable Cox regression analyses with eGFR and TTR estimated time dependently throughout follow-up. We identified 10,423 warfarin-treated AF patients with available international normalized ratio and creatinine tests; 5,527 with eGFR > 60 mL/min/1.73 m2, 4,524 with eGFR 30–60 mL/min/1.73 m2 and 372 with eGFR < 30 mL/min/1.73 m2. Median TTR was 66.7, 61.2 and 49.7% in patients with eGFR > 60, 30–59 and <30 mL/min/1.73 m2, respectively. A TTR < 70% was associated with a higher risk of stroke/thromboembolism (hazard ratio [HR]: 1.39; 95% confidence interval [CI]: 1.20–1.60) and bleeding (HR: 1.22; 95% CI: 1.05–1.42) among patients with eGFR of 30 to 59 and a trend towards higher risk of stroke/thromboembolism (HR: 1.24; 95% CI: 0.86–1.80) and bleeding (HR: 1.17; 95% CI: 0.83–1.65) among patients with eGFR < 30 mL/min/1.73 m2. In conclusion, warfarin-treated AF patients with reduced renal function have suboptimal anticoagulation control which was related to the risk of adverse outcomes.