Background: Previous studies have shown that homocysteine and folate levels in plasma are associated with risk for cardiovascular events and mortality. The aim of this study was to investigate if plasma concentrations of total homocysteine and folate can predict major bleeding, cardiovascular events, and all-cause mortality in patients being treated with warfarin. Methods: In a longitudinal cohort study, 719 patients who were taking warfarin were followed for 3001 treatment years. The following were recorded and classified: major bleeding; cardiovascular events including stroke, arterial emboli, and myocardial infarction (MI); and mortality. Blood samples collected at baseline were analysed for plasma homocysteine and folate levels. Results: After adjustment for age, C-reactive protein, and creatinine, high homocysteine levels were associated with cardiovascular events [hazard ratio (HR) 1.23 per standard deviation (SD); 95% confidence interval (CI): 1.03-1.47], MI (HR 1.38 per SD; 95% CI: 1.03-1.85), and all-cause mortality (HR 1.41 per SD; 95% CI: 1.19-1.68). The highest tertile of folate compared to the lowest tertile was associated with decreased risk for both cardiovascular events (HR 0.64; 95% CI: 0.43-0.91) and MI (HR 0.45; 95% CI: 0.21-0.97). There was no association between major bleeding and homocysteine or folate levels. Conclusions: In patients receiving warfarin treatment, high homocysteine and low folate plasma concentrations are associated with increased risk for cardiovascular events but not major bleeding. For homocysteine levels, there is also an association with all-cause mortality.