There were fewer primary events in patients assigned to warfarin than in those assigned to aspirin (24, 1·8% per year vs 48, 3·8% per year). Risk of stroke rose with age from 2% (age 75-79) to 2.8% (age 85 and older) in the warfarin group and from 2.8% to 5.6% in the aspirin group. There were no significant differences between the groups for hemorrhages with 1.1% annual risk (age 75-79) in the warfarin group, compared to .8% in the aspirin group and 2.9% (age 85 and older) compared to 3.7%. Secondary endpoints did not differ significantly, with 8% risk per year of mortality in the warfarin group, compared to 8.8% in the aspirin group, and 6.1% risk of all non-vascular events compared to 6.3% in the aspirin group. There was a slightly lower risk of vascular events in the warfarin group. Of the 488 randomised to warfarin treatment, 326 (67%) remained on the treatment with 124 of the 127 that did not continue it, taking aspirin and the other 3 put on clopidogrel.