Abstract
OBJECTIVES: To provide greater understanding of the "real world" effect of anticoagulation on stroke risk over several years. DESIGN: Cohort study. SETTING: Anticoagulation and Risk Factors in Atrial Fibrillation Study community-based cohort. PARTICIPANTS: Adults with nonvalvular atrial fibrillation (AF) between 1996 and 2003 (13,559). MEASUREMENTS: All events were clinician adjudicated. Extended Cox regression with longitudinal warfarin exposure was used to estimate cause-specific hazard ratios (HRs) for thromboembolism and the competing risk event (all cause death). The Fine and Gray subdistribution regression approach was used to estimate this association while accounting for competing death events. As a secondary analysis, follow-up was limited to 1, 3, and 5 years. RESULTS: The rate of death was much higher in the group not taking warfarin (8.1 deaths/100 person-years (PY)) than in the group taking warfarin (5.5 deaths/100 PY). The cause-specific HR indicated a large reduction in thromboembolism with warfarin use (adjusted HR = 0.57, 95% confidence interval (CI) = 0.50-0.65), although this association was substantially attenuated after accounting for competing death events (adjusted HR = 0.87, 95% CI = 0.77-0.99). In analyses limited to 1 year of follow-up, with fewer competing death events, the results for models that did and did not account for competing risks were similar. CONCLUSION: Analyses accounting for competing death events may provide a more-realistic estimate of the longer-term stroke prevention benefits of anticoagulants than traditional noncompeting risk analyses for individuals with AF, particularly those who are not currently treated with anticoagulants.