The purpose of this study was to perform a meta-analysis comparing the effectiveness and safety of anticoagulation to antiplatelet therapy for the prevention of thromboembolic events in patients with atrial fibrillation (AF).MEDLINE, Cochrane, EMBASE, and Google Scholar databases were searched for studies published through May 31, 2014. Randomized controlled trials comparing anticoagulants (warfarin) and antiplatelet therapy in patients with AF were included. The primary outcomes were the rates of stroke and systemic embolism. Secondary outcomes included the rates of hemorrhage/major bleeding and death. Pooled odds ratios (ORs) and 95% confidence intervals (CIs) were calculated.Nine reports of 8 trials that enrolled 4363 patients (2169 patients received anticoagulation and 2194 antiplatelet therapy) were included. All of the studies compared adjusted-dose warfarin or with aspirin, and the majority of the patients were >70 years of age. Anticoagulants were titrated to an international normalized ratio (INR) of 2.0 to 4.5, and aspirin was administered at a dosage of 75 to 325 mg/d. Death occurred in 206 participants treated with an anticoagulant and 229 participants treated with antiplatelet therapy. There was no significant difference in the overall stroke rate between the groups (OR = 0.667, 95% CI 0.426–1.045, P = 0.08); however, patients with nonrheumatic AF (NRAF) treated with an anticoagulant had a lower risk of stroke (OR = 0.557, 95% CI 0.411–0.753, P < 0.001). Anticoagulants were associated with a lower risk of embolism (OR = 0.616, 95% CI = 0.392–0.966, P = 0.04), and this finding persisted in patients with NRAF (OR = 0.581, 95% CI 0.359–0.941, P = 0.03). No significant difference in the rate of hemorrhage/major bleeding was noted (OR = 1.497, 95% CI 0.730–3.070, P = 0.27), and this finding persisted on subgroup analysis.Anticoagulants appear to be more effective than aspirin in preventing embolisms in patients with AF, as the risk of bleeding is not increased.