Summary In patients with atrial fibrillation (AF) and an intermediate risk of stroke (CHADS2 score =1), available evidence from clinical trials is inconclusive and the present guidelines for the management of AF indicate that the choice between oral anticoagulant and aspirin in these patients is open. Our goal was to evaluate whether, in patients with AF and only one moderate risk factor for thromboembolism, treatment with an oral anticoagulant is appreciably more beneficial than treatment with an antiplatelet agent. Among 6,517 unselected patients with AF, 1,012 of them (15.5%) had a CHADS2 score of 1 and were liable to treatment with an antiplatelet agent or an anticoagulant. An oral anticoagulant was prescribed for 606 patients (59.9%) and an antiplatelet agent or no antithrombotic treatment for 406 (40.1%). During follow-up (median=793 days, interquartile range=1,332 days), 105 deaths (10.4%) and 19 strokes (1.9%) were recorded. The administration of an anticoagulant was associated with a lower rate of events (relative risk=0.42, 95% confidence interval 0.29–0.60, p<0.0001) than when no anticoagulant was prescribed. Results remained similar after adjustment for age and other confounding factors. In contrast, prescription of an antiplatelet agent was not associated with a lower risk of events. Factors independently associated with an increased risk of events were older age (p<0.0001), concomitant heart failure (p=0.0002), diabetes (p=0.0025), lack of prescription of an anticoagulant (p<0.0001) and permanent AF (p=0.04). Thus, prescription of an anticoagulant is independently associated with a decreased risk of death or stroke among patients with AF and a CHADS2 score =1.