Atrial fibrillation (AF) is suggested to be less common among black and Asian individuals, which could reflect bias in symptom reporting and access to care. In the Asymptomatic AF and Stroke Evaluation in Pacemaker Patients and the AF Reduction Atrial Pacing Trial (ASSERT), patients with hypertension but no history of AF had AF recorded via an implanted pacemaker or defibrillator, thus allowing both symptomatic and asymptomatic AF incidence to be determined without ascertainment bias. The ASSERT enrolled 2,580 patients in 23 countries in North America, Europe, and Asia. AF was defined as device-recorded AF episodes >190/min, lasting either for >6 minutes or >6 hours in duration. All ethnic groups with >50 patients were enrolled. Ethnic groups studied include Europeans (n = 1900), black Africans (n = 73), Chinese (n = 89), and Japanese (n = 105) patients. Compared to Europeans, black Africans had more risk factors for AF such as heart failure (27.8 vs 14.6%) and diabetes (41.7 vs 26.3%). At 2.5 years follow-up, all 3 non-European races had a lower incidence of AF (8.3%, 10.1%, and 9.5% vs 18.0%, respectively, for AF>6 minutes, P < 0.006). When adjusted for baseline difference, Chinese had a lower incidence of AF > 6 minutes (P < 0.007), and Japanese and black Africans had a lower incidence of AF > 6 hours (P < 0.04 and P = 0.057, respectively). Black Africans, Chinese, and Japanese had lower incidence of AF compared to Europeans. In the case of black Africans, this is despite an increased prevalence of AF risk factors.