Current evidence suggests that beta-blocker lower the risk of development of atrial fibrillation (AF) and in-hospital stroke after cardiac surgery. This study was to assess whether beta-blockers could decrease incidence of new-onset AF in patients with end stage renal disease (ESRD). We identified patients from a nation-wide database called Registry for Catastrophic Illness, which encompassed almost 100% of the patients receiving dialysis therapy in Taiwan from 1995 to 2008. Propensity score matching and Cox's proportional hazards regression model were used to estimate hazard ratios (HRs) for new-onset AF. Among 100066 patients, 41.7% received beta-blockers. After a median follow-up of 1500 days, the incidence of new-onset AF significantly decreased in patients treated with beta-blockers (HR = 0.483, 95% confidence interval = 0.437-0.534). The prevention of new-onset AF was significantly better in patients taking longer duration of beta-blockers therapy (P for time trend <0.001). The AF prevention effect remains robust in subgroup analyses. In conclusion, beta-blockers seem effective in the primary prevention of AF in ESRD patients. Hence, beta-blockers may be the target about upstream treatment of AF.