Background Rivaroxaban, apixaban and dabigatran are non-vitamin K antagonist oral anticoagulants (NOACs) that are widely used for treatment or prevention of venous thromboembolism and stroke in patients with atrial fibrillation. Objective To estimate and compare hemorrhagic events report of rivaroxaban, apixaban and dabigatran. Setting FDA Adverse Event Reporting System (FAERS) database. Methods The reporting odds ratio (ROR) was used to assess the signal of hemorrhagic events of different NOACs. Main outcome measure The overall hemorrhagic events and hemorrhagic events in different physiological systems. Results From January 1, 2014 to December 31, 2019, the total number of reports of hemorrhage related to rivaroxaban was 53,085, and the numbers of apixaban and dabigatran were 13,151 and 14,100 respectively. The overall ROR (95% CI) of hemorrhagic events reporting for rivaroxaban versus dabigatran and apixaban versus dabigatran were 1.58 (1.54-1.62) and 0.47 (0.46-0.48) respectively. The ROR (95% CI) for rivaroxaban versus dabigatran in gastrointestinal system, nervous system, renal and urinary system, skin and subcutaneous tissue, and eye system was 1.38 (1.34-1.42), 0.94 (0.90-0.98), 1.07 (1.01-1.13), 0.80 (0.70-0.90), and 1.38 (1.19-1.60) respectively. The RORs (95% CI) for apixaban versus dabigatran in gastrointestinal system, nervous system, renal and urinary system, skin and subcutaneous tissue, and eye system were 0.28 (0.27-0.29), 0.69 (0.66-0.73), 0.31 (0.29-0.34), 0.98 (0.86-1.12), and 1.18 (1.00-1.39), respectively. Conclusions Overall, we found a moderate signal of higher frequency of reporting hemorrhage in rivaroxban compared with dabigatran and decreased hemorrhagic event reporting in apixaban compared with dabigatran. While this potential signal has not been confirmed in clinical trials or observational studies, in clinical practice, attention should be paid to the risk of potential hemorrhage when the patients switch from apixaban to dabigatran or rivaroxban.