Abstract Introduction To study the association of antibiotics with ischaemic stroke/systemic embolism (IS/SE) in patients with atrial fibrillation (AF) receiving anticoagulants. Methods This retrospective population-based cohort study applied electronic health data from a territory-wide medical database. Data were collected for patients with AF who were prescribed non-vitamin K antagonist oral anticoagulants (NOAC) or warfarin between Jan 1, 2010 and Dec 31, 2018. We further identify the patients with and without use antibiotics (any classification of antibiotics with prescription period within 7 days) during the NOAC/warfarin treatment period. The primary outcome is IS/SE and the secondary outcomes including hemorrhagic stroke (HS) and all-cause mortality. Patients were followed up for a median of 3.7 years. A self-control case crossover design was adopted to compare the short-term association between use of antibiotics and outcomes of interest among NOAC and warfarin group. Results This study included 40221 patients, consisted of 30617 NOAC users and 9826 warfarin users, while 9604 patients in NOAC group and 3325 patients in warfarin group had concurrent antibiotics use. At baseline, 46.1% of patients in NOAC group were male and the mean (SD) age was 77.4 (9.8) years, and 56.2% of patients in warfarin group were male and the mean (SD) age was 73.8 (11.3) years. Short-term use of antibiotics is associated with increased risk of IS/SE, HS and all-cause mortality in both NOAC and warfarin groups (NOAC vs. warfarin group: IS/SE: hazard ratio [HR]:3.41 [95%CI, 2.78-4.18]; HR: 2.51 [95%CI, 2.00-3.14]; HS: HR: 2.22 [95%CI, 1.78-2.78]; HR: 2.53 [95%CI, 1.86-3.43]; mortality: HR: 1.96 [95%CI, 1.85-2.08]; HR: 1.87 [95%CI, 1.73-2.02]; all P < 0.001). Conclusion The findings of the present study suggest that the short-term antibiotics use was associated with an increased risk of IS/SE, HS and all-cause mortality among patients in both NOAC and warfarin group. These findings further suggest that use of antibiotics may indicate a higher risk of IS/SE, HS and death for patients with AF, even when receiving NOAC/warfarin and that further investigation is warranted.The study design.The odds ratios of different outcomes.