The purpose of this study was to assess the association between reduced apixaban dose and two outcomes: ischemic stroke/systemic embolism (SE) and major bleeding. We performed a retrospective cohort study within the database of the largest healthcare provider in Israel. We identified all patients diagnosed with atrial fibrillation, who started apixaban treatment between 2013 and 2017. Apixaban users were classified into three dosing regimen groups based on their renal function, age, and weight: standard dose (5 mg b.i.d.), adjusted reduced dose (2.5 mg b.i.d.), and underdosing (2.5 mg b.i.d.). Patients were followed through 2018 for the occurrence of stroke/SE and major bleeding. Of the 27,765 included patients, 13,141 (47%) adequately received standard apixaban dose, 4,739 patients (17%) received adjusted reduced dose, and 9,885 patients (36%) were classified as underdosed. The CHA2DS2-VASc score adjusted hazard ratio for ischemic stroke/SE was 1.1 (95% confidence interval (CI), 0.83-1.43) in the adjusted reduced dose group, and 1.04 (95% CI, 0.83-1.35) in the underdosing group, compared with the standard apixaban dose group. The HAS-BLED score adjusted hazard ratio for any major bleeding was 1.66 (95% CI, 1.32-2.09) in the adjusted reduced dose group, and 1.51 (95% CI, 1.24-1.83) in the underdosing group, compared with apixaban in the standard dose group. Results were similar for major gastrointestinal bleeding and intracranial hemorrhage separately. We conclude that underdosing with apixaban is very common, and may not disproportionately elevate the risk of ischemic stroke. However, albeit halving the dose, patients treated with reduced apixaban dose (adjusted or underdosing) seem to be at higher risk for major bleeding.