The primary study objective is to evaluate if personalized treatment by decision support, based on the biomarker-based risk prediction (ABC-scores) guided strategy, reduces the occurrence of the composite outcome of stroke or death in patients with atrial fibrillation. Approximately 6500 patients will be randomized 1:1 to ABC risk score guided therapy or standard care.

Inclusion criteria:

1. Patients with diagnosis of atrial fibrillation, including newly diagnosed, with or without current oral anticoagulant treatment
2. Signed informed consent

Exclusion criteria:

1. Contraindication for any oral anticoagulant, according to the SmPC (in Sweden FASS)
2. Indication for oral anticoagulant treatment beyond atrial fibrillation, e.g. venous thromboembolism and/or mechanical heart valve prosthesis
3. Currently on treatment with a non-vitamin K antagonist oral anticoagulant (NOAC) and not eligible for change of NOAC drug e.g. drug-drug interactions
4. Concomitant dual antiplatelet treatment
5. Acute coronary syndrome (myocardial infarction or unstable angina) within the last 30 days
6. Participation in anti-thrombotic pharmaceutical trial
7. Planned for AF ablation or AF surgery
8. Haemoglobin <90 g/L
9. Patients who, in the opinion of the investigator, cannot or will not comply with the requirements of the protocol