To determine whether a comprehensive evaluation and customized multilevel educational interventions will increase the rate of use of oral anticoagulants and the adherence and persistence of use in patients with atrial fibrillation. Our hypothesis is that there will be differences in the use of oral anticoagulants and the persistence in patients between the control and interventional group. There will be a greater change in the use of oral anticoagulants over one year in the cohort in the intervention sites than the control sites. This will be due to two factors: greater initiation of oral anticoagulants among patients not treated at baseline and greater persistence of treatment for those on treatment at baseline

Inclusion Criteria:

• 1. Patients over the age of 18 with paroxysmal, persistent or permanent atrial fibrillation who have a 12-lead ECG showing atrial fibrillation and/or more than one rhythm strip showing atrial fibrillation at least 2 weeks apart, not due to a reversible cause (like post-CABG))
• 2. At least one of the traditional CHADS2 risk factors or at least two CHA2DS2 VASc risk factors

Exclusion Criteria:

• Mechanical prosthetic valve
• Clinically unstable at the time of enrollment (ie, with ongoing shock)
• Terminal illness and/or comfort care
• Unable to provide consent (e.g. severe cognitive impairment)
• Patients unable to have one year of follow-up for any reason
• Clear and absolute contraindication to oral anticoagulation (for example, active bleeding or recent life-threatening bleeding such as ICH)