Electronic physician notifications to improve guideline-based anticoagulation in atrial fibrillation: a randomized controlled trial.

backgroundOral anticoagulants reduce the risk of stroke in patients with atrial fibrillation. However, many patients with atrial fibrillation at elevated stroke risk are not treated with oral anticoagulants.

objectiveTo test whether electronic notifications sent to primary care physicians increase the proportion of ambulatory patients prescribed oral anticoagulants.

designRandomized controlled trial conducted from February to May 2017 within 18 practices in an academic primary care network.

participantsPrimary care physicians (n = 175) and their patients with atrial fibrillation, at elevated stroke risk, and not prescribed oral anticoagulants.

interventionPatients of each physician were randomized to the notification or usual care arm. Physicians received baseline email notifications and up to three reminders with patient information, educational material and primary care guidelines for anticoagulation management, and surveys in the notification arm. MAIN MEASURES: The primary outcome was the proportion of patients prescribed oral anticoagulants at 3 months in the notification (n = 972) vs. usual care (n = 1364) arms, compared using logistic regression with clustering by physician. Secondary measures included survey-based physician assessment of reasons why patients were not prescribed oral anticoagulants and how primary care physicians might be influenced by the notification. KEY

resultsOver 3 months, a small proportion of patients were newly prescribed oral anticoagulants with no significant difference in the notification (3.9%, 95% CI 2.8-5.3%) and usual care (3.2%, 95% CI 2.4-4.2%) arms (p = 0.37). The most common, non-exclusive reasons why patients were not on oral anticoagulants included atrial fibrillation was transient (30%) or paroxysmal (12%), patient/family declined (22%), high bleeding risk (20%), fall risk (19%), and frailty (10%). For 95% of patients, physicians stated they would not change their management after reviewing the alert.

conclusionsElectronic physician notification did not increase anticoagulation in patients with atrial fibrillation at elevated stroke risk. Primary care physicians did not prescribe anticoagulants because they perceived the bleeding risk was too high or stroke risk was too low.

trial registrationClinicalTrials.gov identifier NCT02950285.