Project Summary/Abstract Atrial Fibrillation (AF) increases the risk of stroke 5-fold and accounts for roughly 15% of all strokes in the United States. Many individuals may have undiagnosed AF whose arrhythmia does not prompt evaluation either because of minimal symptoms or brief episodes. Oral anticoagulation (OAC) is highly effective at reducing stroke risk in patients with AF, but is only prescribed to individuals with recognized disease. Identifying subjects with undiagnosed AF is important so they may be treated with OAC and prevent strokes from occurring. This project will help advance our knowledge of screening patients for undiagnosed AF in the primary care setting by identifying patients at high risk of developing AF for targeted clinic and home-based screening strategies. Fundamental questions of who to target and how to implement a screening program in the United States remain unanswered. To address these gaps in knowledge, the principal investigator (PI) proposes a career development program that blends rigorous methodologic and content area training with an innovative research agenda. This plan has three scientific objectives: 1) To identify predictors of AF incidence from electronic health record structured and free-text data, 2) To develop and implement an electronic algorithm that improves upon existing models to identify patients at increased risk for AF in a primary care population, and 3) To implement and evaluate feasibility, acceptability, appropriateness, and usability of a clinic and home-based screening program to identify undiagnosed AF. This project aligns with several objectives of the National Heart, Lung, and Blood Institute’s Strategic Vision by implementing novel diagnostic tools to diagnose AF and prevent strokes, by optimizing clinical and implementation research to improve health and reduce disease, and leveraging emerging opportunities in data science, through the use of natural language processing of text data in the electronic health record, to improve identification of patients at high risk for AF. The long-term goal of this career development award is to establish the PI as an independent researcher with expertise in cardiovascular disease and targeting populations for prevention of cardiovascular disease events. Career development activities include training in biomedical informatics, data mining and risk prediction, survey research design, implementation research, and cardiovascular disease pathophysiology and clinical management through formal coursework, clinical shadowing, as well as mentorship by an exceptionally qualified team of senior scientists. Successful completion of this career development proposal will improve prediction of which patients are at the highest risk of developing AF, and evaluate how to use this risk information to implement an effective screening strategy in primary care. Pilot data from this award will lay the groundwork for a highly competitive application for NIH R01 funding.