Background: Atrial fibrillation (AF) is associated with a risk for cognitive impairment and dementia, which is more pronounced in patients with a history of clinical stroke. Anticoagulation use and efficacy impact long-term risk of dementia in AF patients in observational trials. Methods: The Cognitive Decline and Dementia in Patients with Non-Valvular Atrial Fibrillation (CAF) Trial was a randomized, prospective, open-label vanguard clinical study with blinded endpoint assessment involving patients with moderate- to high-risk (CHADS2 or CHA2DS2-Vasc scores of ≥2) non-valvular AF assigned to dabigatran etexilate or warfarin. The primary endpoint was incident dementia or moderate cognitive decline at 24 months. Results: A total of 101 patients were enrolled, of which 50 received dabigatran and 51 warfarin anticoagulation. The mean age was 73.7±6.0 years and 54(53.5%) were male. Prior stroke and stroke risk factors were similar between groups. Average INR over the study was 2.41±0.68 in the warfarin group. No patient experienced a stroke or developed dementia. Mini-Mental Status Evaluation, Hachinski Ischemic scale, cognitive subscale of the Alzheimer’s Disease Assessment Scale, Disability Assessment for Dementia, Quality of Life Improvement as assessed by Minnesota Living with Heart Failure Scale and the Anti-Clot Treatment Scale Quality of Life Survey scores did not vary at baseline or change over 2 years. Biomarker analysis indicated a similar efficacy of anticoagulation strategies Conclusion: Use of dabigatran and well-managed warfarin therapy were associated with similar risks of stroke, cognitive decline, and dementia at 2 years, suggestive that either strategy is acceptable to mitigate these risks. The results of this Vanguard study did not support the pursuit of a larger formally powered study.