Despite a decade of successful clinical trials for stroke prevention, substantial gaps exist in the application and implementation of this information in community practice. The frequency of guideline use is low, and there remains controversy regarding the standard of practice. Patients with stroke may have multiple risk factors and concomitant stroke mechanisms, factors that are not addressed in stroke clinical trials and guideline statements. New guidelines are needed to account for these complexities and to provide primary care physicians a practical means to achieve stroke prevention. We sought to develop guidelines that can be implemented by primary care physicians to enhance the use of medical and surgical measures for recurrent stroke prevention. We sought to test the applicability of current evidence-based guidelines to daily practice with routine and complex patient case scenarios to determine whether these could be simplified into a more easily applied form for primary care physicians. We used RAND/UCLA Appropriateness Methodology to develop guidelines for the use of interventions supported by randomized controlled trials including carotid revascularization, anticoagulant therapy, antiplatelet therapy, and blood pressure management for the prevention of recurrent stroke. After a systematic literature review of randomized clinical trials we developed a comprehensive list of indications or clinical scenarios to capture decision making. A diverse multidisciplinary panel reviewed and rated each indication according to the RAND Appropriateness Method. First, panelists rated each scenario (1-3 for inappropriate, 4-6 for uncertain, and 7-9 for appropriate) without interaction with other panelists. "Appropriate" was defined as the expected health benefit exceeding its expected negative consequences by a sufficient margin. At a formal interactive session, panelists re-rated all indications. Overall carotid endarterectomy was rated as appropriate when there was 50% to 99% ipsilateral symptomatic carotid artery stenosis, inappropriate with <50% or 100% stenosis (total occlusion), and uncertain when the surgical risk was high. Carotid angioplasty was generally rated as of uncertain value. When there was atrial fibrillation, anticoagulation with warfarin was rated as appropriate when there was a low bleeding risk but of uncertain value when the bleeding risk was high. For patients who were not candidates for warfarin therapy, aspirin, aspirin plus extended-release dipyridamole, or clopidogrel were all rated as appropriate initial therapies. Ticlopidine was considered inappropriate and aspirin plus clopidogrel of uncertain value. With the exception of ticlopidine and aspirin, persons with a prior cerebral ischemic event while on aspirin could receive any of the aforementioned antiplatelet agents or combinations and be considered appropriately treated. The panelists rated a blood pressure of <130/80 mm Hg at 1 year after ischemic stroke as the target level and rated any of the following agents as appropriate initial therapies if there was no diabetes mellitus or proteinuria: diuretics, beta-blockers, angiotensin-converting enzyme inhibitors, angiotensin-converting enzyme receptor blockers, or combinations of a diuretic and an angiotensin-converting enzyme inhibitor or angiotensin receptor blocker. Patient risk played a significant role in deterring the panel from recommending certain therapies; however, the presence of atrial fibrillation or large or small cerebral vessel syndromes rarely had significant influence on treatment decisions. Appropriateness was less where bleeding or surgical risk was excessive. Using consensus evidence from clinical trials, we have developed recurrent stroke prevention guidelines for routine and more complex patient scenarios according to appropriateness methodology. Broad application of these guidelines in primary practice promises to reduce the burden of recurrent stroke.