Methods:

The authors used Swedish national registries to link hospitalization for ischemic stroke, including discharge medications, with sociodemographic information. The study population included patients with AF who had their first ischemic stroke between January 2009 and December 2012, and were not treated with an OAC at the time of the stroke. The primary outcome was prescription of an OAC at discharge, dichotomized to yes/no. OACs included warfarin, dabigatran, rivaroxaban, and apixaban. Level of education and disposable income were used to define socioeconomic status. Country of birth was categorized as Swedish, Nordic (other than Sweden), European (other than Nordic countries), and other. Age, living situation prestroke, activities of daily living (ADL) status, level of consciousness on admission, discharge disposition, and medical comorbidities (smoking, congestive heart failure, diabetes, hypertension, and vascular disease) were also captured. The associations between prescription of an OAC and patient characteristics were analyzed by multivariable logistic regression, controlling for year of stroke, sex, age, country of birth, education, income, living alone, dependency in ADLs, discharge disposition, level of consciousness on admission, and medical comorbidities.