OBJECTIVES: According to Canada, US and European guidelines and the Helsingborg Declaration, all eligible stroke patients should receive care in specialized stroke centers. During the last decade, partly due to the Alberta Provincial Stroke Strategy (APSS), 16 stroke centers were established in Alberta. This study examined the effect of admission to stroke centers on mortality for patients with ischemic stroke, compared with admission to non-stroke centers.  METHODS: The study population was identified from the Discharge Abstract Database (DAD) from the province of Alberta, Canada. We included stoke patients with most response diagnostic code I63 (ICD10) with a first admission to acute care hospitals between April 1st 2004 and March 31st 2011. Disease specific co-morbidities were adapted from the literature review, including secondary diagnoses in the DAD. We utilized the triage information from National Ambulatory Care Report System (NACRS) as the proxy of disease severity.  The average marginal effect of stroke center on the 30-days in-hospital mortality was estimated in a bivariate probit model, using differential distance to hospitals as an instrumental variable to correct potential pre-hospital selection bias, adjusting for age, sex, co-morbidities, and disease severity. RESULTS: Among 9152 patients, 6405 (70%) were admitted to stroke centers (n=16) and 2747 (30%) to non-stroke centers. The overall unadjusted 30-day all-cause mortality rate was 9.8% for patients first admitted to stroke centers and 11.1% for patients admitted to non-designated hospitals. Adjusting patient characteristics and other factors, we found first admission to a stroke center was associated with a 6.4% (95%CI: -1.2%, -11.5%) absolute reduction in 30-day all-cause in-hospital mortality compared to non-stroke centers. CONCLUSIONS: In an observational study, we provided new evidence to support the role of stroke center on the reduction in mortality in a universal publicly funded health care system.