In this large cohort of stroke patients with detailed socio-demographic and clinical information, we found substantial differences in risk factors and stroke characteristics among those identified as East Asian, South Asian and White ethnic groups. However, processes of stroke care delivery and stroke outcomes were similar among ethnic groups.
Our finding of more frequent and more severe ICH in East Asian patients compared to White or South Asian patients is consistent with previous studies [11–13, 22]. It has been speculated that higher rates of hemorrhagic stroke in East Asian patients is related to a greater prevalence of hypertension and a greater propensity to experience hemorrhagic stroke at lower systolic blood pressure levels than other Western populations [23, 24]. In our study, both a pre-stroke history of hypertension and baseline systolic blood pressure on presentation were similar among ethnic groups, while alcohol intake – another potential risk factor for hemorrhagic stroke – was lower in East Asian compared to other ethnic groups. South Asian patients with ICH were 10 years younger on average compared to White or East Asian patients, but tended to have strokes of lesser severity than those seen in White or East Asian patients. This younger age at presentation with ICH in the South Asian patients is consistent with other stroke studies [7, 25]. Our study found that baseline risk factors that could be associated with premature strokes, such as hypertension, dyslipidemia and diabetes, were more common in South Asian compared to other ethnic groups. However, other risk factors for ICH were paradoxically favorable in South Asian patients –fewer comorbid conditions overall, less smoking, and lower alcohol intake.
In our cohort with ischemic stroke, we also found that South Asian patients had a significantly higher prevalence of diabetes, dyslipidemia, and hypertension compared with other ethnic groups. This constellation of disproportionately elevated vascular risk factors is consistent with other studies of heart disease and ischemic stroke in South Asian populations [7–10, 25–27]. Our findings of a higher prevalence of atrial fibrillation, alcohol consumption, and tobacco use in White patients with ischemic stroke are also consistent with other reports [1, 7].
Ensuring a high standard of pre-hospital and hospital stroke care regardless of ethnicity is a critical goal for health care systems. In our study, all ethnic groups studied had significant delays in hospital arrival raising concern for a lack of recognition of stroke symptoms, failure to understand stroke as an emergency, or poor access to transportation. Despite differences in socioeconomic status, South Asian and East Asian patients were just as likely or more likely to receive appropriate emergency and in-hospital stroke management as well as secondary stroke prevention. East Asian patients were more likely admitted to the ICU. This was likely due to their higher occurrence of hemorrhagic strokes and lower levels of consciousness on admission. These findings are in contrast to two single center studies that found South Asian patients were less likely to receive lipid lowering therapy and East Asian patients were less likely to receive antiplatelet agents [7, 10]. The lack of ethnic differences in quality of care in our study may reflect inter-regional/ inter-institutional differences or that our cohort was derived from multiple specialized stroke centers within an organized regional system of stroke care and in the context of a universal health care system.
Despite differences in baseline risk factors, short term and long term prognosis after ICH and ischemic stroke was generally similar among the identified ethnic groups. This extends the work by previous groups that only examined short- term outcomes in East Asian patients  and death certificate analyses for hemorrhagic stroke . Previous studies reported conflicting findings with one study observing increased 30-day mortality in South Asian patients with ischemic stroke , and two studies demonstrating lower 28 day mortality in South Asian patients for all stroke  and lower nine-month mortality in Asian patients with ischemic stroke compared to White patients . However, these studies had limited adjustment for key prognostic factors or did not separate different types of Asian patients despite significant differences in major Asian subgroups.
The strengths of this study include the rigorous case ascertainment of stroke using established and uniform definitions, the multi-center, multi-ethnic cohort and an extensive adjustment for confounding prognostic factors. However, there are several limitations to be noted. First, the stroke registry only had ethnicity recorded for 60% of patients and thus, these results need to be confirmed in population-based studies. However, this analysis still represents one of the largest clinically detailed evaluations of ethnic differences in stroke care and long term prognosis in these populations. Second, our cohort comprised patients seen at urban tertiary stroke centers with neurologist and CT scanning capability therefore, our results may not be generalized to settings without these capabilities. Finally, we were unable to adjust for year of immigration or generation status. However, given that the largest Asian immigration waves in Canada were in the last 35 years , our cohort is likely mostly comprised of first generation immigrants.