Incidence of Stroke in the Aboriginal and Non-Aboriginal Populations of Australia: A Data Linkage Study

Abstract

BACKGROUND: Most estimates of stroke incidence among Aboriginal and Torres Strait Islander (hereinafter Aboriginal) Australians are confined to single regions and include small sample sizes. We aimed to measure and compare stroke incidence in Aboriginal and non-Aboriginal residents across central and western Australia. METHODS: Whole-population multijurisdictional person-linked data from hospital and death datasets were used to identify stroke admissions and stroke-related deaths (2001–2015) in Western Australia, South Australia, and the Northern Territory. Fatal (including out-of-hospital deaths) and nonfatal incident (first-ever) strokes in patients aged 20-84 years were identified during the 4-year study period (2012–2015), using a 10-year lookback period to exclude people with prior stroke. Incidence rates per 100 000 population/year were estimated for Aboriginal and non-Aboriginal populations, age-standardized to the World Health Organization World Standard population. RESULTS: In a population of 3 223 711 people (3.7% Aboriginal), 11 740 incident (first-ever) strokes (20.6% regional/remote location of residence; 15.6% fatal) were identified from 2012 to 2015, 675 (5.7%) in Aboriginal people (73.6% regional/remote; 17.0% fatal). Median age of Aboriginal cases (54.5 years; 50.1% female) was 16 years younger than non-Aboriginal cases (70.3 years; 44.1% female; P <0.001), with significantly greater prevalence of comorbidities. Age-standardized stroke incidence in Aboriginal people (192/100 000 [95% CI, 177–208]) was 2.9-fold greater than in non-Aboriginal people (66/100 000 [95% CI, 65–68]) aged 20-84 years; fatal incidence was 4.2-fold greater (38/100 000 [95% CI, 31–46] versus 9/100 000 [95% CI, 9–10]). Disparities were particularly apparent at younger ages (20–54 years), where age-standardized stroke incidence was 4.3-fold greater in Aboriginal people (90/100 000 [95% CI, 81–100]) than non-Aboriginal people (21/100 000 [95% CI, 20–22]). CONCLUSIONS: Stroke occurred more commonly, and at younger ages, in Aboriginal than non-Aboriginal populations. Greater prevalence of baseline comorbidities was present in the younger Aboriginal population. Improved primary prevention is required. To optimize stroke prevention, interventions should include culturally appropriate community-based health promotion and integrated support for nonmetropolitan health services.