摘要：Objectives. We evaluated trends and disparities in stroke death rates for American Indians and Alaska Natives (AI/ANs) and White people by Indian Health Service region. Methods. We identified stroke deaths among AI/AN persons and Whites (adults aged 35 years or older) using National Vital Statistics System data for 1990 to 2009. We used linkages with Indian Health Service patient registration data to adjust for misclassification of race for AI/AN persons. Analyses excluded Hispanics and focused on Contract Health Service Delivery Area (CHSDA) counties. Results. Stroke death rates among AI/AN individuals were higher than among Whites for both men and women in CHSDA counties and were highest in the youngest age groups. Rates and AI/AN:White rate ratios varied by region, with the highest in Alaska and the lowest in the Southwest. Stroke death rates among AI/AN persons decreased in all regions beginning in 2001. Conclusions. Although stroke death rates among AI/AN populations have decreased over time, rates are still higher for AI/AN persons than for Whites. Interventions that address reducing stroke risk factors, increasing awareness of stroke symptoms, and increasing access to specialty care for stroke may be more successful at reducing disparities in stroke death rates. Stroke is the fourth leading cause of death for all persons in the United States and the seventh leading cause of death among American Indian and Alaska Native (AI/AN) persons. 1,2 In 2010, stroke caused 559 deaths among AI/AN persons. 1 Recent National Vital Statistics System (NVSS) data have typically reported stroke death rates for AI/AN persons to be lower than those for the overall population and non-Hispanic White people. 1 However, the prevalence of stroke is higher among the AI/AN population than among any other racial/ethnic group (i.e., in 2010, 5.9% of AI/AN persons and 2.4% of non-Hispanic Whites had a history of stroke). 3 In addition, when compared with other racial groups, AI/AN persons report high prevalence of risk factors for stroke, including diabetes, obesity, current smoking, and physical inactivity. 4–7 AI/AN populations have a younger mean age at time of death caused by stroke than White populations. 7,8 Among younger age groups, stroke incidence 9 and mortality (based on NVSS data) 10 are higher among AI/AN adults than among non-Hispanic White adults. On the basis of NVSS data for 1995 to 1998, stroke death rates for those aged 25 to 64 years were higher for AI/AN persons than for non-Hispanic White persons, 10 and the Strong Heart Study found a higher incidence of stroke among AI/AN persons than non-Hispanic White persons among those aged 45 to 74 years. 9 In addition, the Strong Heart Study found higher case fatality rates among their AI/AN participants aged 45 to 74 years at enrollment; 1-year case fatality rates were 1.5 times the rates reported in other studies of majority White participants. 9 Though NVSS data show lower overall stroke death rates for AI/AN persons than for non-Hispanic White persons, when restricted to areas served by the Indian Health Service (IHS), stroke death rates among AI/AN men aged 35 to 74 years were comparable with rates for all races (i.e., IHS AI/AN men = 73.6 per 100 000 population; US men, all races = 66.8 per 100 000 population) and significantly higher than those for women (IHS AI/AN women = 66.7 per 100 000; US women, all races = 51.0 per 100 000). 11 One explanation for the lower stroke death rates reported by national vital event data is that misclassification of AI/AN persons in the national mortality data sets may result in underestimated death rates for the group. 12,13 In this study, we address this issue by using a data set for which IHS patient data have been linked to the NVSS database to provide more accurate classification of AI/AN individuals. In addition, geographic disparities in stroke death 11,14,15 and risk factors for stroke 14,16 have been reported for AI/AN persons. Therefore, we examine regional disparities in stroke death by stratifying these unique data according to IHS region and restricting analyses to IHS Contract Health Services Delivery Area (CHSDA) counties.
