摘要：Objectives. We evaluated heart disease death rates among American Indians and Alaska Natives (AI/ANs) and Whites after improving identification of AI/AN populations. Methods. Indian Health Service (IHS) registration data were linked to the National Death Index for 1990 to 2009 to identify deaths among AI/AN persons aged 35 years and older with heart disease listed as the underlying cause of death (UCOD) or 1 of multiple causes of death (MCOD). We restricted analyses to IHS Contract Health Service Delivery Areas and to non-Hispanic populations. Results. Heart disease death rates were higher among AI/AN persons than Whites from 1999 to 2009 (1.21 times for UCOD, 1.30 times for MCOD). Disparities were highest in younger age groups and in the Northern Plains, but lowest in the East and Southwest. In AI/AN persons, MCOD rates were 84% higher than UCOD rates. From 1990 to 2009, UCOD rates declined among Whites, but only declined significantly among AI/AN persons after 2003. Conclusions. Analysis with improved race identification indicated that AI/AN populations experienced higher heart disease death rates than Whites. Better prevention and more effective care of heart disease is needed for AI/AN populations. Heart disease is the leading cause of death and a major cause of medical expenditures in the United States. 1,2 Among American Indians/Alaska Natives (AI/ANs), death rates for heart disease have been consistently equal to or lower than that of the overall US population and non-Hispanic Whites, especially among older age groups (75 years and older) as reported by National Vital Statistics System (NVSS) data. 3 However, from 1997 to 2011, the self-reported prevalence of heart disease in the National Health Interview Survey was either higher or similar among AI/AN persons compared with other populations. 3 Moreover, the most recent Behavioral Risk Factor Surveillance System data showed that prevalence of coronary heart disease for 2006 to 2010 was much higher among AI/AN populations than other groups. 4 In addition, the Strong Heart Study found that the incidence of coronary heart disease events in participating AI/AN communities was 2 times higher than estimates from other large cohort studies using similar case ascertainment methods. 5 Available data also indicated that the prevalence of risk factors for heart disease and stroke—including smoking, 6,7 physical inactivity, 7 and obesity 6,7 —among AI/AN persons was comparable to or higher than the other high-risk racial groups such as Blacks or African Americans and Native Hawaiians or other Pacific Islanders. A diabetes epidemic among AI/AN individuals in the last couple of decades also coincided with increasing prevalence of heart disease in this group. 6,8–10 In 2003, the prevalence of having 2 or more heart disease and stroke risk factors in AI/AN persons was similar to that in Blacks, but was higher than that in Whites, Hispanics, and Asians. 11 The overall age-adjusted heart disease death rate in the US population has declined steadily since the late 1950s 1 ; however, available data indicate that the decline may be slower among AI/AN persons compared with both the US general population and Whites. 3 However, the incidence of coronary heart disease may be increasing. 5 In addition, geographic disparities in mortality among AI/AN individuals 12,13 and disparities between AI/AN persons and Whites in risk factors 13,14 for heart disease have been observed. No published studies have compared the risk-adjusted case-fatality rates between AI/AN and White persons as a possible explanation for this paradox of a lower heart disease death rate among AI/AN individuals, but a higher prevalence of heart disease and its risk factors. Racial misclassification in mortality and population data, however, is a likely explanation. 15 A recent study indicated that up to 30% of AI/AN persons were misclassified or underreported on death certificates. 15,16 Given the high prevalence of diabetes in the AI/AN population, differences in the practice of coding diabetes versus heart disease as the underlying cause of death (UCOD) on the death certificate may also play a role in this paradox. 17 An accurate accounting of the disease burden and proper trend analysis is critical in making public health policy decisions and allocating resources. In the present analysis, we addressed this issue by using a dataset that links the Indian Health Service’s (IHS’s) patient registry data and the National Death Index (NDI) database to achieve more accurate racial classification of AI/AN persons. We analyzed both the UCOD and multiple causes of death (MCOD) on death records. We described disparities in heart disease death rates by age, gender, region, and year of death for AI/AN individuals compared with Whites, and evaluated whether racial misclassification might explain previous observations that heart disease mortality was similar among AI/AN and White persons, although AI/AN populations appeared to have higher prevalence of risk factors for heart disease.
