Background and Purpose Because of the paucity of published information, this report seeks to better characterize the pattern of stroke occurrence and risk factors among Native Americans in the United States.
Methods Data from the US Vital Statistics System and two National Health and Nutrition Examination Surveys were analyzed.
Results Stroke was a leading cause of death among US Native Americans in 1990. In persons aged 45 and over, stroke was the cause of 6% of deaths in Native Americans and 7% of deaths in whites. The percentage of stroke deaths due to hemorrhagic stroke was higher in Native Americans than whites. In 1988 through 1990, stroke death rates were similar in Native Americans and whites under age 65 but lower in Native Americans at ages 65 years and over. High prevalence of diabetes, smoking, and obesity may contribute to stroke mortality in Native Americans.
Conclusions Targeted research, innovative analyses of existing data, and use of ongoing surveys and the Census should be considered in the study of the epidemiology of stroke, other leading causes of death, and risk factors in Native Americans. Continued hypertension detection and treatment efforts are needed for Native Americans as for other groups. Smoking cessation and prevention should receive high priority in Native American populations.
Little information has been published about the epidemiology of stroke in native populations of North America.1 In 1973 through 1975, age-specific death rates for cerebrovascular disease were higher in Native Americans compared with all races below age 65 years and lower at age 65 and above.2 Between 1958 and 1987 in New Mexico, Native Americans had lower age-adjusted death rates from cerebrovascular disease than Hispanics or non-Hispanic whites.3 Age-adjusted rates for Native Americans in New Mexico increased between 1958 and about 1970 and declined thereafter to levels approximately 40% to 47% below those in 1958. US stroke death rates for Native Americans declined between 1969 to 1971 and 1979 to 1981, as did proportional mortality.4 Because of the paucity of available information, data from the National Center for Health Statistics were examined to better characterize the pattern of stroke occurrence and risk factors among Native Americans in the United States.
Subjects and Methods
Deaths from cerebrovascular disease were enumerated using published and unpublished data from the National Center for Health Statistics5 6 for International Classification of Diseases, 9th Revision (ICD-9) codes 430 through 438. Numbers of deaths were divided by counts or estimates of resident population obtained from the US Bureau of Census to produce death rates. The standard population for age adjustment by the direct method was the 1940 resident population of the United States.
The first National Health and Nutrition Examination Survey (NHANES I) was conducted on a nationwide, multistate probability sample of 28 043 persons.7 8 The sample was drawn from the civilian, noninstitutionalized population of the United States excluding Alaska, Hawaii, and reservation lands of Native Americans between the ages of 1 and 74 years. Of these persons, 70.4% were examined between April 1971 and June 1974. An additional sample of 3059 persons aged 25 to 74 years was selected for detailed examination between July 1974 and September 1975. At household interview, adult participants or parents or guardians of child participants were asked, “Which of the groups on this card best describes your ancestry or national origin?” The question was asked only of participants not visually identified by interviewers as black. Among persons aged 12 to 74 years, the present analysis is limited to 240 men and nonpregnant women who identified themselves as Native Americans. The composition of the sample was affected by deliberate oversampling of the poor, preschool children, women of childbearing age (20 to 44 years), and the elderly (65 to 74 years), all groups thought to be at high nutritional risk. Details of the plan, sampling, operation, and response have been published, as have procedures used to obtain informed consent and to maintain confidentiality of information obtained.7 8
Medical history and demographic and behavioral information was collected before the examination by household interview of participants or of their parents or guardians.7 8 Classification of race (white, black, and other) was made by observation. Examinations were carried out in a mobile examination center.7 8 9 10 The physician measured blood pressure with the subject seated, using a standard sphygmomanometer (only the first reading was used in this study).9 Blood samples were obtained, and frozen serum was sent to the Centers for Disease Control and Prevention for biochemical tests.10 Body mass index was computed as weight in kilograms divided by height in meters squared. Smoking history was obtained for a subsample of 20% of persons aged 25 to 74 years examined in 1971 to 1974, and for all persons in the additional sample of 1974 to 1975.7 8
Conducted in 1976 to 1980, the second National Health and Nutrition Examination Survey (NHANES II) used similar methods.11 Unlike NHANES I, the target population included Alaska and Hawaii. Of 18 209 sample persons aged 18 to 74 years, 68.7% were examined. Preschool children, the elderly (60 to 74 years), and the poor were oversampled. Details of the plan, sampling, operation, and response have been published, as have procedures used to obtain informed consent and to maintain the confidentiality of information obtained.11 Among persons aged 12 to 74 years, 226 identified themselves as Native Americans.
All descriptive statistics for risk factors were computed by standard methods.12 Data for Native Americans in this paper were not weighted to produce estimates for the reference population. Analyses were hindered by the small sample size and must be interpreted with caution. The samples did not resemble the total Native American population in several demographic characteristics. For example, in 1970 about 60% of US Native Americans lived on reservations, with 44.9% living in urban areas with a population of 2500 or more and 19.9% in central cities of standard metropolitan statistical areas (SMSAs).13 14 For NHANES I, only persons not residing on a reservation or in Alaska at the time of the survey were eligible. Furthermore, certain subgroups were oversampled as described above. Throughout this paper, the terms “Native American,” “American Indian,” and “American Indian/Alaska Native” are considered to be synonymous.
Fig 1⇓ shows age-specific death rates for cerebrovascular disease from 1988 to 1990.5 In persons under age 65 years, death rates were similar in Native Americans and whites. At ages 65 and over, Native Americans had rates that were substantially below those of whites, especially at ages 85 and over. Rates in Native American men and women were similar below age 65 but higher in men than women at ages 65 and over, especially at ages 65 to 74 and 75 to 84.
Fig 2⇓ shows age-adjusted death rates for Native Americans, blacks, and whites aged 45 years and over for 1980 to 1990.5 Age-adjusted rates declined in each sex/race group (for men: Native American, −23.6%; black, −20.5%; and white, −26.8%; for women: Native American, −16.7%; black, −23.0%; and white, −26.1%).
Cerebrovascular disease is a leading cause of death in Native Americans, ranking eighth in males and fourth in females in 1990.5 The top three causes of death in males were diseases of the heart, unintentional injuries, and malignant neoplasms. For females, they were diseases of the heart, malignant neoplasms, and unintentional injuries. The proportion of deaths from all causes that were attributed to cerebrovascular disease was slightly lower in Native Americans than whites at ages 65 years and over but slightly higher at ages 45 to 64 (Table 1⇓). The proportion of deaths due to heart disease was lower in Native Americans than in whites in both age groups. For all ages in 1989, the ratio of cerebrovascular disease to coronary heart disease deaths was 0.29 in Native Americans, 0.28 in whites, and 0.43 in blacks.
The percentage of total deaths from cerebrovascular disease assigned to various diagnostic subgroups is shown in Table 2⇓. Compared with those of whites, Native American deaths were more likely to be coded as subarachnoid hemorrhage and less likely as cerebral thrombosis. This may reflect in part the differing population age structures.
The sample of 240 persons aged 12 to 74 years in NHANES I included 43 persons aged 12 to 17 years. In this sample, 26% lived in central cities of SMSAs, and 51% lived in rural places with a population of less than 2500. The regional distribution was as follows: Northeast, 9%; Midwest, 19%; South, 45%; and West, 27%. Among whites, only 33% lived in rural areas, while the regional distribution was Northeast, 25%; Midwest, 26%; South, 22%; and West, 27%. NHANES regions are defined elsewhere.15 All but 17 self-identified Native Americans were classified white by interviewer observation. This may indicate that most were not readily identifiable as American Indians to the predominantly white interviewers. Family incomes of 23% of subjects fell below the federal poverty limit. Family incomes were less than $4000 in 25%, less than $7000 in 44%, greater than $10 000 in 29%, and greater than $15 000 in 12%. Among whites, only 13% had family incomes less than $4000. Among persons aged 18 to 64 years, 83% of 57 men reported working, the remainder not being employed; of 109 women, 41% reported working, 55% reported keeping house, and 4% reported neither activity.
The sample of 226 persons in NHANES II aged 12 to 74 years included 28 persons aged 12 to 17 years. In this sample, 19% lived in central cities of SMSAs, and 58% lived in rural places with a population of less than 2500. The regional distribution was as follows: Northeast, 6%; Midwest, 23%; South, 45%; and West, 27%. All but 26 persons were classified as white by interviewer observation (6 black and 20 other race). Family incomes of 27% of subjects fell below the Federal poverty limit compared with only 9% of whites. Family incomes were less than $4000 in 15%, $10 000 or greater in 51%, and $15 000 or greater in 27%. Among persons aged 18 to 64, 74% of 89 men with employment data reported working, the remainder not being employed; of 80 women with employment data, 43% reported working, 54% reported keeping house, and 3% reported neither activity.
Among persons aged 18 to 74 years in NHANES I, 13 (18.1%) of 72 men and 22 (17.6%) of 125 women had been told they had high blood pressure. Of those with a history of high blood pressure, 37% reported taking blood pressure medication. Of 71 men aged 18 to 74 with blood pressure data, 23 (32.3%) had systolic blood pressure (SBP) ≥140 mm Hg, and 21 (29.6%) had diastolic blood pressure (DBP) ≥90 mm Hg. Of 124 women aged 18 to 74 with complete data, 28 (22.6%) had SBP ≥140 mm Hg, and 20 (16.1%) had DBP ≥90 mm Hg.
Among persons aged 18 to 74 years in NHANES II, 26% of 198 (25% of men, 26% of women, and 41% of persons over 55) had been told they had high blood pressure. Of the 32 with a history of high blood pressure, 23 (72%) reported taking blood pressure medication. Among men, 37% had SBP of ≥140 mm Hg, and 34% had DBP of ≥90 mm Hg. The percentages of women were 23% and 20%, respectively.
Estimates of mean blood pressure by age and sex for self-identified American Indians are shown in Fig 3⇓. Sex-specific means showed males to have slightly higher DBP than females in every age group except those 55 to 64 years old. SBP tended to be higher in males up to age 44 and lower thereafter. Differences between mean blood pressures of Native Americans in NHANES II and those of whites in NHANES II were small and inconsistent (Fig 3⇓). SBP was significantly (P≤.01) correlated with age and resting pulse rate in men and with age, weight, body mass index, pulse rate, and serum cholesterol levels in women aged 18 to 74. DBP was significantly correlated with age and serum cholesterol in men and with age, weight, body mass index, and pulse in women.
Among persons aged 18 to 74 in NHANES I, 1 (1.4%) of 72 men and 6 (4.8%) of 125 women reported being told by a doctor that they had diabetes mellitus. In NHANES II, 12 (6.1%; 6.8% of men and 5.3% of women) of 198 persons aged 18 to 74 years reported a history of diabetes.
Of 72 men aged 18 to 74 years in NHANES I, 16 (22.2%) had serum cholesterol concentrations ≥240 mg/dL. Of 125 women aged 18 to 74 years, 32 (25.6%) had serum cholesterol concentrations ≥240 mg/dL. Of 198 persons aged 18 to 74 years in NHANES II, 20% (20% of men and women) had serum total cholesterol levels of ≥240 mg/dL. An additional 21% had levels of 200 to 239 mg/dL.
Mean serum cholesterol levels rose with age (Fig 4⇓). This was true in men below age 65 and women. Sex differences were inconsistent across age groups. Differences between mean serum cholesterol levels in Native Americans and whites in NHANES II were inconsistent because of the small sample size (Fig 4⇓). Serum cholesterol level was significantly correlated with age, body mass index, and DBP in men and with age, SBP, and pulse in women aged 18 to 74 (not shown).
Of 82 American Indians aged 25 to 74 years in the detailed examination subsample of NHANES I, 49 (59.8%) reported current cigarette smoking. Smoking was more prevalent in men (28 of 37, 75.7%) than in women (21 of 45, 46.7%). Male smokers reported smoking an average of 23.3 cigarettes per day compared with 17.9 for women. Overall, 70.8% of smokers smoked 20 or more cigarettes per day (77.8% of men and 61.9% of women). In NHANES II, 102 (52%) of 198 persons aged 18 to 74 reported current smoking, 63% at ages 18 to 34, 58% at ages 35 to 54, and 30% at ages 55 to 74 (61% of men and 41% of women). Among current smokers, men smoked an average of 23 cigarettes per day and women 18 cigarettes per day. These smoking rates were consistently higher than those of whites aged 18 to 74 in NHANES II (38.6% of men and 30.5% of women).
Other Risk Factors
Of 72 Native American men aged 18 to 74 years in NHANES I, 18 (25.0%) had a body mass index ≥27.8 kg/m2 and hence might be considered overweight.15 Of 125 women, 37 (29.6%) had a body mass index of ≥27.3 kg/m2 and might be considered overweight. In NHANES II, 22% of 103 men and 26% of 95 women were overweight by the same criteria. The prevalence of overweight in whites aged 18 to 74 in NHANES II was 19% in men and 21% in women. Mean body mass index by age and sex is shown in Fig 5⇓. Marked increases occurred between ages 12 to 17 and 18 to 24 (not shown). Because of the small sample of Native Americans, differences in mean body mass index between Native Americans and whites in NHANES II were inconsistent. Neither weight nor body mass index was significantly correlated with age in men or women aged 18 to 74 in NHANES I.
Of persons aged 18 to 74 years, 75% of men and 59% of women reported drinking alcohol in NHANES I. Of 93 women aged 18 to 44, 24 (25.8%) had used birth control pills in the preceding 6 months. Among persons aged 18 to 74 years in NHANES I, 57% of men and 40% of women reported high nonrecreational activity, and 26% of men and 14% of women reported high recreational activity; the remainder reported moderate or low levels. In NHANES II, 38% reported little or no exercise for recreation, and 11% reported low nonrecreational activity, similar to whites (36% and 14%, respectively).
Heart disease was the leading cause of death in Native Americans in 1969 to 1971, 1979 to 1981, and 1989 to 1991.1 2 3 4 5 In 1989 to 1991, stroke was the fourth leading cause in Native American women and the eighth in men.5 Death rates for stroke were similar in Native Americans and whites at age 45 to 64 years but lower in Native Americans at ages 65 and over. Age-adjusted stroke death rates declined between 1980 and 1990 for Native Americans, whites, and blacks. Morbidity data are lacking. Current risk factor levels do not offer a ready explanation of differences in death rates between Native Americans and whites over age 65. Racial differences in risk factors before 1970 may be important for persons over 65, although they are difficult to document.
As previously noted, mean blood pressure levels of American Indians and whites in NHANES I showed no consistent differences.2 9 16 Blood pressure rose with age in Native Americans as in whites. Compared with serum cholesterol estimates published for US whites in NHANES I, unweighted means for American Indians tended to be lower among men at all ages and similar in younger women and lower in older women.10 Serum cholesterol rose with age until age 55 to 64 in Native Americans as in whites. Smoking rates in Native Americans were higher than in whites.5 Mean body mass index was not consistently different from that of the general population sampled by NHANES I or II.15 Unlike whites, the mean body mass index of Indians in the sample was not greater at higher ages, perhaps reflecting a cohort effect. Although not shown in the figures, major increases in body mass index, blood pressure, and serum cholesterol levels were seen between the ages 12 to 17 and 18 to 24 years in young Native Americans.
Comparisons With Other Reports
In 32 reservation states in 1985 to 1987, the death rate from cerebrovascular disease in American Indians and Alaskan Natives at ages 45 to 64 years was 35.4 per 100 000, the same as the US rate for all races.17 At ages 65 and over, the rate was 296.5 per 100 000, 30% lower than the US all-races rate. These statistics are consistent with the data shown in Fig 1⇑. Stroke was the sixth leading cause of death at ages 45 to 64 and the third leading cause at ages 65 and over among Native Americans in reservation states. In 1987, the age-adjusted stroke mortality rate for the Indian Health Service area population (American Indian/Alaskan Native population in counties served by Indian Health Service facilities) was 33.1 per 100 000 for all areas; it was 36.8 per 100 000 when three areas with apparent underreporting of Indian race on death certificates were excluded.18 This rate was 21% higher than the US all-races rate of 30.3 per 100 000. The service area rate is consistently higher than the rate in the reservation states, perhaps reflecting socioeconomic and health status factors as well as possible reporting differences. How this rate compared with that of the white population in the service areas is unknown. Death rates were lower in the Southwestern Indian Health Service areas than in other areas. A study of death rates of Alaskan Natives in 1980 to 1986 revealed age-adjusted stroke death rates that were slightly higher than those for nonnative Alaskans (rate ratios: male, 1.13; female, 1.03) because of a higher rate of deaths coded as cerebral hemorrhage (rate ratio, 1.36) (ICD-9 430-432).19
In New Mexico, age-adjusted death rates for cerebrovascular disease in Native Americans increased between 1958 and 1970 and declined thereafter to levels 40% to 47% below those in 1958.3 Native Americans had lower rates than Hispanics or non-Hispanic whites. However, in New York State exclusive of New York City in 1980 to 1986, Native American males and females had a standardized proportional mortality ratio of 1.40 (95% confidence limits [CL], 0.99 to 1.93) and 1.15 (95% CL, 0.84 to 1.57), respectively, for cerebrovascular disease (76 deaths) (ie, apparent higher rates in Native Americans than the total population). However, ratios for total cardiovascular disease were less than 1.0 (apparent lower rates in Native Americans than the total population).20 Although these reports are not directly comparable and confirmatory studies are needed, such a regional difference would not be inconsistent with Indian Health Service statistics.
Data on stroke incidence and prevalence in Native Americans are lacking. An estimated 24 014 Native Americans with any diagnosis of cerebrovascular disease were discharged from nonfederal, acute-care hospitals in 1987 to 1990 based on data from the National Hospital Discharge Survey. This is a minimal estimate, since in administrative records many Native Americans are misclassified as white, and federal hospitals of the Indian Health Service are not included in this survey. The rate per 100 000 Indian population of hospital discharges with stroke diagnoses that were funded by the Indian Health Service (an underestimate since it excludes Veterans Administration, self-paid, or third party–paid discharges) was only half (54%) as great as the rate of discharges with stroke diagnoses for all races from US nonfederal hospitals in 1983.21 Hospital discharge rates were lowest in the Southwestern Indian Health Service areas. A higher rate of hospitalization for bleeding intracranial aneurysms was reported for Greenlandic Eskimos than Caucasian Danes.22
Previous reviewers have suggested substantial regional and intertribal variation in stroke risk factor levels based on limited published data.16 21 Except for diabetes in the Pima, in recent decades Southwestern Indians probably had lower levels of several risk factors than non-Southwestern Indians, US whites, and US blacks, despite a high prevalence of obesity.2 16 21 Data are particularly scarce for Native Americans not living on reservations. A 1980 survey of urban American Indians in Minnesota found a higher prevalence of diabetes, cigarette smoking, and obesity compared with whites and a similar prevalence of high blood pressure and serum cholesterol levels.23 Among children, Native Americans had higher SBP than whites because of greater obesity.24 Many of these findings were consistent with those in NHANES for Native Americans not living on reservations. A few studies have suggested higher rates of hypertension and hypertensive mortality among men under age 55 in American Indians compared with whites.25 Heavy alcohol consumption in this group might play a role in producing hypertension and cardiac and cerebrovascular mortality and morbidity.21 Serum cholesterol levels in a recent study of Navaho were similar to NHANES II levels in men but lower in women over age 55.26 Although several of the leading causes of mortality and morbidity in Native Americans are nutrition related, only limited data were available on the nutrient intake of American Indians.16 27
The 1992 National Health Interview Survey estimates were consistent with a smoking rate among American Indians and Alaskan Natives over age 18 years (36.5%; 95% confidence interval, ±7.6) that was the highest of any racial group (whites, 26.2%).28 Estimates were nearly identical for men and women. This prevalence of smoking was similar to that in persons of all races living below the poverty level (37.0%). However, like the NHANES estimates, these must be used with caution because of small numbers in the sample. A telephone survey of 1055 American Indian respondents aged 18 and older in 1985 to 1988 found that the percentage of current smokers varied by region (for men: Southwest, 18.1%; Plains, 48.4%; West, 25.2%; other, 38.0%; for women: Southwest, 14.7%; Plains, 57.3%; West, 31.6%; other, 30.7%).29 Rates were consistently higher than for whites in each area except the Southwest.
Bias in comparisons of death rates for Native Americans and whites may arise because persons identified as Native American or American Indian in data from the US Census (denominator of death rates) are sometimes misreported as white on the death certificates, causing death rates to be underestimated by 22% to 30%.30 Trends in death rates must be interpreted with great caution for Native Americans because the population estimates for Native Americans increased by 45% between the decennial censuses of 1980 and 1990 due to improved enumeration techniques and the increased propensity for people to identify themselves as American Indian in the 1990 Census.5 It is not known whether a parallel change occurred in information on death certificates. Examination of proportional mortality offers important information complementary to that deriving from death rates.4 Proportional mortality analyses are appropriate because in 1989 only 3% of deaths of American Indians had nonspecific diagnoses, ie, coded to symptoms, signs, and ill-defined conditions (ICD-9 780-799).
In the 1980 US Census, an estimated 76% of 1 366 676 Native Americans lived away from reservations, including 2.8% in Alaskan Native villages, 8.2% in historic areas of Oklahoma, and 2.1% in tribal trust lands.31 Median incomes were higher and poverty rates lower among those living away from compared with on reservations. Unemployment rates and percentage of housing units lacking complete plumbing were much higher on reservations.31 Unfortunately, none of the national surveys of the National Center for Health Statistics obtained data on the American Indian population that were adequate for providing reliable population estimates for cardiovascular risk factors. Reservations were not included. In some surveys, American Indians were not specifically identified. If identified, the numbers of persons sampled were small. Nevertheless, in view of the paucity of data on cardiovascular risk factors in American Indians not living on reservations, an analysis was undertaken of data from NHANES I and II. Results of this and other analyses may suggest strategies for providing better information in the future as well as hypotheses for specialized studies of cardiovascular disease in Native American.
The limitations of this analysis of risk factor data must be stressed. Apparent differences among sample subgroups may be due to bias arising from nonresponse, sample selection procedures, and errors in reporting of ancestry, among other sources. Furthermore, in this descriptive analysis no attempt was made to control for confounding by factors other than age and sex. The small sample size resulted in large standard errors of means and low statistical power to detect differences among subgroups. The results should not be generalized to any local American Indian population. Published national estimates for whites from NHANES II were plotted in Figs 3 through 5⇑⇑⇑ to provide reference lines, even though use of weighted estimates and their standard errors for formal statistical testing for the small population subgroup of Native Americans was not considered appropriate.32 33 34 Nevertheless, in view of the paucity of data on cardiovascular risk factors in American Indians not living on reservations, such analyses may be of use to researchers. This is one of the few studies in which risk factors in Native Americans and whites in the same communities were measured using a single standardized protocol.23
At the least, data from NHANES I and II can be used as pilot study samples of over 200 persons of self-reported American Indian ancestry with large amounts of data collected on each individual. In fact, many published studies of Native Americans have had samples of similar size.26 35 Like data from pilot surveys or pretests, such samples can be used for hypothesis-generating analyses and for planning sample size and content of future surveys in nonreservation populations. Planning of future nutrition surveys in American Indians might be assisted by pilot analyses of the extensive nutrition data of the NHANES surveys, analyses which were beyond the scope of the present report. Unlike many published data, NHANES data are widely available as public-use data sets.36
Inclusion of ancestry questions and the pooling of data from multiple cycles of past and future interview and examination surveys might, over time, contribute to the knowledge of cardiovascular risk status of American Indians not living on reservations. Some health questions (eg, smoking, high blood pressure, and diabetes history) might be included for persons of American Indian or Alaskan Native ancestry in a future census as a way to obtain data on a large number of American Indians and Alaskan Natives living both on and off reservations. Determining disease prevalence and risk factor distributions in specific Native American populations requires specialized population-based surveys, such as the Strong Heart Study, and ongoing analyses of data from the Indian Health Service and the National Medical Expenditure Survey.37 38
Stroke is a leading cause of death among US Native Americans. In 1988 to 1990, stroke death rates were similar in Native Americans and whites under age 65 but lower in Native Americans at age 65 and over. A high prevalence of diabetes, smoking, and obesity may contribute to stroke mortality in Native Americans. Few studies have been published on stroke in Native Americans. Cohort studies, well-designed case control studies, and innovative analyses of national, state, and Indian Health Service databases are needed to further define the epidemiology of stroke in US Native Americans. Continued hypertension detection and treatment efforts are needed for Native Americans as for other groups. Smoking cessation and prevention should receive high priority in Native American populations.
Reprint requests to Richard F. Gillum, MD, Centers for Disease Control and Prevention, Office of Analysis, Epidemiology, and Health Promotion, National Center for Health Statistics, 6525 Belcrest Rd, Hyattsville, MD 20782.
- Received November 7, 1994.
- Revision received December 23, 1994.
- Accepted December 23, 1994.
- Copyright © 1995 by American Heart Association
Yatsu FM. Strokes in Asians and Pacific-Islanders, Hispanics, and Native Americans. Circulation. 1991;83:1471-1472.
Kuller LH. Stroke report. In: Report of the Secretary’s Task Force on Black and Minority Health, Vol 4. Cardiovascular and Cerebrovascular Diseases, Part 2. Washington, DC: US Dept of Health and Human Services; 1986:478-510.
Gillum RF. Ischemic heart disease mortality in American Indians, United States, 1969-71 and 1979-81. Am Heart J. 1988;115: 1141-1144.
National Center for Health Statistics. Health, United States, 1992. Hyattsville, Md: US Public Health Service; 1993.
National Center for Health Statistics. Vital Statistics of the United States, 1988, Vol II, Mortality, Part A. Washington, DC: Public Health Service; 1991. US Dept of Health and Human Services publication (PHS) 91-1101.
National Center for Health Statistics, Miller HW. Plan and operation of the Health and Nutrition Examination Survey, United States, 1971-73. Vital Health Stat 1. No 10a and b. Washington, DC: Health Services and Mental Health Administration; Feb 1973. US Dept of Health, Education, and Welfare publication (HSM) 73-1310.
National Center for Health Statistics, Engel A, Murphy RS, Maurer K, Collins E. Plan and operation of the HANES I Augmentation Survey of Adults 25-74 Years, United States, 1974-75. Vital Health Stat 1. No 14. Washington, DC: Public Health Service; June 1978. US Dept of Health, Education, and Welfare publication (PHS) 78-1314.
National Center for Health Statistics, Roberts J, Maurer K. Blood pressure levels of persons 6-74 years, United States, 1971-74. Vital Health Stat . No 203. Washington, DC: Health Resources Administration; Sept 1977. US Dept of Health, Education, and Welfare publication (HRA) 78-1648.
National Center for Health Statistics, Abraham S, Johnson CL, Carroll MD. Total serum cholesterol levels of adults 18-74 years, United States, 1971-74. Vital Health Stat . No 205. Washington, DC: Public Health Service; April 1978. US Dept of Health, Education, and Welfare publication (PHS) 78-1652.
National Center for Health Statistics, McDowell A, Engel A, Massey JT, Maurer K. Plan and operation of the Second National Health and Nutrition Examination Survey, 1976-80. Vital Health Stat 1. No 15. Washington, DC: Public Health Service; July 1981. US Dept of Health and Human Services publication (PHS) 81-1317.
Allen A, ed. SAS User’s Guide: Basics, 1982 Edition. Cary, NC: SAS Institute Inc; 1982:501-512.
Bureau of the Census. 1970 Census of Population, Vol 1: Characteristics of the Population. PC(1)-B1. Washington, DC: US Government Printing Office; 1972.
Bureau of the Census. 1970 Census of Population, Vol 2: Subject Reports: American Indians. PC(2)-1F. Washington, DC: US Government Printing Office; 1973.
National Center for Health Statistics, Fulwood R, Johnson C. Height and weight of adults ages 18-74 years by socioeconomic and geographic variables, United States. Vital Health Stat . No 224. Washington, DC: Public Health Service, Aug 1981. US Dept of Health and Human Services publication (PHS) 81-1674.
Kumanyika SK, Savage DD. Ischemic heart disease risk factors in American Indians and Alaska Natives. In: Report of the Secretary’s Task Force on Black and Minority Health, Vol IV. Cardiovascular and Cerebrovascular Disease, Part 2. Washington, DC: US Dept of Health and Human Services; 1986:444-473.
Indian Health Service. Indian Health Service Regional Differences in Indian Health, 1990. US Dept of Health and Human Services, Public Health Service; 1990.
Indian Health Service. Trends in Indian Health, 1990. US Dept of Health and Human Services, Public Health Service; 1990.
Mahoney MC, Ellrott MA, Michalek AM. A mortality analysis of Native Americans in New York State, 1980-86. Int J Epidemiol. 1989;18:403-412.
Welty TK, Coulehan JL. Cardiovascular disease among American Indians and Alaska Natives. Diabetes Care. 1993;16(suppl 1):277-283.
Kristensen MO. Increased incidence of bleeding intracranial aneurysms in Greenlandic Eskimos. Acta Neurochirurgica. 1983; 67:37-43.
Gillum RF, Prineas RJ, Palta M, Horibe H. Blood pressure of urban Native American school children. Hypertension. 1980;2: 744-749.
Summary Report of the Second National Conference on High Blood Pressure Control in Native American Communities. Bethesda, Md: National Institutes of Health; 1982.
Sugarman JR, Warren CW, Oge L, Helgerson SD. Using the behavioral risk factor surveillance system to monitor year 2000 objectives among American Indians. Public Health Rep. 1992;107: 449-456.
US Congress, Office of Technology Assessment. Indian Health Care. OTA-H-290. Washington, DC: US Government Printing Office; April 1986:59-82.
Drizd T, Dannenberg AL, Engel A. Blood pressure levels in persons 18-74 years of age in 1976-80 and trends in blood pressure from 1960 to 1980 in the United States. National Center for Health Statistics. Vital Health Stat 11 (234), 1986.
Carroll M, Sempos C, Briefel R, Gray S, Johnson C. Serum lipids of adults 20-74 years: United States, 1976-80. National Center for Health Statistics. Vital Health Stat 11 (242), 1993.
Najjar MF, Rowland M. Anthropometric reference data and prevalence of overweight, United States, 1976-80. National Center for Health Statistics. Vital Health Stat 11 (238), 1987.
Young TK. The Health of Native Americans. New York, NY: Oxford University Press; 1994:113-138.
Kovar MG. Data systems of the National Center for Health Statistics. National Center for Health Statistics. Vital Health Stat 1 (23), 1989.
Lee ET, Welty TK, Fabsitz R, Cowan LD, Le NA, Oopik AJ, Cucchiara AJ, Savage PJ, Howard BV. The Strong Heart Study. A study of cardiovascular disease in American Indians: design and methods. Am J Epidemiol. 1990;132:1141-1155.
Johnson A, Taylor A. Prevalence of Chronic Diseases: A Summary of Data From the Survey of American Indians and Alaska Natives. Rockville, Md: Public Health Service; 1991. National Medical Expenditure Survey Data Summary 3, Agency for Health Care Policy and Research publication 91-0031.