HERBERT W.NICKENS 1947–1999
The health section of America Becoming: Racial Trends and Their Consequences is dedicated to the memory of Herbert W.Nickens, M.D., M.A., who passed away on March 22, 1999. In his all too brief lifetime, Dr. Nickens devoted his academic and professional interests toward improving the health status of racial and ethnic minorities in the United States. After serving on the staff of the landmark Secretary’s Task Force on Black and Minority Health, Dr. Nickens became the first director of Department of Health and Human Services’ Office of Minority Health. In that role, he was pivotal in crafting the programmatic themes for that office—many of which continue to this day. As the first Vice President for Community and Minority Programs for the Association of American Medical Colleges (AAMC), Dr. Nickens devoted much of his professional energy to increasing the number of underrepresented minority applicants to medical school. His passion for social justice brought forth the Project 3000 by 2000, which reignited the institution’s long-standing commitment to increasing minority representation in medicine. Dr. Nickens’s interests did not, however, focus solely on the educational pipeline. He continued to maintain his focus on eliminating minority health disparities in his scholarly writings, such as the co-authorship of one of the chapters in this report, and in AAMC programming, such as the Health Services Research Institute for Minority Medical School Faculty. Herbert W.Nickens will be remembered for many things—but most importantly for his pursuit of equity in health.
Racial and Ethnic Differences in Health: Recent Trends, Current Patterns, Future Directions
Raynard S.Kington and Herbert W.Nickens
The volume of research addressing racial and ethnic differences in health has grown dramatically in recent decades as more and better data have become available. Health results from the interaction of intrinsic genetic and biological factors with environmental and social factors distributed throughout every facet of life. Health also influences every facet of life: it affects one’s ability to work, to socialize, to think, to learn, to communicate, to reproduce. Because of the complex causal pathways and widespread influences, no single discipline is sufficient to address the problem of racial and ethnic differences in health status.
All of the behavioral and social sciences—anthropology, psychology, sociology, economics, and demography—have contributed to the understanding of racial and ethnic health differentials. The public health and medical sciences provide different and often complementary perspectives. Researchers grounded in epidemiology, genetics, molecular biology, physiology, and clinical medicine have all described aspects of health and disease that differ by race or ethnicity and have posited theories to account for those differences. In attempting to synthesize these data into a coherent whole, we have decided against presenting our findings within a single theoretical framework. Instead, we chose an ecumenical approach, accepting core insights from multiple disciplines to understand the evidence on why racial and ethnic groups differ in health status and what can be done to address the disparities.
Over the past century, a series of reports have addressed the health of minorities in the United States: from the health-related papers in the Atlanta University Series (1896, 1906), to the comprehensive review by Dorn (1940) on the health of Blacks written as a background paper for Gunnar Myrdal’s An American Dilemma, to the groundbreaking 1985 Report of the Secretary’s Task Force on Black and Minority Health (U.S. Department of Health and Human Services, 1985). Until the 1960s and 1970s, minority health studies in the United States focused on Blacks. The exception was an extensive body of anthropological literature on American Indian health, dating from the 1920s (see the review in Young, 1994, pp. 25–28). Much of it suggests that the introduction of infectious diseases from European colonists led to a significant decline in the American Indian population; that decline did not begin to reverse itself until the end of the nineteenth century. Other studies of the health status of non-Black minorities in the nineteenth and early twentieth centuries were likely carried out, but probably limited to a single locale.
Systematic registration of deaths in the nation began only about 50 years ago when all of the United States was included in the Death Registration Area1 (Ewbank, 1987), and all of the historic studies show that Black death rates were substantially higher than those for Whites. Disparities lessened somewhat, however, when Blacks and Whites at similar socioeconomic position (SEP) were compared (e.g., Holland and Perrott, 1938). In the nineteenth century, Black death rates decreased along with those for Whites. Although Black death rates remained at a higher level, there were fewer deaths overall from infectious diseases because of improvements in factors such as water and sanitation and in child care. Ewbank (1987) concludes that in the second half of the nineteenth century, declines in Black mortality were minimal, but life expectancy increased from about 35 to 54 years between 1900 and 1940; thus, life expectancy for Blacks in 1940 was approximately equal to what life expectancy was for Whites in 1920.
Race as a Concept in Health Studies
Race as a factor in health care and health outcomes is increasingly being recognized and addressed by researchers. The modern concept of
race was developed during a period of intense global exploration and European colonialism. Hierarchical schemes of classification rooted in dubious theories of biology and genetics were invented to classify individuals as both foreign and inferior. This strategy served as a scientific and moral enabler of colonial economic activity, for which enslaving people was accepted as an economic expedient. The concept of race, however, has been under increasing assault in the last half of the twentieth century from at least two different fronts. First, the Civil Rights Movement and its aftermath led to the end of legal segregation. Subsequently, since the 1960s, there has been increasing racial integration in the U.S. mainstream. Second, intense and sustained immigration has led to a more diverse U.S. population, resulting in a society so multiracial, simplistic racial classification and hierarchies are no longer sustainable.
The concept of “race” as it has been typically used in health research, however, has at least five flaws:
“Race” does not reflect meaningful biological or genetic distinctions. Race classifications are based on an individual’s appearance; and although physiognomic differences are genetic, the vast proportion of genetic variation is not so visible, and occurs within races, not between them (Lewontin, 1972).
Race classification schemes are highly imprecise (LaVeist, 1994; Williams et al., 1993; Williams, 1996). In the case of determining the race of a child, such a classification may rely on arbitrary rules based on factors such as race or nationality of the mother, race or nationality of the father, or the race the mother or father if either was non-White, etc. In determining race on a death certificate or health record, judgment is often made visually by a clerk or a funeral director.
Our current concept of race is distorted by America’s Black-White experience. This is especially problematic with regard to trying to classify “Hispanics,” who may be either Black or White, though the vast majority of Hispanics self-describe as White (U.S. Bureau of the Census, 1999).
The idea of a pure race is problematic. Each human being more than likely embodies a variety of racial groups. U.S. citizens who self-identify as Black, for example, are likely genetic admixtures of White, American Indian, and African in widely different proportions.
The widespread analysis of data by race may steer investigators away from thinking more deeply about actual underlying variables of interest such as SEP, disenfranchisement, and cultural values. In an examination of the use of “race” in the journal Health Services Research over a 25-year period, Williams (1994) noted that about 66 percent of empirical studies included “race,” but only 13 percent defined the term or had a justifiable reason for using it.
Quality of Health and Health-Care Data
One of the primary challenges in research addressing racial and ethnic health disparities is the quality of data on the health status and health care of many minority populations. Data accuracy is especially problematic for Hispanics, Asian and Pacific Islanders, and American Indians and Alaska Natives. In the 1980s, fewer than half of the states recorded Hispanic ethnicity on death certificates. By 1996, the District of Columbia and all states except Oklahoma reported Hispanic ethnicity on death certificates (National Center for Health Statistics, 1998). Sorlie and colleagues (1992a) compared reported race and ethnicity from study data with death certificate data for 21,000 deaths in the National Longitudinal Mortality Survey and found that while agreement on race was 99.2 percent for Whites and 98.2 percent for Blacks, agreement for American Indians and Alaska Natives was 73.6 percent and for Asians and Pacific Islanders 82.4 percent. Agreement for ethnicity was 89.7 percent for Hispanics. Hahn and Eberhardt (1995) analyzed 1990 death certificate data, correcting for both census undercount and race misclassification. Changes in the estimates of American Indian and Alaska Native mortality were particularly striking; estimated life expectancy at birth decreased from 73.8 years to 71.0 years for men and from 81.6 years to 79.4 years for women. The change resulted in this population going from having a reported health advantage compared to Whites to having a disadvantage. Mortality rates that are not adjusted for misclassification underestimate the mortality of these populations and lead to overestimates of life expectancy.
Protocols for collecting racial and ethnic data also change over time. Kovar and Poe (1985) considered changes in data collection for the National Health Interview Survey (NHIS). From its inception in 1957 until 1976, NHIS racial categorization depended on interviewer observation, and all racial and ethnic categories other than Black and White were identified as “Other Races.” In 1976, in addition to interviewer-observed race, the NHIS added a single question allowing the respondent to identify one of the following categories as his or her national origin or ancestry: countries of Central or South America, Chicano, Cuban, Mexican, Mexicano, Mexican-American, Puerto Rican, Other Spanish, Other European (such as German, Irish, English, or French), Black/Negro/Afro-American, American Indian/Alaskan Native, and Asian and Pacific Islander (such as Chinese, Japanese, Korean, Filipino, Samoan). In 1978, to comply with guidelines established by the U.S. Office of Management and Budget (OMB) Directive 15, that question was then changed to two parts: the first addressed Hispanic ethnicity; the second addressed race. Persons who self-identified as being of Hispanic ancestry could report any of the following groups: Puerto Rican, Cuban, Mexican/Mexicano,
Mexican American, Chicano, “Other Latin American,” or “Other Spanish.” Racial categories to choose from were American Indian or Alaska Native, Asian and Pacific Islander, Black, White, or other. In 1992, NHIS expanded the subgroups of Asians and Pacific Islanders; in addition to Chinese, Filipino, Korean, and Japanese, respondents could choose Vietnamese, Asian Indian, Samoan, Guamanian, and “Other Asian and Pacific Islander.” (Beginning with the 2000 Census, all federal data systems will over several years change to comply with the revised OMB Revision directive that mandates federal agencies that collect data on race to allow persons to self-identify as being a member of more than one racial group [Office of Management and Budget, 1997].)
Although data from longitudinal cohort studies or interview surveys that rely on self-report of race and ethnicity should not be subject to significant misclassification errors, these longitudinal data sets do not allow assessment of trends in differences between races and ethnic groups.
Measures of Health Status
“Health” extends far beyond simple measures of the risk of disease and death. It encompasses an array of quality-of-life dimensions that include physical and social functioning and mental well-being. The primary measures of health we used in this overview are life expectancy, all-cause mortality, and, when available, respondent-rated general health status. Life expectancy is the estimated average number of years of life remaining for a person at a given age, and life expectancy at birth is a good summary measure of mortality differences across the life cycle. We use age-adjusted mortality rates,2 wherever possible, to allow appropriate comparisons because the age structure varies substantially across racial and ethnic groups. We use only all-cause mortality because subsequent chapters address cause-specific rates. Respondent-rated3 general health status is measured by the response to a standard survey question that asks a person to grade his or her overall health, usually on a scale of excellent, very good, good, fair, and poor. How a respondent answers
that single question captures many dimensions of health including health-related quality-of-life and functional status; how a respondent answers has also proved to be a significant predictor of mortality even after accounting for standard socio-demographic factors (Idler and Benyamini, 1997).
PATTERNS OF RACIAL AND ETHNIC DIFFERENCES IN MORTALITY AND GENERAL HEALTH STATUS
The story of health in the United States in the twentieth century can be summarized by two central findings. First, simply and clearly, across age groups, across racial and ethnic groups, among males and females, by almost all indicators, the health of the U.S. population has improved dramatically. Second—and complex to the degree that it defies straightforward description—in spite of the improvements, there have been remarkably persistent differences in health status by race and ethnicity, with the members of many, but not all, racial and ethnic minorities experiencing worse health than the majority population but also with substantial heterogeneity within any one group. Both findings are illustrated by statistics listed in Table 11–1.
Members of the largest minority group of the twentieth century, Blacks, have experienced poorer health compared to Whites throughout the century. Members of the fastest growing minority group in absolute numbers, Hispanics, have enjoyed relatively good health by some measures but contain within their ranks substantial variation—e.g., health of
TABLE 11–1 Age-Adjusted Death Rates, 1980–1996 (per 100,000 person years)
American Indian/ Alaska Native
Asian or Pacific Islander
SOURCE: National Center for Health Statistics (1998).
Puerto Ricans is comparatively poorer, whereas health of Mexican Americans and Cubans is comparatively better. For Asians and Pacific Islanders the picture is varied as well, though many Asian and Pacific Islander subgroups have comparatively good health. American Indians and Alaska Natives, like Blacks, continue to experience poorer health than Whites, though in recent periods the health gap between Whites and American Indians and Alaska Natives has narrowed considerably.
Over the course of the twentieth century, life expectancy at birth for Blacks has more than doubled; for Whites it increased by about 60 percent. In 1900, life expectancy for a Black person was 33 years and 47.6 years for a White person; by 1950, life expectancy had risen to 60.7 years for Blacks and 69.1 years for Whites (National Center for Health Statistics, 1998). The Black-White gap had decreased to its lowest point, 5.8 years, around 1984, and then began to increase due to high death rates among Blacks from HIV, homicides, diabetes, and pneumonia. (A decrease in life expectancy was extremely unusual for a large subpopulation in a developed country.) Another factor in the increased differential was that mortality from heart disease decreased at a much faster rate for Whites than for Blacks (Kochanek et al., 1994). By 1989 the gap was at 7.0 years, where it remained until 1995. By 1996 life expectancy increased to 70.2 years for Blacks and 76.8 years for Whites (National Center for Health Statistics, 1998). Comparison of age-adjusted all-cause mortality rates for Blacks presents a similar picture. In 1950, the mortality rate for Blacks was 1,236.7 per 100,000; the rate for Whites was 800.4. As shown in Table 11–1, by 1996, the rate for Blacks had fallen by almost 40 percent to 738.3; the rate for Whites was 466.8 (National Center for Health Statistics, 1998).
One fact is true for all groups: women live longer than men. Relative gains in mortality and life expectancy have not been experienced equally, however, by Black men and women. In 1900, the life expectancy at birth of Black females was 1 year greater than for Black males; for White females life expectancy was 2.1 years longer than for White men. Among Blacks the differential increased to 3.8 more years for females in 1950,8.3 more in 1970, and peaked at about 9.2 years in 1991. In 1996 it decreased to 8.1 years. Among Whites the differential peaked in 1970 at 7.6 years and has steadily declined since then, primarily because of reductions in cardiovascular disease among men (National Center for Health Statistics, 1998; Waldron, 1993).
National survey data on respondent-reported general health status also suggest a poorer health-related quality of life for Blacks compared to Whites. The age-adjusted percentage of Blacks who reported “fair” or
“poor” health in 1995 was 15.4 percent compared to 8.7 percent for Whites (National Center for Health Statistics, 1998), and it has remained around 15 to 17 percent during the last decade. Among the elderly age 75 or older, the Black-White gap in the percentage reporting “fair” or “poor” health narrowed from 18.8 percent to 13.1 percent between 1987 and 1995. Among Black persons age 75 and older the percentage dropped from 52.4 to 44.4; among Whites, from 33.2 percent to 31.2 percent.
Although the Black population is often thought of as a culturally distinctive, homogenous group, there are in fact significant differences among subgroups. Several studies have described wide variation in mortality rates among Blacks in different parts of the county, with some areas far behind others in recent improvements in health (McCord and Freeman, 1990; Geronimus et al., 1996). A growing literature has also described differences in health between Blacks who were born in the South and Blacks born in other parts of the country (Fang et al., 1996; Schneider et al., 1997; Kington et al., 1998). Most, but not all, of this literature has found worse health for Blacks born in the South even if they have migrated. These findings may be related to poor early-life health conditions in the South or perhaps to more subtle psychosocial factors (Schneider et al., 1997).
A related body of literature is now describing the health of recent immigrants of African descent, most notably those from the Caribbean but more recently those from sub-Saharan Africa. In 1980, approximately 3.1 percent of Black Americans were foreign born. It is estimated that by 2010, over 10 percent of the Black population in the United States will be recent immigrants or their descendants (Reid, 1986). We know of no national studies of West Indian or African immigrants, but Fang and colleagues (1997) analyzed data on mortality in New York and found that U.S.-born Blacks had between 1.5 and 2 times the age-adjusted mortality rate of foreign-born Blacks (67 percent of whom were from the West Indies, 12 percent from South America, and 5 percent from Africa). David and Collins (1997) analyzed data on the birth weight of infants, a potent indicator of infant mortality risk, and found that among infants born in Chicago between 1985 and 1990, those born to African-born Black women were likely to have a birth weight more similar to infants of U.S.-born White women rather than to babies born to U.S.-born Black women. Literature on the relatively good health of foreign-born Blacks in the United States has been used to dispel the idea that Black health status is genetically based. Comparisons across Black subgroups may provide important additional insights into the causes of racial and ethnic differences in health.
By some measures of general health status, Hispanics experience relatively good health, but closer scrutiny of the data reveals a more complicated pattern. Life expectancy at birth for Hispanics in 1990 was estimated to be 79.1 years, 3 years longer than the 76.1 years estimated for Whites (Erikson et al., 1995). That 79.1 years, however, is almost certainly an overestimate because it is based on data that exclude several states, including New York, where over 40 percent of Puerto Ricans live, and this group generally has poorer health status than other Hispanic subgroups. NCHS mortality data also suggest that the health of the Hispanic population has improved. Between 1985 and 1994–1996, the age-adjusted mortality rate for Hispanic males decreased from 524.8 to 501.1, and for Hispanic females from 286.6 to 270.34 (National Center for Health Statistics, 1998). For non-Hispanic White males over this period, the mortality rate dropped from 669.7 to 602.8 and for females from 385.3 to 365.5. Analyses by Sorlie et al. (1993) and Liao et al. (1998) of longitudinal data on national sample cohorts revealed a similar pattern of lower mortality rates for Hispanics compared to non-Hispanic Whites.
In 1990, Mexican Americans comprised the largest subpopulation of Hispanics, 63 percent, Puerto Ricans comprised 11 percent, Cubans, 5 percent, and a combined category of “all other Hispanics” including immigrants from Central and South America comprised the remaining 21 percent (US. Bureau of the Census, 1991). Analyses of data sets that have included a sufficient number of members of each subgroup to compare have found substantial differences across the Hispanic subgroups. For example, Sorlie and colleagues (1993) analyzed data from the National Longitudinal Mortality Study and compared the age-adjusted mortality rate of Hispanic subpopulations and the non-Hispanic population; they found that the ratio for Hispanics compared to non-Hispanics was 0.74:1 for men and 0.82:1 for women (ratios less than 1 indicate better health than Whites), but the ratio varied substantially across age groups. For example, among men older than age 65, Mexicans, Puerto Ricans, Cubans, and a final category of “all other Hispanics” each had substantially lower mortality rates than non-Hispanics. In contrast, among those aged 25 to 44, no Hispanic subgroup had a mortality risk significantly different than Whites. Liao and colleagues (1998) analyzed linked mortality data from the 1986–1990 NHIS and describe a similar pattern of overall lower
mortality for Hispanics, but with a pattern of significantly higher mortality among young Hispanic men and lower mortality among older Hispanics.
Mortality data present a picture of relatively good health for Hispanics on average; however, survey data on respondent-reported general health present a different picture. Data from the 1992–1994 NHIS revealed that Hispanics were more likely to report “fair” or “poor” health (14.7 percent) than non-Hispanic Whites (10.4 percent). Among Hispanic subgroups, Puerto Ricans were more likely to report “fair” or “poor” health (17.4 percent) than Mexicans (15.3 percent) or Cubans (13.3 percent). A similar pattern of relatively poor respondent-rated health status for Hispanics was found in the 1987–1988 National Survey of Families and Households (Ren and Amick, 1996).
American Indians and Alaska Natives
Much of the data on the health of American Indians and Alaska Natives comes from the Indian Health Service (IHS), which provides healthcare services primarily to the American Indian and Alaska Native populations living on or near reservations (Indian Health Service, 1996; Kunitz, 1996) —about 60 percent of all Native Americans residing in the United States. The consensus is that American Indians and Alaska Natives have significantly poorer health than Whites, and data from the IHS, which generally provides a reasonable picture of its service population, corroborate that consensus. Their data also suggest that the gap between American Indians and Alaska Natives and Whites is narrowing.5 In 1940, life expectancy at birth for American Indians and Alaska Natives was 51 years compared to 64.2 years for Whites, a gap of 13 years (Indian Health Service, 1991). By the period of 1972 to 1974, life expectancy for the IHS population had increased to 63.5, leaving about an 8-year gap. By 1991– 1993, the gap had narrowed to about 3.3 years, with a life expectancy of 73.2 years for the IHS population (Indian Health Service, 1996).
Although more than half of American Indians live in urban areas, IHS provides services in relatively few urban areas (Indian Health Service, 1996). Because the IHS data are the primary source of high-quality data on the health of American Indians, little is known about the health of urban American Indians and Alaska Natives. In one of the few studies of urban American Indians, Grossman and colleagues (1994) compared health indicators among urban American Indians, Blacks, and Whites in
the Seattle, Washington, metropolitan area and American Indians in seven rural Washington counties with reservation lands. In general, urban American Indians in this study had a pattern of health similar to that of Blacks—i.e., poorer than Whites; but no consistent differences were found between urban and rural American Indians.
Asian Americans and Pacific Islanders
Immigrants from Asia and the Pacific Islands have been a part of this country for more than a century, but national data on the relative health status of this population have been collected only within the past thirty years (Yu and Liu, 1992). Although national vital statistics are likely to underestimate mortality for Asians and Pacific Islanders, it is still accepted that as a group Asian Americans live longer than Whites. In 1992, the estimated life expectancy for Asian Americans between birth and age 1 was estimated to be 80.3 years compared to 75.1 years for Whites (Hoyert and Kung, 1997). Such a statistical comparison, however, masks the diversity of among Asian and Pacific Islander subgroups. The only reasonable summary statement about the relative health of Asian and Pacific Islander populations is that some of the subgroups enjoy good health compared to the country as a whole, while other subgroups have poorer health.
In 1990 the three largest subgroups among Asian Americans were Chinese Americans (24 percent), Filipinos (20 percent), and Japanese (12 percent) (U.S. Bureau of the Census, 1993). The remaining 44 percent included Vietnamese, Koreans, Hawaiians, Samoans, and Guamanians, and smaller subgroups such as Cambodians and Hmong. Each of these subgroups has a distinctive culture, language, and immigration history. Not surprisingly, health status varies substantially across these groups, as demonstrated by an analysis of 1992 mortality data from seven states that account for 73 percent of the Asian and Pacific Islander population (Hoyert and Kung, 1997). At one extreme, Japanese, Asian Indians, and Koreans had mortality rates of 298.8, 272.2, and 292.3, respectively, substantially lower than the population-wide average of 523.6. At the other extreme, Hawaiians and Samoans had rates of 901.4 and 907.7, respectively. A similar pattern of significantly lower mortality rates for Chinese, Asian Indians, and Japanese was found in California from 1985 to 1990 (Wild et al., 1995).
Wide variations in respondent-rated general health status were in data from the 1992–1994 NHIS (Kuo and Porter, 1998). Compared to 8.9 percent of non-Hispanic Whites reporting “fair” or “poor” general health, 17.2 percent of Vietnamese reported “fair” or “poor” health. The rate for Asian Indians was similar to that for non-Hispanic Whites; for Chinese it was 7.4 percent, for Filipinos, 7.4 percent, and for Japanese, 6.1 percent.
Specific Measures of Mortality and General Health Status
Infant mortality has long been considered a key indicator of a population’s well being. Low birth weight (LBW) is the most important risk factor for infant mortality. Neonatal infant mortality is generally believed to result from factors intrinsic to the overall health of the pregnancy, whereas postneonatal mortality is considered to be more related to environmental factors in the infant’s first year of life. Low birth weight is the most important risk factor for infant mortality and may result because an infant is born too small (small for gestational age [SGA]) and/or too soon (preterm). Data patterns of LBW, SGA, and preterm births are complex and vary among various racial and ethnic groups and subgroups, a full discussion of which is beyond the scope of this overview.
Less complex are data patterns indicating that infant mortality rates are dropping in the United States. As with other health data, it is not possible to establish historical trends for most non-Black minorities because such information has only been available recently; but recent data given in Table 11–2 show current patterns illustrating substantial variation in rates between both groups and subgroups. What is startling is that in 1996 Blacks still had the highest rates at 14.7 per 1,000 live births compared to 6.1 for Whites (National Center for Health Statistics, 1998; MacDorman and Atkinson, 1998). More startling is the fact that the ratio of Black-to-White infant mortality has increased—from 1.6:1 in 1950 to about 2.4:1 in the 1990s (National Center for Health Statistics, 1998). Historical data for American Indians and Alaska Natives documents a substantial improvement in the infant mortality rates, and, in particular, the neonatal infant mortality rates (Indian Health Service, 1996). The neonatal infant mortality rate for American Indians and Alaska Natives remains somewhat higher than that of Whites, but the postneonatal infant mortality rates for American Indians and Alaska Natives are about 2.5 times that of Whites (National Center for Health Statistics, 1998).
Mortality Among the Elderly
The fastest growing age group of the U.S. population is people 85 years old and older, and the growth rate for racial and ethnic minorities is substantially higher than for Whites (Angel and Hogan, 1994; U.S. Bureau of the Census, 1996). The elderly are most likely to experience health problems and, because of their burgeoning number, are likely to consume more and more health-care resources. An increasing percentage of research over the last two decades has addressed the health of the elderly,
TABLE 11–2 Infant, Neonatal, and Postneonatal Mortality Rates by Race or Hispanic Origin of Mother: United States, 1995
minority elderly in particular; and it has found that patterns of differences in health across racial and ethnic groups persist into old age.
In I960, life expectancy at age 65 was 14.4 years for Whites and 13.9 years for Blacks—a gap of 0.5 years (National Center for Health Statistics, 1998). By 1990, life expectancy for both groups at this age had increased: at age 65 life expectancy was 17.3 years for Whites and 15.4 years for Blacks, an increase in the gap to 1.9 years.
There has been some debate as to whether the health disadvantages of minorities increase in old age because of a compounding of the effects of age and race, the “double jeopardy” hypothesis (Dowd and Bengston, 1978). Evidence of this hypothesis is inconsistent. Ferraro and Farmer’s (1996) longitudinal analysis of national data found that at every age Blacks had poorer health than Whites but did not find Blacks becoming more worse off, relative to Whites, as they age. Gibson (1991) reviewed analyses of several national data sets and described a bimodal distribution of
health among older Blacks—a younger group was, on average, relatively sick and an older group was, on average, relatively hardier. For many years, analyses of mortality data have suggested that at older ages, the Black-White mortality gap narrows and eventually crosses over so that the life expectancy of Blacks older than age 75 to 80 exceeds that of Whites. Preston et al. (1996) ascribed mortality cross-over results primarily to differential misreporting of age; Corti and colleagues (1999) suggest that the cross-over may be real.
Mortality rates for elderly Hispanics consistently have been found to be lower than rates for elderly Whites (Sorlie et al., 1993; Liao et al., 1998; National Center for Health Statistics, 1998). Among those age 65 to 74, mortality rates in 1994–1996 for Hispanic males was 2,307.8, for Hispanic females, 1,387.0, for White males, 3,225.0, and for White females, 1,940.8 (National Center for Health Statistics, 1998). Two recent national surveys, however, one of persons age 50 to 60 and another of persons age 70 and above, indicate poorer respondent-rated general health for older Hispanics compared to Whites (Smith and Kington, 1997b).
Mortality data on American Indians and Alaska Natives older than age 65 suggest relatively small differences in health compared to Whites. In 1991–1993, the mortality rate for American Indians and Alaska Natives age 65 and older was 4,255.2; for Whites in this age group, it was 4,871.2 (Indian Health Service, 1996). The general pattern of overall better health for Asians and Pacific Islanders and wide differences among these subgroups persists in old age. For example, life expectancy for persons between ages 65 and 74 in 1992 was 17.2 years for Whites and 19.9 years for all Asians and Pacific Islanders. Among Asian and Pacific Islander subgroups, Chinese, Japanese, Asian Indians, and Filipinos in the older age groups have longer life expectancies than Whites; Hawaiians and Samoans have lower life expectancies (Hoyert and Kung, 1997).
Disability and Functional Status
Disability addresses the impact of disease and poor health on a person’s ability to handle demands made on that person in particular environments. One of the most widely accepted frameworks for understanding disability was proposed by Nagi (see review, Nagi, 1976). It describes disability as the end result of a process that leads from pathology in a bodily function, to impairment of a physical structure or function, to a limitation in the ability to perform tasks, and finally to a limitation in performing an expected role. Although disability has a much higher prevalence among the elderly, its impact on individuals and families may be even greater when it occurs in working-age adults (DeJong et
al., 1989). Most of the data related to disability and the closely related concept of physical function are from elderly populations.
NHIS is a primary source of data on disability rates. One general measure of disability is activity limitation—i.e., a long-term reduction in the ability to perform the usual age-appropriate activities because of chronic disease or impairment. In 1995 Blacks were more likely than Whites to report an activity limitation (age-adjusted percentage 17.4 percent versus 13.6 percent) (National Center for Health Statistics, 1998). For this paper we analyzed NHIS data for 1992–1994 and found Hispanics overall reported a major activity limitation at a rate similar to non-Hispanic Whites, at about 15 percent age-adjusted, with Puerto Ricans at 15.9 percent, and Cubans and Mexicans at about 10 percent. In general, Asian and Pacific Islanders are less likely to report a disability compared to Whites. Kuo and Porter (1998) report that among Asians and Pacific Islanders, in the NHIS 1992–1994 surveys, an age-adjusted 9.3 percent reported an activity limitation, compared to about 15 percent for non-Hispanic Whites. Although there was variation among Asian and Pacific Islander subgroups, all had percentages less than Whites. The highest percentage was for Vietnamese at 13.2 percent. John (1996) reports high rates of disability among American Indians and Alaska Natives. Across age groups, American Indians and Alaska Natives report higher rates of activity limitations caused by chronic medical conditions (Centers for Disease Control and Prevention, 1992). Among persons age 60 and older, 44 percent report a work disability compared to 29 percent of non-Hispanic Whites (U.S. Bureau of the Census, 1994).
A growing amount of evidence suggests an overall lowering of disability rates among older Americans (e.g., Freeman and Martin, 1998). Relatively few studies have specifically addressed trends in relative disability rates across racial and ethnic groups. Clark (1997) compared data from the 1982, 1984, and 1989 National Long Term Care Surveys and found that Black-White disparities in disability increased substantially over this period. Freeman and Martin (1998), using data from the 1984 and 1993 Survey of Income and Program Participation, present a different pattern. Their findings suggest that disability rates for persons more than 49 years old were higher for both Blacks and Hispanics compared to Whites, but that disability rates in all groups have decreased since the 1980s.
CAUSAL PATHWAYS FOR DIFFERENCES IN HEALTH STATUS
Over the past 20 years, research on differences in health status among racial and ethnic groups has shifted from descriptive studies of differences in outcomes toward analytical studies of potential underlying
causes for the differences. Delineating the causal pathways requires pursuing an often frustratingly ill-defined course and piecing together an array of scientific clues. The evidence is overwhelmingly observational and often cross-sectional, with all of the attendant limitations of those types of data; and measures of many important variables are imprecise. Long periods of time generally separate the initiation of a causal factor and its impact on health; to date, no comprehensive studies trace the consequences of any one factor through the complete pathway to the clinical and biological manifestations of impaired health. Therefore, the process of reaching reasonable conclusions about why Blacks have poorer health than Whites, or why Mexican Americans have overall better health than Puerto Ricans, or why Japanese but not Vietnamese have relatively good health requires intellectual leaps, and this literature has been criticized for that reason (e.g., Kaufman and Cooper, 1995).
That said, in the following sections, we review evidence relating observed racial and ethnic differences in health to SEP; health-risk behaviors; psychosocial factors, including stress and racism; access to health care; acculturation; genetic factors; and environmental and occupational exposures to the observed racial and ethnic differences in health. These groups of factors are addressed separately, but many are intimately related to each other. For example, dietary patterns are closely related to socioeconomic position, which in turn may be related to environmental exposures.
In his groundbreaking 1899 study The Philadelphia Negro, DuBois compared mortality rates of Blacks and Whites at different socioeconomic levels. One hundred years later, with increasingly more complex statistical techniques and more comprehensive data sets, the basic premise of DuBois’s analysis remains a cornerstone of research efforts to understand racial and ethnic differences in health in the United States.
We use the term socioeconomic position (SEP), as defined by Krieger, Williams, and Moss (1997) and Haan and Kaplan (1985), to capture the range of dimensions that describe an individual’s relative and absolute position in society—financial and human resources such as income, wealth, and educational skills; social position or class; occupational prestige; and contextual variables such as neighborhood poverty concentration. SEP also captures the dynamic nature of these dimensions over the life stages from fetal life to old age. Although the range of concepts covered by the term is wide, SEP has been measured, in almost all studies until very recently, by a few relatively simple variables—yearly gross income, education, and occupation. One common analytic approach for
studying racial and ethnic differences in health has been first to describe unadjusted differences in a health outcome and then statistically to adjust for SEP and assess whether the differences remain. Unfortunately, when differences in health between groups persist after controlling for standard SEP measures—such as educational attainment—residual differences are often inappropriately attributed to genetic or biologic differences (Kaufman et al., 1997).
Attention has focussed on SEP as a factor in racial and ethnic differences in health for two reasons. First, over the past 150 years, a strong relationship between SEP and health has been found in an overwhelming amount of research (Antonovsky, 1967; Marmot et al., 1987; Bunker et al., 1989; Williams, 1990; Feinstein, 1993; Adler et al., 1994; Evans et al., 1994; Preston and Taubman, 1994; Anderson and Armstead, 1995; Blane et al., 1996; Wilkinson, 1996). The evidence is succinctly summarized in the opening sentence to a recent editorial in The Lancet (1995): “The rich are healthy; the poor are not.” Second, many racial and ethnic groups with poorer health are also more likely to have low SEP.
Although an association between socioeconomic position and health is well-documented, questions remain about the relative contribution of the two causal directions of the relationship. Lower SEP may lead to poor health, and conversely, poor health may also lead to low SEP. We focus here on the causal direction from SEP to health. (For a discussion of theoretical and analytic implications of this issue, see Garber, 1989, Smith and Kington, 1997b, and Ettner, 1996.)
Measurement and Conceptualization of SEP
Most health databases include only simple measures of SEP such as income and education. The lack of comprehensive measures of SEP in health databases limits researchers’ ability to make definitive determinations about the extent to which differences in health status are related to SEP (Kreiger et al., 1997; Kaufman et al., 1997). For example, relatively few health studies include data on wealth, even though wealth may be a better measure of lifetime economic resources than income. Presently, racial and ethnic differences in wealth in the United States are significant, exceed the differences in income, and appear to be growing (e.g., Wolff, 1998, Smith, Volume II, Chapter 4). A person’s SEP may also change over time. Most research on the relationships between SEP, race, ethnicity, and health relies on assessments of SEP at a single point in time, overlooking the cumulative and dynamic nature of SEP (Singer and Ryff, 1997); and recent analyses of data sets that include longitudinal data on SEP suggest that duration of SEP matters (Lynch et al., 1997; McDonough et al., 1997; Hart et al., 1998). Health status today may reflect circumstances or expo-
sures decades old. Conditions in early childhood or even in utero may set health trajectories for chronic conditions much later in life (Barker, 1994; Peck, 1994). Given the dramatic changes in SEP among racial and ethnic groups in this century, particularly for Blacks, the lack of long-term measures of SEP is especially problematic and could prove detrimental to addressing racial and ethnic disparities.
The characteristics of the area where a person lives may also be SEP determinants of an individual’s health (Haan and Camacho, 1987; Anderson et al., 1997). Community-level characteristics encompass factors such as environmental risks and housing quality, and thus are relevant within the context of understanding the role of SEP. Community-level characteristics related to SEP may be especially important for understanding differences in health status because of the persistent degree of residential segregation among some racial and ethnic groups. A growing body of literature has described a strong correlation between various measures of societal-level inequality and mortality. Wilkinson (1996) and other researchers have proposed that inequality at the societal level rather than at the individual level may be the primary pathway for the relationship between SEP and health. Understanding the role of societal-level inequality may be especially relevant in light of growing income inequality between and within some racial and ethnic groups (Pappas et al., 1993).
Summary of Findings on SEP
The bulk of the research addressing the role of socioeconomic position in explaining racial and ethnic differences in health has focused on differences between Blacks and Whites (Williams and Collins, 1995; Smith and Kington, 1997b; Dressler, 1993; Kaufman et al., 1997, 1998; Lillie-Blanton et al., 1996). Most of the reviews—including studies of all-cause mortality as well as infant mortality, general health and function, and specific disease prevalence and incidence—come to the conclusion that SEP (typically education and income) accounts for a substantial part of the observed Black-White differences in health status. Many studies of all-cause mortality (Menchik, 1993; Keil et al., 1992; Rogers, 1992; Sorlie et al., 1992b; Otten et al., 1990) and studies of functional status and general health in adults (Schoenbaum and Waidmann, 1997; Smith and Kington, 1997a; Kington and Smith, 1997; Guralnik et al., 1993) have found that SEP is an important underlying factor. SEP as typically measured, however, appears to play a more limited role in explaining Black-White differences in infant mortality and prevalence of specific diseases such as hypertension and diabetes (Lillie-Blanton et al., 1996; Smith and Kington, 1997b; Williams and Collins, 1995; Schoenbaum and Waidmann, 1997).
Less attention has focused on Hispanics, probably because the health
disadvantage is not as consistent. In one of the few studies addressing the role of SEP in explaining health status differences between Hispanics and Whites, Wei and colleagues (1996) analyzed longitudinal data from the San Antonio Heart Study and found that although foreign-born Mexican Americans had mortality rates similar to non-Hispanic Whites, U.S.-born Mexican Americans had a significantly higher mortality rate. Controlling for SEP measured by income, education, and an occupational prestige index accounted for all of the mortality disadvantage for U.S.-born Mexican Americans. In their analysis of mortality data from the National Longitudinal Mortality Study, Sorlie and colleagues (1993) found that the overall Hispanic mortality advantage increased after adjusting for family income. Given the lower SEP of Hispanics in general, this finding suggests that SEP operates within the Hispanic population as it does in general but that other factors are likely to account for the patterns of mortality advantage. To explain this finding of better health among some Hispanic subgroups in spite of worse SEP—what has been called the “Hispanic paradox”or “epidemiological paradox” —several theories have been proposed including selective migration to this country of relatively healthier immigrants, strong social networks of Hispanic immigrant communities, and better diets (Markides and Coreil, 1986). Hayes-Bautista (1992) and Liao and colleagues (1998) have questioned the pervasive acceptance of the intellectual foundations of the “paradox” nomenclature, suggesting that it is based on a simplistic model of causation and a narrowly defined and measured notion of SEP.
SEP also plays a role in rates of low birth weight (LEW) in babies, but the effect of SEP is variable. Among Blacks, Whites, and American Indians and Alaska Natives, LEW rates are inversely related to the mother’s level of education: i.e., the higher the level of education, the lower the rate of LEW. American Indians and Alaska Natives have less of a gradient than Blacks or Whites. Hispanics and Asians and Pacific Islanders in general do not show this pattern; they tend to have similar and low rates of LEW across the educational spectrum. Within these minority groups, however, Hawaiians do show a strong inverse relationship, as do Puerto Ricans and Cubans. To further illustrate the complex interactions of race and SEP, LEW rates are higher among the most educated segment of Blacks than among the least educated segment of any other race or ethnic group (National Center for Health Statistics, 1998).
Four health behaviors—smoking, dietary intake, physical activity, and alcohol intake—are particularly important determinants of health. Another important determinant of health, though not a behavior, is obesity.
Obesity, however, is related to both diet and physical activity behaviors; it is also related to the risk for conditions such as diabetes and hypertension, which are important contributors to racial and ethnic differences in health status. A substantial body of evidence suggests significant differences across racial and ethnic groups for a range of health behaviors; increasing attention has been focused by government and private health care entities on public health interventions to discourage risk-increasing health behaviors and promote risk-lowering behaviors (U. S. Department of Health and Human Services, 1990).
Cigarette smoking is a causal factor for some cancers, respiratory disease, and cardiovascular disease; and different racial and ethnic groups have distinctly different risks for these diseases. Scientific evidence on racial and ethnic differences in smoking was extensively reviewed in a report from the Office of the Surgeon General (U.S. Department of Health and Human Services, 1998); that report is considered to be the definitive synthesis of evidence on this topic and is, therefore, summarized here.
Among racial and ethnic groups, rates of smoking are highest among American Indians and Alaska Natives, high among Blacks and Southeast Asian men, and lowest among Asian and Pacific Islander and Hispanic women. Among adults during 1994 and 1995, age-adjusted percentages for current smoking among adults were: American Indian and Alaska Native, 36; Blacks 26.5; Whites, 26.4; Hispanics, 18.0, and Asians and Pacific Islanders, 14.2. From 1978 through 1995, smoking declined among Blacks, Asian, and Pacific Islanders, and Hispanic adults. After a period of several years of decline among all groups, however, during the 1990s smoking increased among Black and Hispanic adolescents. Particularly relevant to Hispanic health and the health of other largely immigrant populations is the finding that as immigrant populations acculturate they adopt the smoking habits of the broader community. For example, evidence suggests that smoking increases among Hispanic women as they become more acculturated (Haynes et al., 1990).
The Surgeon General’s report concluded that differences in smoking patterns across racial and ethnic groups were directly related to differences in patterns of diseases related to smoking such as lung cancer, but there is little research on how much of the racial and ethnic differences in various smoking related diseases can be attributed to the behavior (Chen, 1993).
We know of no research that has specifically addressed the relationship between observed racial and ethnic differences in health and differences in diet. In the face of growing evidence relating diet to health risks, the National Academy of Sciences developed guidelines to promote healthy dietary intake; they recommend that people consume less fat, cholesterol, sodium, and red meat, and more carbohydrates and fiber from fruits, vegetables, grains, and legumes (National Research Council, 1989). In recent years, Americans in general have improved their dietary habits, with notable decreases in fat and cholesterol intake (Norris et al., 1997); however, the overall dietary quality remains low for all groups.
The analysis of racial and ethnic differences in diet is complicated by large differences in diet among persons of differing SEP (James et al., 1997). Popkin and colleagues (1996, 1997) compared survey data for Whites and Blacks between 1965 and 1989–1991 and calculated a diet quality index that scores overall compliance with eight dietary recommendations of the National Academy of Sciences; they found that in 1965 the dietary quality of Whites and Blacks was similar and that overall quality improved for both Blacks and Whites. There were only small differences across racial and socioeconomic groups, and Blacks at high socioeconomic levels had the highest quality diets.
Data from NHIS in 1987 and 1992 indicate Hispanics consumed the least fat as a percentage of energy in both time periods and reduced fat and cholesterol intake more than Whites or Blacks. Blacks reduced fat and cholesterol intake substantially less than Whites. There was little change for the population as a whole in fiber intake, but Hispanics in 1992 consumed more fiber than Whites or Blacks (Norris et al., 1997).
Few studies have comprehensively described the dietary habits of either American Indians and Alaska Natives or Asians and Pacific Islanders, and we found none that described trends. The studies that have addressed these groups typically have been restricted to one subgroup such as a single American Indian and Alaska Native tribe or an Asian and Pacific Islander group in one city. A recent review of studies of dietary habits of American Indians and Alaska Natives concluded that in general this population has a diet that is moderately high in total fat and saturated fat and low in polyunsaturated fat and fiber, but it was noted that in spite of these patterns, American Indians and Alaska Natives overall have low rates of two common diet-related diseases, heart disease and cancer (Bell et al., 1997).
Although most research on racial differences in dietary habits has focused on consumption of high-risk foods such as fat and cholesterol
and related risks for obesity and cardiovascular disease, insufficient food intake may also be an important health-risk factor, especially for women of child-bearing age and children. From 1988 through 1994, the overall prevalence of food insufficiency, as measured by reports of a family sometimes or often not having enough food to eat, was about 4.1 percent (Alaimo et al., 1998). Insufficiency was primarily related to poverty, but there were differences across racial and ethnic groups. Non-Hispanic Black families were more likely to report food insufficiency than non-Hispanic White families, primarily due to income and other socio-demographic differences. Even after controlling for these factors, however, Mexican families were twice as likely to report food insufficiency as non-Hispanic White families. It is not known to what extent differences in food insufficiency contributed to differences in health outcomes.
The relationship between physical activity and good health has been clearly demonstrated (U.S. Department of Health and Human Services, 1990, 1996). Lack of physical activity has been shown to be related to overall mortality as well as to the risk of chronic diseases such as cardiovascular disease, non-insulin dependent diabetes, osteoarthritis, and depression (U.S. Department of Health and Human Services, 1996). Rates of physical activity clearly differ across racial and ethnic groups. In 1990, only 34.3 percent of Blacks and 34.9 percent of Hispanics older than age 18 reported exercising or playing sports regularly compared to 41.5 percent of Whites and 41.2 percent of all non-Hispanics older than age 18 (Piani and Schoenborn, 1993). The one exception to this general pattern across age groups and genders was among men between ages 18 and 29; in this group Blacks were more likely to report regular exercise or playing sports than Whites.
Relatively little is known about the physical activity patterns of American Indians and Alaska Natives and Asians and Pacific Islanders. An analysis of national data from the 1991–1992 Behavioral Risk Factor Surveillance System compared across racial and ethnic groups the percentage of persons with a sedentary lifestyle, defined as reporting fewer than three 20-minute sessions of leisure-time physical activity per week (Morbidity and Mortality Weekly Report, 1994). The study found sedentary lifestyle rates of 64 to 65 percent for both Asians and Pacific Islanders and American Indians and Alaska Natives, compared to 56 percent for Whites, 62 percent for Hispanics, and 68 percent for Blacks. In all groups, with the exception of Asian and Pacific Islander men, the prevalence of sedentary lifestyle was lowest among those with the highest education. We found no comprehensive analyses comparing rates of physical activity across
subgroups of Asians and Pacific Islanders or American Indians and Alaska Natives.
Obesity, defined as a body-mass index of =30.0, although not a behavior, is an important risk factor for a range of important chronic diseases and is likely related to both dietary intake and physical activity behaviors. With obesity, genetic factors are also likely to play a causal role (Rosenbaum et al., 1997a). The prevalence of overweight and obesity has increased considerably across the U.S. population since at least 1960, and the general trends have been similar across racial and ethnic groups for whom trend data are available (Flegal et al., 1998). National data on body weight for Blacks and Hispanics have been available only since the late 1970s, and there are few data sets on body weight for Asians and Pacific Islanders and American Indians and Alaska Natives. Non-Hispanic Blacks and Mexicans are substantially more likely to be obese, with most of the difference among women. During 1988 through 1994, 22.4 percent of non-Hispanic White women were obese, compared to 37.4 percent of non-Hispanic Black women and 34.2 percent of Mexican women (Flegal et al., 1998). Furthermore, analysis of data on adolescents suggested that the rate of obesity increases in second- and third-generation immigrants (Popkin and Udry, 1998). Although data from other racial and ethnic groups are more limited, they suggest a high prevalence of obesity as well among Puerto Ricans, American Indians and Alaska Natives, Native Hawaiians, and Samoans, again with strikingly higher rates among women (Kumanyika, 1993; McGarvey, 1991; Aluli, 1991; Broussard et al., 1991).
Few studies have specifically addressed the contribution of racial and ethnic differences in obesity to the differences in prevalence of chronic medical conditions. Obesity is a particularly important risk factor for adult-onset diabetes, and Blacks, Hispanics, and American Indians all are at increased risk (Centers for Disease Control and Prevention, 1992). Two studies of the contribution of obesity to Black-White differences in diabetes found that the obesity contributed to, but did not explain all of, the differences in diabetes (O’Brien et al., 1989; Cowie et al., 1993).
Excessive alcohol intake is an important health-risk factor related to numerous health outcomes including chronic diseases such as cirrhosis, pancreatitis, and cardiomyopathy as well as injuries and sexually transmitted diseases. Alcohol is also a factor in a range of adverse pregnancy outcomes and fetal abnormalities. There are substantial differences in
alcohol intake across racial and ethnic groups. In 1990, Blacks were less likely than Whites (4.3 percent versus 5.8 percent) and Hispanics were less likely than non-Hispanics (4.6 percent versus 5.6 percent) to report excessive alcohol intake (here defined as an average of one ounce or more of alcohol per day in a two-week period) (Piani and Schoenborn, 1993); likewise, Blacks were less likely than Whites (46.4 percent versus 63.1 percent) to report at least one drink of beer, wine, or liquor over two weeks, and Hispanics were less likely than non-Hispanics (52.3 percent versus 61.4 percent) to report at least one drink of beer, wine, or liquor over a two-week period. Data from the 1982–1984 Hispanic Health and Nutrition Examination Survey suggest that among Hispanic subgroups, Mexican and Puerto Rican men have the highest rates of excessive alcohol intake (Aguirre-Molina and Caetano, 1994).
Excessive alcohol intake is a particularly important problem among American Indian and Alaska Native populations. From 1991 through 1993 their mortality rate from alcoholism was 38.4, 6.3 times the rate among Whites (Indian Health Service, 1996). Though the rate remains high, both the absolute alcoholism mortality rate and the rate relative to Whites has decreased in recent years. From 1979 through 1981, the alcoholism mortality rate among American Indians was 59.0, 8.6 times the rate in Whites. Although there is no question that alcoholism has a significant impact on the health of American Indian populations, surveys of drinking patterns, most of which are restricted to a single tribe or reservation, present a complex picture. A recent review of the epidemiologic evidence on drinking among American Indians concluded that (1) there is wide variation in drinking patterns across subgroups and periods of time, (2) some recent studies suggest less drinking on average among American Indians compared to the general U.S. population, (3) urban Indian populations tend to have a higher prevalence of drinking than reservation populations, (4) Indian women are more likely than other U.S. women to abstain, and (5) among those Indians who drink, there is a high prevalence of heavy drinkers, particularly binge drinkers (May, 1996).
Although the deleterious health effects of excessive alcohol intake are clear, the overall health implications of alcohol intake for a population are complicated by the well-described relationship between moderate alcohol intake and decreased mortality, especially for cardiovascular disease.
The evidence suggests significant differences in favorable and unfavorable health-related behaviors across racial and ethnic groups (Myers et al., 1995). Relatively few studies have attempted to estimate the contribution of behavioral differences to differences in outcomes such as mor-
tality. Otten and colleagues (1990) analyzed longitudinal data beginning in 1971 through 1975 for a national cohort of persons and estimated the percentage of the Black-White differential in mortality that could be attributed to six risk factors: smoking, alcohol intake, cholesterol level, body-mass index, and systolic blood pressure. They found that 31 percent of the mortality difference could be attributed to these factors, 38 percent to family income, and the remaining 31 percent was unexplained. Berkman and Mullen (1997) analyzed data from the Established Population for Epidemiologic Studies of the Elderly in New Haven and concluded the Black-White differences in health behaviors were unlikely to account for Black-White differences in health status. Similarly, an analysis of mortality differences between U.S.-born and foreign-born Mexicans and non-Hispanic Whites in the San Antonio Heart Study found that health-risk factors, including smoking, physical activity, and alcohol intake, added little to explaining differences beyond SEP (Wei et al., 1996). Lantz and colleagues (1998) attempted to assess the contribution of risk behaviors to health differences among socioeconomic groups and also found that smoking, alcohol intake, sedentary lifestyle, and relative body weight did not account for those differences in mortality. Though the importance of reducing unfavorable health behaviors and promoting favorable health behaviors throughout the general population is unquestioned, and differences in health behaviors probably contribute to racial and ethnic differences in health, health behaviors alone are unlikely to account for the bulk of the racial and ethnic differences in general health status.
Psychosocial Factors: Stress and Racism
Researchers also have long speculated that the stressful effects of racial and ethnic discrimination and marginalization in American society could underlie some of the racial differences in health. Indeed, in recent decades, studies have begun to describe increasingly complex physiologic pathways that connect stress with adverse health outcomes (Weiner, 1992). Much of the literature on the relationship between stress and racial differences in health has specifically addressed the role of stress in the risk for hypertension, which has a much higher prevalence among Blacks and is thought to be causally related to a range of psychological factors.
One issue that has been examined is whether race or SEP is the source of stress; conflicting results have been reported (Kessler and Neighbors, 1986; Cockerham, 1990; Fiscella and Franks, 1997). A substantial body of work has arisen from surveys in which people are asked about their perceptions related to racism or discrimination; then the correlation between their responses and an objective health outcome is measured. One of the principal barriers to this area of research is the subjective nature of stress,
and the circular reasoning associated with it: something is stressful if it makes someone feel stressed and vice versa. Again the results of these studies are inconclusive. It is also difficult to determine the direction of the causal relationship: poor health can create a predisposition to experience life as stressful (Kennedy et al., 1997; Jackson et al., 1996).
Another group of studies have examined patterns of coping styles and their relation to health outcomes (Krieger and Sidney, 1996; Strogatz et al., 1997). However, a complication of interpreting this research is that coping styles are not value free. In general, American society favors active coping, even in a situation in which one is lower in the power structure. Perhaps the most colorfully described coping style explored has been John Henryism (named for the folk hero) (James, 1994) —i.e., the tendency on the part of some Black men to adapt by overcompensating, to overcome racial stereotypes by laboring self-destructively and then succumbing to ill health from overwork.
A related set of studies looked at the effect on health outcomes of what might be termed an imposed coping style: workplace conditions. One study examined job-related decision latitude and job strain; for Black men, but not Black women, high decision latitude was associated with a lower prevalence of hypertension (Curtis et al., 1997). Another body of research has addressed whether similar measures of control in the workplace also underlie socioeconomic differences in health outcomes (e.g., Marmot et al., 1997). Finally, there are a substantial number of studies that have tried to reverse the causal chain, for example, using stress reduction as a treatment for hypertension (e.g., Schneider et al., 1995).
Access to Health Care
Historically, much of the public policy discussion about differences in health across sociodemographic groups has been devoted to the role of health care, especially differences in access—i.e., the extent to which an individual is able to obtain a reasonable level of medical services given his or her medical care needs. Although health insurance is an important determinant of access, it is not the only determinant. Other access factors include willingness of providers to accept certain forms of insurance, insurance co-payments, travel costs, the cultural competence of health care providers, and the knowledge of, and attitudes toward, the patient and the health of patients and their families. All these factors affect access regardless of whether an individual has health insurance. One of the primary reasons for focusing on health insurance, however, is the prominent role that the federal and state governments have played in providing health insurance, especially for vulnerable populations including the elderly and the poor.
We know of no empirical research that has attempted to estimate explicitly how much racial and ethnic differences in access contribute to differences in health status. For decades, the conventional wisdom in the public health community has been that most of the improvements in health in this century were the result of society-level improvements in nutrition and housing and public health activities such as improved sanitation, rather than the result of medical interventions (e.g., McKeown, 1979; McKinlay and McKinlay, 1977). More recently, however, researchers have criticized these conclusions, suggesting that, in fact, a significant part of the recent improvements in health are indeed the result of medical interventions (Bunker, 1995; Andrulis, 1998). Furthermore, a number of studies suggest that lack of insurance coverage is related to a range of clinical outcomes, including mortality (Braveman et al., 1989; Shea et al., 1992; Franks et al., 1993a, 1993b; Ayanian et al., 1993; Braveman et al., 1994).
Prenatal care is often a prominent part of discussions of LBW and infant mortality in part because prenatal care is one of the few direct interventions available in attempting to lower LBW and infant mortality rates. The percent of live births in which prenatal care began during the first trimester is 84 percent for Whites, 81 percent for Asians, 72 percent for Hispanics, 71 percent for Blacks, and 67.7 percent for American Indians (National Center for Health Statistics, 1998). Unfortunately the data do not support the assertion that prenatal care even if universally provided would dramatically change infant mortality rates (Mustard and Roos, 1994; Fiscella, 1995). In addition to representing a set of health-care services, use of prenatal care may also be an indicator of women who differ along a number of dimensions from women who obtain prenatal care late or not at all.
Clearly, there is wide consensus in this country in favor of providing reasonable access to health care for everyone, to maximize health and quality of life and minimize pain and suffering. The fundamental question of the role of medical care in eliminating differences across groups is important because it may predict the relative effectiveness of alternative strategies targeted to reduce racial and ethnic differences overall. There is a growing compilation of high-quality evidence of the impact of both medical and public health interventions on a range of diseases that are major determinants of mortality and likely causal factors in racial and ethnic differences in health. A reasonable conclusion is that access to effective medical interventions plays a role in the differences in health. How much of a role is unclear.
In the United States, health insurance is probably the major determinant of access, and health insurance serves primarily by reducing the out-of-pocket cost of medical services. Most working-age Americans are in-
sured by private insurance obtained through employment, and racial and ethnic differences in employment patterns underlie some of the differences in insurance patterns. The vast majority of elderly persons are primarily insured by Medicare with most elderly having secondary coverage by private insurance, often related to previous employment. The majority of children receive insurance coverage through their parents’ employment-obtained insurance, but a large minority of children are covered by Medicaid. The type of health insurance coverage a person has may be another important determinant of access because reimbursement rates vary by insurance type. In particular, Medicaid reimbursement rates are substantially lower than those of private insurers. As a result, providers may limit access to Medicaid recipients and the quality of services may also be lower (Braveman et al., 1994; Medicaid Access Study Group, 1994).
Among racial and ethnic minority groups in the United States, Hispanics have the lowest rates of health insurance coverage. In 1995, 30.8 percent of Hispanics younger than age 65 were uninsured, compared to 12.7 percent of non-Hispanic Whites (National Center for Health Statistics, 1998). Furthermore, there is substantial variation across Hispanic subpopulations; 17.8 percent of Puerto Ricans, 21.6 percent of Cubans, 35.4 percent of Mexican Americans, and 29 percent of other Hispanics were uninsured in 1995 (National Center for Health Statistics, 1998). As is true for the elderly in every racial and ethnic group, the vast majority of Hispanic elderly, though still a lower percentage than Whites and Blacks, are covered by Medicare; for example, in 1989, of those age 65 and older, 96 percent of non-Hispanic Whites, 95 percent of non-Hispanic Blacks, and 91 percent of Hispanics were covered by Medicare (General Accounting Office, 1992). There are also substantial differences in insurance coverage among Hispanic subgroups correlated with foreign birth. During 1989 and 1990, 40.8 percent of foreign-born Hispanics age 18 or older were uninsured compared to 24.8 percent of U.S.-born Hispanics (Thamer et al., 1997).
The distinctive pattern of insurance coverage among Hispanics below age 65 results from two factors (Valdez et al., 1993; General Accounting Office, 1992); first, Hispanics are more likely to work in lower paying jobs in smaller firms and in industries less likely to offer health insurance coverage; second, as a group, Hispanics, and especially Mexicans (only 19 percent of whom were covered by Medicaid), are more likely to reside in states with more stringent Medicaid eligibility criteria—e.g., Texas and Florida. Puerto Ricans, on the other hand, 36.3 percent of whom are cov-
ered by Medicaid, are concentrated in New York and New Jersey—two states with less restrictive Medicaid eligibility requirements.
Another indicator of access to health care is whether one has a regular source of medical care. By this measure as well, Hispanics have less access to care compared to Whites. In 1993, among persons ages 18 to 64, 71.9 percent of Hispanics reported a regular source of care compared to 84.2 percent of non-Hispanic Whites; for Mexicans the number was 69.5 percent (Bloom et al., 1997).
With the enactment of the legislation creating Medicare and Medicaid in 1965, substantially more Americans in general, Blacks in particular, began to enjoy the benefits of health insurance coverage. That legislation, along with the 1964 Civil Rights Act, played an important role in eliminating segregation in health-care facilities (Halperin, 1988; Thomson, 1997). Implementation of Medicare and Medicaid effected a substantial narrowing of gaps between Blacks and Whites in coverage and access (Davis et al., 1987). Nevertheless, some differences remain, mostly in the types of insurance coverage. Among persons younger than age 65 in 1996, 55.4 percent of non-Hispanic Blacks and 78.5 percent of non-Hispanic Whites had private insurance, 24.2 percent of Blacks and 7.5 percent of Whites had public insurance, and 18.9 percent of Blacks and 12.9 percent of Whites had no insurance (National Center for Health Statistics, 1998). Of those older than age 65, 42 percent of non-Hispanic Blacks and 77.9 percent of non-Hispanic Whites had private insurance in addition to Medicare.
Black adults report having a regular source of care at frequencies similar to Whites. Between ages 18 and 64, about 83.1 percent of non-Hispanic Blacks and 84.2 percent of non-Hispanic Whites report having a regular source of care, while above age 64, about 94 percent of both Whites and Blacks report having a regular source of care (Bloom et al., 1997; Cohen et al., 1997).
Asians and Pacific Islanders
Asians and Pacific Islanders in general are more likely to be uninsured than Whites. In 1996, 18.6 percent of Asians and Pacific Islanders younger than age 65 were uninsured compared to 15.4 percent of Whites (National Center for Health Statistics, 1998). In 1996, Asians and Pacific Islanders were slightly more likely (12.4 percent) to report coverage by Medicaid or another public program compared to Whites (9.3 percent) under age 65 and were less likely to have private insurance (67.8 percent
versus 74.2 percent). As with Hispanics, where a person was born also was a factor in the percentage uninsured. In 1989 and 1990, of those older than age 18, 20.9 percent of foreign-born Asians and Pacific Islanders were uninsured. Of that same age group, about 11 percent of both U.S.-born Asians and Pacific Islanders and U.S.-born Whites were uninsured (Thamer et al., 1997).
As with health status, there is significant variation in access across Asian and Pacific Islander subgroups. Comparing pooled data from the 1993–1996 NHIS on age-adjusted insurance rates for those younger than age 65, we found the variation striking. On average, about 19.2 percent were uninsured in this period and 68.1 percent had private insurance. Of the subgroups, Japanese had the highest rate of private insurance (86.5 percent) and the lowest rate of uninsured (10.3 percent). The rate of private insurance among Korean Americans was 56.7 percent; uninsured was 36.8 percent. An alternative measure of access is the length of time since a person last saw a physician; analysis of 1992–1994 NHIS data on this access measure revealed similar variation across Asian and Pacific Islander subpopulations (Kuo and Porter, 1998). Japanese and Vietnamese had the highest age-adjusted percentage of persons who had been seen by a physician within a year (76.7 percent); Koreans and Chinese had the lowest percentage (69.7 percent). On average, Asians and Pacific Islanders reported a slightly longer period of time since the last physician visit compared to Whites. Adjusting for age, 73.4 percent of Asians and Pacific Islanders had seen a physician within the last year compared to 79.6 percent of Whites.
American Indians and Alaska Natives
American Indians and Alaska Natives have had the option of receiving health care from the federal government since the early nineteenth century. In 1995, IHS operated 38 hospitals and 112 health centers, which, in general, offer a more limited range of services than other community health settings (Kunitz, 1996). Relatively little is known about the access to health care services for Native Americans who are not covered by the IHS. In 1987, the National Medical Expenditures Survey (NMES) was used to obtain information about American Indians and Alaska Natives eligible for IHS services. Results showed that 55 percent had no health insurance other than the IHS. Even the elderly population was less well covered than other elderly populations; only 85 percent were covered by Medicare, compared to over 96 percent for other groups (Cunningham and Schur, 1991).
Trends in Insurance Coverage
Analyses of gross trends in insurance coverage suggest an increase in the percentage of persons without insurance in recent decades. Determining detailed trends in insurance coverage across racial and ethnic groups, however, is difficult because different surveys often ask different questions about coverage, which may affect estimates, and because surveys may lack sufficient sample sizes to estimate changes in coverage over time for small groups. Estimates of the extent of the change in insurance coverage across racial and ethnic groups in recent decades vary depending on the data source and the time period. Berk and colleagues (1996) compared data from the 1977 National Medical Care Expenditures Survey, the 1987 NMES, and the 1989 and 1992 NHIS and found that for the entire U.S. population, although the percentage of uninsured increased (from 12.2 percent in 1977 to 17.1 percent in 1992), the rate of increase varied substantially across racial and ethnic groups. They estimated that the increase in uninsured Hispanics accounted for almost 40 percent of the total increase over the covered time periods, with Mexican Americans alone accounting for 27 percent of the increase.
Since the 1980s, the changes in the number of Americans without health insurance are not as striking. Comparison of data from the 1987 NMES with 1996 Medical Expenditures Panel Survey data on similar questions reveals increases in the rates of uninsured among the non-institutionalized population of Hispanics. In 1987, percentages were: Hispanics 31.5 percent and Blacks 22.0 percent. The comparable 1996 percentages were: Hispanics, 33.5 percent and Blacks, 22.9 percent (Short et al., 1989; Vistnes and Monheit, 1997). Comparison of NHIS data over a similar time period: in 1984, the percentages uninsured were Hispanics 29.0 percent, Asians and Pacific Islanders 17.8 percent, non-Hispanic Blacks 19.2 percent, and non-Hispanic Whites 11.6 percent; in 1996, Hispanics 31.6 percent, Asians and Pacific Islanders 18.6 percent, non-Hispanic Blacks 18.9 percent, and non-Hispanic Whites 12.9 percent (National Center for Health Statistics, 1998).
Analyses of data from the Current Population Survey between 1989 and 1996 revealed that the largest increase in percentage of persons uninsured was among Blacks (2.4 percent) (Carrasquillo et al., 1999). Among Hispanics, the percentage uninsured did not change significantly over this period, but because of the increase in the overall size of the Hispanic population, Hispanics accounted for a large increase in the absolute number of uninsured.
Quality of Care
Because of concerns about the rising costs of health care, public attention has expanded beyond access to address the issue of the quality of health-care services, and with increased attention has come increasing evidence of differences in the quality of health care provided across racial and ethnic groups. A large number of studies have documented differences in the usage of a range of health-care services among racial and ethnic groups, and most have found that Blacks are less likely to receive a number of procedures (Escarce et al., 1993; McBean and Gornick, 1994; Carlisle et al., 1995; Ford and Cooper, 1995). Schulman et al. (1999) conducted a study involving physicians and identical presentations of patients with chest pain; the only variation in the presentation was the patient’s race and gender. Differences by race and gender were found in the proportion of patients referred for cardiac catheterization. This compelling evidence suggests that bias at the level of the health-care system may be a reflection of biases in physicians’ clinical judgments and diagnoses. Differences in usage, however, can also be the result of a range of other factors including differences in access, differences in clinical characteristics of diseases, and differences in patient preferences. (An overview of racial and ethnic differences in utilization is included in the paper by Oddone and colleagues in this volume).
Relatively few studies have explicitly compared the quality of care across racial and ethnic groups. In an analysis of data based on the review of more than 30,000 New York State medical records in 1984, Burstin and colleagues (1992) found that compared to a reference group of Whites and others, Blacks were not more likely to experience substandard care as measured by the overall rate of adverse medical events, but were more likely to experience adverse medical events due to negligence. The difference, however, was no longer significant after accounting for lack of insurance and low income, both of which were related to greater likelihood of substandard care.
In another study based on reviews of medical records of a sample of Medicare recipients between 1981 and 1986, Kahn and colleagues (1994) found that within each category of hospital (rural, urban teaching, and urban non-teaching), Blacks were more likely to have poorer processes of care even after controlling for a range of patient and hospital characteristics. An interesting finding of this study was that overall both Blacks and patients from poor neighborhoods did not experience poorer care because they tended to receive care in urban teaching hospitals, which delivered better care than the other types of hospitals. More recently, Pappas and colleagues (1997) analyzed data on the 1990 rates of hospitalization for conditions that may be avoidable if appropriate ambulatory care were
provided and found that even among persons privately insured, Blacks under age 65 were more likely than Whites to experience avoidable hospitalizations.
There is little research addressing the quality of care of Hispanics, Asians and Pacific Islanders, or American Indians and Alaska Natives. In one of the few assessments of quality of care specifically addressing Hispanics, researchers at UCLA reviewed medical records from emergency room patients with isolated long-bone fractures and found that even after controlling for a range of potentially confounding factors, Hispanic patients were twice as likely to receive no pain medication (Todd et al., 1993). Another small study of pain control among cancer patients found a similar pattern of insufficient pain control for Hispanics and other minorities (Cleeland et al., 1997). A small number of the studies of utilization of procedures have also addressed racial and ethnic differences other than Black-White differences. For example, Carlisle and colleagues (1995) reviewed hospital discharge data for use of invasive cardiac procedures in Los Angeles between 1986 and 1988 and found that, as in other study samples, controlling for a range of sociodemographic and clinical variables, Blacks and Hispanics were also less likely to receive the procedures compared to Whites and Asians.
Managed care has raised new concerns. As more and more persons have become covered by managed-care organizations, fears have been raised that by placing barriers to access, managed-care organizations will exacerbate existing racial and ethnic differentials in access to health care and possibly increase differentials in quality of care, although few data are available on this topic (Rosenbaum et al., 1997b).
Any conclusion regarding the contribution of differences in health care access to differences in health requires substantial speculation. In general, many racial and ethnic minority groups have less access to health care than the majority population, and these differences are probably related to differences in health outcomes.
The Role of Minority Health-Care Professionals in Access and Provision
The relationship between an available supply of minority physicians and access to care has been buttressed by several studies that support the conclusion that minority physicians are much more likely to serve minority patients, and to serve poorer patients (Moy and Bartman, 1995; Komaromy et al., 1996). Advocates for increasing minority representation among health-care professionals make the case that minority professionals have a broad salutary effect on minority patients’ health (Nickens, 1991). For that reason, among others, underrepresentation of minorities in
the health professions has occupied a substantial part of the discussion of minority access to health care and minority health in general, particularly since the Civil Rights movement of the 1960s.
In 1964, only 2.2 percent of the 32,000 students enrolled in the nation’s then 83 allopathic medical schools were Black. The only two Black medical schools then in existence, Howard and Meharry, enrolled 76 percent of these students—i.e., on average, each of the other 81 schools enrolled only 1 Black student every two years (Hutchins et al., 1967). In 1971, the first year for which there are data for minority groups other than Blacks, only 19 Mexican Americans, 14 mainland Puerto Ricans, and 2 American Indian physicians graduated from U.S. medical schools. (These groups plus Blacks are the four groups considered by the Association of American Medical Colleges to be underrepresented in medicine). Such low enrollment numbers provided an obvious target for physicians and others seeking to obtain racial equity in the health profession; since then, substantial energy has been directed toward increasing minority representation in medicine (Nickens et al., 1994).
One way of quantifying racial and ethnic representation in medical professions is by calculating race/ethnic-specific physician-population ratios. For example, Libby et al. (1997) estimate that in 1990 there were 124 Hispanic, 69 Black, 46 American Indian, 875 Asian, and 241 non-Hispanic White active physicians per 100,000 persons. This would seem to create a somewhat artificial, some would say objectionable, implication—i.e., that only those of a given racial/ethnic group can or should treat members of that group. Obviously, that is not true. By quantifying minority representation, an indication of access to care is provided, and considering the fact that medical professions tend to rate high on professional indexes, an indication of the group’s progress on the socioeconomic ladder.
Although arguments over what is the appropriate physician-population ratio are far from settled, Hispanic, Black, and American Indian and Alaska Native physician supply numbers are far below any that could reasonably be advanced as adequate. At the other extreme, Asians now represent about one in five new entrants to U.S. allopathic medical schools, about six times their representation in the population at large (Association of American Medical Colleges, 1998a).
Health professions across the board have similar levels of underrepresentation of Hispanics, Blacks, and American Indians and Alaska Natives. Blacks, for example, are 3.6 percent of physicians, 3.1 percent of dentists, 1.7 percent of veterinarians, and 3.4 percent of podiatrists (Association of American Medical Colleges, 1998a). Underrepresented minorities are even more profoundly underrepresented among Ph.D.s in science than within health professions (Malcolm et al., 1998). Though hard data
are lacking, many in the health research community believe that minority researchers may be more likely to be interested in problems related to minority and disadvantaged populations than nonminorities. If this is true, then addressing the research agenda implicit in the many unanswered questions we pose in this paper will be much more difficult without a racially and ethnically diverse research workforce.
Acculturation is the process by which members of immigrant or other marginalized groups assimilate into the broader, majority culture. The range of processes for minority groups includes the immigration and assimilation experiences of Hispanics and Asians and Pacific Islanders, the road from slavery to the present for Blacks, the American Indian’s history of subjugation and repression, and the incorporation of indigenous Alaska Native and Hispanic populations. Unfortunately, our understanding of the acculturation process and how it affects health is still rudimentary, and researchers have yet to develop a clear and consistent framework that captures the complexity of the processes (Palinkas and Pickwell, 1995).
In contrast to many issues related to minority health, studies of acculturation have been focused primarily on non-Black minorities. Studies have typically used indicators such as foreign versus U.S. birth or fluency in English as measures of acculturation. More elaborate acculturation scales measure factors such as media preferences; friendship networks; ethnic identification of self, mother, and father; and subscription to values that are thought to be associated with the “traditional” culture of origin (e.g., Hazuda et al., 1988).
Typically studies in acculturation and its relation to health status measure the relationship between the degree of acculturation and outcomes such as health behaviors—e.g., smoking and use of prenatal care (Zambrana et al., 1997), obesity (Hazuda et al., 1988), low birth weight (Cobas et al., 1996), and rates of chronic diseases (Palinkas and Pickwell, 1995; Wei et al., 1996). Many of these studies have shown a correlation between adverse health effects and acculturation, but there are suggestions that the overall correlation may mask a more complex process (English et al., 1997). Superficially, it would appear that there is a clear-cut causal link between acquiring health-risk behaviors such as smoking and drug use and the fact that these behaviors are often more common in the United States than in immigrants’ countries of origin (Haynes et al., 1990); however, this explanation overlooks the fact that acculturation is not a passive acquisition process. Individuals interact with the majority culture and make choices about which aspects of the new culture to adopt. Immi-
grants often come to this country seeking the American lifestyle and so may choose elements of that lifestyle that are both obvious and available to them. A more nuanced understanding of acculturation as a reflection of active choices is needed.
Perhaps no topic covered in this paper elicits more heated and emotional debate than the question of the role of fundamental biologic and genetic differences as the cause of racial and ethnic differences in health. One reason for concern when this topic is raised is the lengthy history in this country of the misuse of science to support of racism (e.g., see Krieger, 1987). Because of that history, many reviews of racial differences in health have tended to dismiss genetic factors as important determinants of racial and ethnic differences in health. Recent advances in science, however, point to growing evidence that support a role for genetic factors in the etiology of a wide range of diseases, often via complex pathways in which multiple genetic factors are likely to interact with each other and with environmental factors. Although the role of genetic factors in determining the risk of disease in individuals is becoming clearer, the role of genetic factors in explaining differences in health across populations remains quite obscure (Baird, 1994). A detailed discussion of this issue is beyond the scope of this paper. However, worthy of note is that genetic factors alone cannot explain the trends in racial and ethnic differences in health during this century. In diseases in which genetic factors are known to play a role, these trends are likely to be explained by changes in environmental factors which then have their greatest impact on those persons who are at greatest risk genetically.
A recent review of the evidence on the role of genetic factors in explaining racial and ethnic differences in health of the elderly focused on Black-White differences in hypertension and diabetes, two genetically complex and environmentally influenced diseases that differ in risk between Blacks and Whites (Neel, 1997). The review concluded that the interaction of genetic and environmental factors is sufficiently complex that there is little prospect of disentangling the role of genetic factors in explaining racial differences in the near future.
Although clear-cut answers are unlikely to appear for decades, if ever, the volume of research addressing this issue is growing rapidly. For example, the apolipoprotein E (apoE)-e4 gene allele has been found in several studies to be a risk factor for Alzheimer’s disease, an important disease in the elderly population (Jarvik, 1997). Studies, however, have found that the apoE-e4 allele appears to be less closely related to the disease among Blacks (Farrer et al., 1997). A recent population-based longitudi-
nal study of a sample of elderly in New York City compared the risk for Alzheimer’s disease associated with apoE-e4 allele across racial and ethnic groups and found that although the presence of the allele was a risk factor for Alzheimer’s disease among Whites, it was not a risk factor among Hispanics and Blacks (Tang et al., 1998). Furthermore, the study found that both Blacks and Hispanics in the sample were at significantly greater risk of Alzheimer’s disease and speculated that other genes may account for this risk. With regard to other health factors, studies have suggested that genetically based racial differences in the metabolism of nicotine might be related to observed Black-White differences in smoking patterns (Perez-Stable et al., 1998; Caraballo et al., 1998). Regarding hypertension, a large body of literature has begun to address the role of various genetic factors in explaining racial differences in the etiology of that disease (Cooper and Rotimi, 1994). The potential for further study addressing the role of genetic factors in racial and ethnic differences in health is likely to increase greatly with the growing number of large population studies that include clinical and genetic data as well as socioeconomic and environmental data on sufficiently large samples of diverse racial and ethnic groups.
Environmental and Occupational Exposures
Over the past few decades, increasing attention has focused on differential environmental and occupational exposures to hazards across sociodemographic groups. Few studies have rigorously assessed the role that exposure to environmental hazards plays in racial and ethnic differences in health status. Several studies have addressed the location of hazardous waste sites. In 1983, the General Accounting Office reviewed the racial and economic characteristics of the communities surrounding four hazardous waste landfills and found that Blacks made up the majority of the population in three of the four sites (General Accounting Office, 1983). Four years later, a frequently cited and controversial report by the United Church of Christ Commission for Racial Justice (1987) also suggested that hazardous waste sites were more likely to be located in areas with large minority populations. A more recent analysis of hazardous waste sites did not find a distinctive pattern of locating these sites in communities with higher concentration of racial and ethnic minorities (Anderton et al., 1994).
Environmental exposures are not confined to hazardous waste sites, however. Exposure of children to lead from a variety of sources in a variety of settings may be an especially important health-risk factor that differs across racial and ethnic groups. Studies suggest that Black and Hispanic children are at much greater risk for exposure to lead than White
children (Brody et al., 1994; U.S. Department of Health and Human Services, 1988; Carter-Pockras et al., 1990). Although detailed data are not available for a wide range of environmental risk factors, the preponderance of evidence suggests that racial and ethnic minorities are exposed to higher levels of environmental risks (Institute of Medicine, 1999).
Employment patterns differ substantially across racial and ethnic groups, and some racial and ethnic minorities are more likely to hold jobs that expose individuals to higher risks for injury, illness, and death (Fried-man-Jimenez and Ortiz, 1994; Robinson, 1984). Relatively few studies have specifically addressed this issue, however, and none has specifically estimated the contribution of differences in occupation-related risk to overall differences in health. In the few studies that have addressed the issue, the findings vary across data sources, health outcomes, and time periods. National data from the 1970s and 1980s suggested that Blacks had higher rates of disabling occupational injuries than Whites (Robinson, 1984, 1987). Data on work-related injuries and illnesses in California during 1986 also suggested higher rates for Hispanics and Blacks compared to non-Hispanic Whites (Robinson, 1989). A review of occupational cancer epidemiology studies from the mid-1980s found a paucity of data on racial differences in occupational cancer risks but concluded that there was evidence suggesting higher risks for non-Whites (Kipen et al., 1991).
Other studies found mixed results on racial differences in occupational injuries. Data from the 1983–1987 NHIS revealed that working Blacks reported fewer injuries requiring medical attention or restriction of usual activities than Whites and that Blacks and Whites reported similar rates of injuries at work (Wagener and Winn, 1991). An analysis of data on a national sample of persons age 51–61 from 1992–1993 did not find significantly higher reporting of occupational injuries for either Blacks or Hispanics (Zwerling et al., 1996). A recent analysis of data on fatal occupational injuries in North Carolina, however, found that Blacks were more likely than Whites to die from an injury on the job and that much of the difference was due to racial differences in the types of jobs held (Loomis and Richardson, 1998), and national data on fatal injuries during 1980– 1985 also found a slightly higher risk for Black workers (Bell et al., 1990).
Although there appear to be racial and ethnic differences in levels of exposure to environmental and occupational risks, there are not enough data to come to any strong conclusion about the relative contribution of exposure to occupational and environmental hazards to overall racial and ethnic differences in health. (Although occupational and environmental factors probably do account for some racial and ethnic differences in health, we believe that they are unlikely to make a large contribution to the overall patterns found.) Much remains to be learned about both occupational and environmental health risks, and more rigorous studies of
these risks across racial and ethnic groups will greatly improve our understanding of causal pathways for differences in health (Northridge and Shepard, 1997; Johnson and Coulberson, 1993; Institute of Medicine, 1999).
…with economic and industrial progress of the colored population its death-rate will gradually approach nearer to that of the White population.
—Trask, U.S. Public Health Service (1917)
…The prospect of improvement for Negro health that would result from changes in the economic and occupational environment is much less optimistic. It seems inescapable that low incomes derived from unskilled occupations will show a large degree of prevalence among Negroes for a long time.
—Tibbits, U.S. Public Health Service (1937)
Predicting the future trends of racial and ethnic differences in health in the United States has never been easy. So many variables come into play to complicate our predictions, and the health trends are affected by trends in every other dimension of life. The fundamental problem is that we do not understand the present very well. Key questions remain unanswered:
Why are some Black mortality and morbidity rates so high, with Black infant mortality, in particular, seemingly so refractory to the usually mitigating effects of higher SEP?
Why are American Indian mortality rates not worse than they are, given the historical and current deprivations suffered by Indian peoples?
Why are Hispanic mortality rates so low, seemingly so refractory to the aggravating effects of low SEP?
What are the intermediate steps that translate low SEP into poor health and do they vary among different racial and ethnic groups? How do psychological and social factors such as self-image and values interact with the more material aspects of deprivation?
Although we have speculated about the answers to some of these questions, definitive answers are at present beyond the reach of the scientific evidence. These are large, complex questions, and subsumed under them are hundreds of potential research questions that need to be asked and answered. Having looked back to the recent past, while we are reas-
sured by general trends of improvements in health, perhaps the single most disturbing finding was the unprecedented downturn in life expectancy for Blacks during the 1980s. That downturn provides a cautionary lesson for future predictions: the health of racial and ethnic minorities in the United States will not inevitably improve with the mere passage of time. A multitude of factors, many outside of the control of the public health and medical communities, will interact to determine the trajectories, and the health of populations can deteriorate, as we have seen in recent years of transition in Russia (Walberg et al., 1998). Accepting our tenuous insight into current mechanisms of minority health status, it is nonetheless useful to look ahead.
Trends in Racial Differences in Socioeconomic Position Suggest Increases in Racial Health Disparities
Socioeconomic position will continue to play a large role in explaining many of the health differences among racial and ethnic groups, and evidence of growing Socioeconomic inequality does not bode well for the future (Karoly, 1996; Wolff, 1998; Smith, Volume II, Chapter 4). Blacks, American Indians, and Hispanics all have high poverty rates and substantially lower levels of educational attainment than do Whites and Asians and Pacific Islanders. Moreover, inasmuch as educational attainment drives income in our society, the economic trajectory of these minority groups is worrisome as well. To the extent that disparities in economic resources and education increase, the health of racial and ethnic minorities who remain in a disadvantaged SEP will worsen. Particularly disturbing is the large number of minority children who live in poverty. The high rates of childhood poverty may be setting in place trajectories for a cohort of persons who will carry with them high risks for poor health if and when they reach middle and old age.
Hispanic Health May Worsen with Acculturation
The health status of Hispanics is of particular concern given their growing numbers in the population. The anomalous current combination of low SEP and relatively low death rates among Hispanics may amount to a form of “epidemiologic levitation.” Hispanic high school completion rates are even lower than those for Blacks and American Indians. Because the acculturation literature suggests that the protective effect of recent immigration wears off with time, there is a real possibility that Hispanic health status could worsen.
Changing Definitions of Race and Ethnicity May Reshape Our Understanding of Racial Differences in Health
Another source of uncertainty is presented by the now familiar projection that by 2050 minorities will represent about half of the U.S. population. But the relatively high intermarriage rates for American Indians and Alaska Natives, Hispanics, and Asians and Pacific Islanders with the White population suggests that we will not have a mosaic of neatly partitioned racial and ethnic groups; rather, we will have what has been termed a “beige continuum” in which the lines among groups are quite unclear.
This projection for American Indians and Alaska Natives, Hispanics, and Asians and Pacific Islanders, however, contrasts with the likely scenario for Blacks. Blacks have a relatively low intermarriage rate, suggesting increased social partitioning for Blacks. Such a potential for a future nation divided into beige and Black will force us to reconsider our simplistic notions of race and ethnicity and its relationship to health (Lind, 1998).
Racial and Ethnic Groups Will Become More Heterogeneous
Minority elderly populations are growing at a rate faster than the White population, and larger numbers of racial and ethnic minorities will reach old age with distinctive life and health histories. The characteristics of immigrants will continue to change, each cohort bringing distinctive health and social characteristics. Similarly, as has happened throughout the history of this country, different cohorts of Blacks will have increasingly diverse race-related life experiences. The urban American Indian population may grow and present unique challenges. With the increasing heterogeneity within racial and ethnic groups, sweeping generalizations about a group’s health will become of increasingly limited value as a means to explain causal pathways and promote better health outcomes.
The Research Agenda for the Future
As difficult as it is to predict future trends, it is equally difficult to determine the best course of action for reducing health disparities. What is clear is that more research is needed.
Study the Role of Culture in Determining Health
We know little about the processes underlying the acculturation of immigrants and other minority populations and the relationship between these processes and adoption of health-related behaviors. With the expec-
tation that Hispanics will become the largest minority group during the early twenty-first century, how acculturation occurs may have a significant impact on the health of that population. Improved understanding of acculturation will be relevant not only for other immigrant populations such as Asians and Pacific Islanders but also for other groups including American Indians and Alaska Natives and Blacks.
Assess the Effectiveness of Public Health Interventions in Reducing Disparities
In completing this review, we were surprised by the dearth of literature that rigorously addresses the extent to which recent large public health initiatives have affected racial differences in health. For example, how much have the interventions to control blood pressure in Blacks contributed to recent improvements in cardiovascular mortality? Such analyses will provide essential information on which broad health interventions have an effect on disparities, why they do, and how; they will help in the planning of future efforts to reduce disparities. Recent efforts in Europe to understand the evidence on the effect of public health interventions on socioeconomic disparities in health may serve as a model (e.g., Arblaster et al., 1996)
Unify Research on Causal Factors and Their Biological Pathways
The overwhelming majority of the research on causal pathways for racial differences has been piecemeal, artificially separating groups of factors that are closely, and often causally, related to each other as well as to health outcomes. For example, it is unlikely education and income will change independent of nutrition and access to health care. The traditional, piecemeal approach limits our ability to assess the effectiveness of changing various factors to reduce disparities; more methodological and empirical work is needed to provide a more realistic understanding of the interactive effects of causal factors. This will likely require the development of a series of from-birth cohort study populations from diverse racial and ethnic groups. Similar cohorts in England and other European countries have greatly increased our understanding of how SEP affects health across the life cycle.
In the absence of changes in the social environment, isolated public health and medical interventions are unlikely to eliminate racial and ethnic disparities, especially in light of the importance of SEP in determining
health (Marmot, 1998). Nevertheless, we have chosen to restrict our comments here to health-care policy options as traditionally defined.
Improve Health Care Access and Monitor Differences in Health-Care Quality
Differentials in access to health care play a role in perpetuating racial and ethnic differences in health, and recent initiatives aimed at improving access, especially for poor children, are likely to affect the trajectories of poor minority populations. Because access so heavily relies on health insurance, of particular concern is the high percentage of Hispanics, especially immigrants, who are uninsured. We also cannot ignore the evidence of racial and ethnic differences in quality of care delivered. Suggestions that race and ethnicity be among the variables included in efforts to monitor the quality of health care in managed care organizations warrant serious consideration (Smith, 1998; Lavizzo-Mourey and Mackenzie, 1996).
Promote Targeted Health Behavior Interventions
Although health-risk behaviors alone probably cannot account for the disparities in health among racial and ethnic groups, they likely contribute to the differences; and we can do something about them. A growing body of research is available on public health interventions that are most likely to work in increasing the adoption of health-promoting behaviors and the cessation of health-reducing behaviors. Certain trends warrant consideration for targeted public health interventions, in particular recent increases in smoking among Black and Hispanic adolescents.
Increase Minority Representation Among Health-Care Providers
The low educational attainment of Blacks and Hispanics serves as a powerful barrier to producing a diverse health professional and scientific workforce (Nickens et al., 1994). Higher education institutions have been able to admit substantial numbers of Hispanics, Blacks, and American Indians by engaging in affirmative action. However, that these groups score substantially lower than Whites and Asians on standardized tests has provided numerical ammunition for the opponents of affirmative action to assert that it is both unfair and admits unqualified students (Association of American Medical Colleges, 1998b). Critics of affirmative action have gone further, attacking the legitimacy of any minority-targeted programming, which complicates national ameliorative efforts. One of the issues that underlies the battle over affirmative action is a debate
about the nature of race and ethnicity in contemporary America. Implicit in the arguments that race should not be a consideration in, for example, educational programming decisions, is an assertion that race no longer is determinative of life chances and thus is not a relevant way to distinguish among people or is not a suitable criterion on which to give minorities a “plus factor” in, for example, an admissions process. This general assertion that “race no longer matters” is not sustained by this review; with regard to health status, race and ethnicity remain important factors.
Future trends for racial and ethnic disparities are uncertain, but we see little evidence for optimism that disparities will be eliminated in the near future. Most of the research evidence suggests that the current relative health status of some minority groups is substantially “locked in” by poorly understood economic, social, educational, and medical structural disadvantages. The only viable key is an understanding of how all these variables interact and the national resolve to open the door to change.
We are particularly grateful for the assistance of the following individuals: at the National Center for Health Statistics: Charlene Brock, Harnethia Cousar (Research Library); Dawn Hutcherson, Kathryn Porter, Yolanda Cowan (DHES); Anjum Hajat, Jacqueline Lucas, Ann Hardy (DHIS); Jeff Mauer (DVS); Diane Wagener, Elsie Pamuk, Diane Makuc (OAEHP); Jennifer Madans (DVHSS); and Ed Sondik, Director; at the Agency for Health Care Research and Quality: Jessica Vistnes; at the Association of American Medical Colleges: Willa Owens, Lois Bergeisen; and at other institutions: Jay Kaufman.
Adler, N., T.Boyce, M.Chesney, S.Cohen, S.Folkman, R.Kahn, and S.Syme 1994 Socioeconomic status and health: The challenge of the gradient. American Psychologist 49:15–24.
Aguirre-Molina, M., and R.Caetano 1994 Alcohol use and alcohol-related issues. In Hispanic Health in the United States: A Growing Challenge, C.Molina and M.Aguirre-Molina, eds. Washington, D.C.: American Public Health Association.
Alaimo, K., R.Briefel, E.Frongillo, and C.Olson 1998 Food insufficiency exists in the United States: Results from the Third National Health and Nutrition Examination Survey (NHANES III). American Journal of Public Health 88:419–426.
Aluli, N. 1991 Prevalence of obesity in a native Hawaiian population. American Journal of Clinical Nutrition 53:1556–1560S.
Anderson, N., and C.Armstead 1995 Toward understanding the association of socioeconomic status and health: A new challenge for the biopsychosocial approach. Psychosomatic Medicine 57:213–225.
Anderson, R., P.Sorlie, E.Backlund, N.Johnson, and G.Kaplan 1997 Mortality effects of community socioeconomic status. Epidemiology 8:42–47.
Anderton, D., B.Anderson, J.Oakes, and M.Fraser 1994 Environmental equity: The demographics of dumping. Demography 31(2):229–248.
Andrulis, D. 1998 Access to care is the centerpiece of the elimination of socioeconomic disparities in health. Annals of Internal Medicine 129:412–416.
Angel, J., and D.Hogan 1994 The demography of minority aging populations. In Minority Elders: Five Goals Toward Building a Public Policy Base. Washington, D.C.: Gerontological Society of America.
Antonovsky, A. 1967 Social class, life expectancy, and overall mortality. Milbank Memorial Fund Quarterly 45:31–73.
Arblaster, L., M.Lambert, V.Entwistle, M.Forster, D.Fullerton, T.Sheldon, and I.Watt 1996 A systematic review of the effectiveness of health services interventions aimed at reducing inequalities in health. Journal of Health Services Research and Policy 1(2):93– 103.
Association of American Medical Colleges (AAMC) 1998a Minority Students in Medical Education: Facts and Figures XI. Washington, D.C.: Association of American Medical Colleges.
1998b Questions and Answers on Affirmative Action in Medical Education. Washington, D.C.: Association of American Medical Colleges.
Atlanta University 1896 Mortality Among Negroes in Cities. Atlanta University Publications No. 1. Atlanta: Atlanta University Press.
1906 The Health and Physique of the Negro American. Atlanta University Publications No. 11. Atlanta: Atlanta University Press.
Ayanian, J., B.Kohler, T.Abe, and A.Epstein 1993 The relation between health insurance coverage and clinical outcomes among women with breast cancer. New England Journal of Medicine 329:326–331.
Baird, P. 1994 The role of genetics in population health. In Why Are Some People Healthy and Others Not? The Determinants of Health of Populations, R.Evans, M.Barer, and T. Marmor, eds. New York: Aldine De Gruyter.
Barker, D. 1994 Mothers, Babies, and Disease in Later Life. London: British Medical Journal Publishing Group.
Bell, C., N.Stout, T.Bender, C.Conroy, W.Crouse, and J.Myers 1990 Fatal occupational injuries in the United States, 1980 through 1985. Journal of the American Medical Association 263:3047–3050.
Bell, R., E.Mayer-Davis, Y.Jackson, and C.Dresser 1997 An epidemiologic review of dietary intake studies among American Indians and Alaska Natives: Implications for heart disease and cancer risk. Annals of Epidemiology 7:229–240.
Berk, M, A.Albers, and C.Schur 1996 The growth of the U.S. uninsured population: Trends in Hispanic subgroups, 1977–1992. American Journal of Public Health 86:572–575.
Berkman, L., and J.Mullen 1997 How health behaviors and the social environment contribute to health differences between Black and White older Americans. In Racial and Ethnic Differences in the Health of Older Americans, L.Martin and B.Soldo, eds. Washington, D.C.: National Academy Press.
Blane, D., E.Brunner, and R.Wilkinson, eds. 1996 Health and Social Organization: Toward a Health Policy for the 21st Century. New York: Routledge.
Bloom, B., G.Simpson, R.Cohen, and P.Parsons 1997 Access to health care. Part 2: Working-age adults. National Center for Health Statistics. Vital Health Statistics 10(197).
Braveman, P., G.Oliva, M.Miller, R.Reiter, and S.Egerter 1989 Adverse outcomes and lack of health insurance among newborns in an eight-county area of California, 1982 to 1986. New England Journal of Medicine 321:508– 513.
Braveman, P., V.Schaff, S.Egerter, T.Bennett, and W.Schecter 1994 Insurance-related differences in the risk of ruptured appendix. New England Journal of Medicine 331:444–449.
Brody, D., J.Pirkle, R.Kramer, K.Flegal, T.Matte, E.Gunter, and D.Paschal 1994 Blood lead levels in the U.S. population. Phase 1 of the Third National Health and Nutrition Examination Survey (NHANES III, 1988 to 1991). Journal of the American Medical Association 272:277–283.
Broussard, B., A.Johnson, J.Himes, R.Fichtner, and F.Hauck, et al. 1991 Prevalence of obesity in American Indians and Alaska Natives. American Journal of Clinical Nutrition 53:1535–15425.
Bunker, J. 1995 Medicine matters after all. Journal of the Royal College of Physicians of London 29(2):105–112.
Bunker, J., D.Gomby, and B.Kehrer, eds. 1989 Pathways to Health—The Role of Social Factors. Menlo Park, Calif.: Henry J.Kaiser Family Foundation.
Burstin, H., S.Lipsitz, and T.Brennan 1992 Socioeconomic status and risk for substandard medical care. Journal of the American Medical Association 268:2383–2387.
Caraballo, R., G.Giovino, T.Pechacek, and P.Mowery, et al. 1998 Racial and ethnic differences in serum cotinine level of cigarette smokers. Third National Health and Nutrition Examination Survey 1988–1991. Journal of the American Medical Association 280:135–139.
Carlisle, D., B.Leake, and M.Shapiro 1995 Racial and ethnic differences in the use of invasive cardiac procedures among cardiac patients in Los Angeles County, 1986 through 1988. American Journal of Public Health 85(3):352–356.
Carrasquillo, O., D.Himmelstein, S.Woolhandler, and D.Bor 1999 Going bare: Trends in health insurance coverage, 1989 through 1996. American Journal of Public Health 89:36–42.
Carter-Pokras, O., J.Pirkle, G.Chavez, and E.Gunter 1990 Blood lead levels of 4–11-year-old Mexican American, Puerto Rican, and Cuban children . Public Health Reports 105:388–393.
Centers for Disease Control and Prevention (CDC) 1992 Chronic Disease in Minority Populations. Atlanta: Centers for Disease Control and Prevention.
Chen, V. 1993 Smoking and the health gap in minorities. Annals of Epidemiology 3:159–164.
Clark, D. 1997 U.S. trends in disability and institutionalization among older Blacks and Whites. American Journal of Public Health 87:438–440.
Cleeland, C, R.Gonin, L.Baez, P.Loeher, and K.Pandya 1997 Pain and treatment of pain in minority patients with cancer. Annals of Internal Medicine 127:813–816.
Cobas, J., H.Balcazar, M.Benin, V.Keith, and Y.Chong 1996 Acculturation and low-birthweight infants among Hispanic women: A reanalysis of HHANES data with structural equation models. American Journal of Public Health 86:394–396.
Cockerham, W. 1990 A test of the relationship between race, socioeconomic status, and psychological distress. Social Science and Medicine 31:1321–1326.
Cohen, R., B.Bloom, G.Simpson, and P.Parsons 1997 Access to health care. Part 3: Older adults. National Center for Health Statistics. Vital Health Statistics 10(198).
Cooper, R., and C.Rotimi 1994 Hypertension in populations of West African origin: Is there a genetic predisposition? Journal of Hypertension 12:215–227.
Corti, M., J.Guralnik, L.Ferrucci, G.Izmirlian, S.Leveille, M.Pahor, H.Cohen, C.Pieper, and R.Havlik 1999 Evidence for a Black-White crossover in all-cause and coronary heart disease mortality in an older population: The North Carolina EPESE. American Journal of Public Health 89:308–314.
Cowie, C., M.Harris, R.Silverman, E.Johnson, and K.Rust 1993 Effect of multiple risk factors on differences between Blacks and Whites in the prevalence of non-insulin dependent diabetes mellitus in the United States. American Journal of Epidemiology 137(7):719–732.
Cunningham, P., and C.Schur 1991 Health care coverage: Findings from the Survey of American Indians and Alaska Natives. DHHS Pub. No. (PHS) 91–0027. National Medical Expenditures Survey Research Findings 8, Agency for Health Care Policy and Research. Rockville, Md.: Public Health Service.
Curtis, A., S.James, T.Raghunathan, and K.Alcser 1997 Job strain and blood pressure in African Americans: The Pitt County Study. American Journal of Public Health 87(8):1297–1302.
David, R., and J.Collins 1997 Differing birth weight among infants born of U.S.-born Blacks, African-born Blacks, and U.S.-born Whites. New England Journal of Medicine 337:1209–1214.
Davis, K., M.Lillie-Blanton, B.Lyons, F.Mullan, N.Powe, and D.Rowland 1987 Health care for Black Americans: The public sector role. Milbank Memorial Fund Quarterly 65(Suppl 1):213–247.
DeJong, G., A.Batavia, and R.Griss 1989 America’s neglected health minority: Working-age persons with disabilities. Milbank Memorial Fund Quarterly 67(Suppl 2, Part 2):311–351.
Dorn, H. 1940 The Health of the Negro, A Research Memorandum. Carnegie-Myrdal Study. New York: Schomburg Center for Research in Black Culture.
Dowd, J., and V.Bengston 1978 Aging in minority populations: An examination of the double jeopardy hypothesis. Journal of Gerontology 33(3):427–436.
Dressler, W. 1993 Health in the African American community: Accounting for health inequalities. Medical Anthropology Quarterly 7:325–345.
DuBois, W.E.B. 1899 The Philadelphia Negro: A Social Study. Philadelphia: University of Pennsylvania Press.
English, P., M.Kharrazi, and S.Guendelman 1997 Pregnancy outcomes and risk factors in Mexican Americans: The effect of language use and mother’s birthplace. Ethnicity and Disease 7:229–240.
Erikson, P., R.Wilson, and I.Shannon 1995 Years of healthy life. Statistical Notes No. 7. Hyattsville, Md.: Centers for Disease Control and Prevention, National Center for Health Statistics.
Escarce, J., K.Epstein, D.Colby, and J.Schwartz 1993 Racial differences in the elderly’s use of medical procedures and diagnostic tests. American Journal of Public Health 83:948–954.
Ettner, S. 1996 New evidence on the relationship between income and health. Journal of Health Economics 15:49–66.
Evans, R., M.Barer, and T.Marmor, eds. 1994 Why Are Some People Healthy and Others Not? The Determinants of Health of Populations. New York: Aldine De Gruyter.
Ewbank, D. 1987 History of Black mortality and health before 1940. Milbank Memorial Fund Quarterly 5:100–128.
Fang, J., T.Madhavan, and M.Alderman 1996 The association between birthplace and mortality from cardiovascular causes among Black and White residents of New York City. New England Journal of Medicine 335:1545–1551.
1997 Nativity, race and mortality: Influence of region of birth on mortality of U.S.-born residents of New York City. Human Biology 69(4):533–544.
Farrer, L., L.Cupples, J.Haines, B.Hyman, and W.Kukull 1997 Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease: A metaanalysis. Journal of the American Medical Association 278:1349–1356.
Feinstein, J. 1993 The relationship between socioeconomic status and health—A review of the literature. Milbank Memorial Fund Quarterly 71:279–323.
Ferraro, K., and M.Farmer 1996 Double jeopardy to health hypothesis for African Americans: Analysis and critique. Journal of Health and Social Behavior 37:27–43.
Fiscella, K. 1995 Does prenatal care improve birth outcomes? A critical review. American Journal of Obstetrics and Gynecology 85(3):468–479.
Fiscella, K., and P.Franks 1997 Does psychological distress contribute to racial and socioeconomic disparities in mortality? Social Sciences and Medicine 45:1805–1809.
Flegal, K., M.Carroll, R.Kuczmarski, and C.Johnson 1998 Overweight and obesity in the United States: Prevalence and trends, 1960–1994. International Journal of Obesity 22:39–47.
Ford, E., and R.Cooper 1995 Racial and ethnic differences in health care utilization of cardiovascular procedures: A review of the evidence. Health Services Research 30:237–251.
Franks, P., C.Clancy, and M.Gold 1993a Health insurance and mortality. Evidence from a national cohort. Journal of the American Medical Association 270:737–741.
Franks, P., C.Clancy, M.Gold, and P.Nutting 1993b Health insurance and subjective health status: Data from the 1987 National Medical Expenditure Survey. American Journal of Public Health 83:1295–1299.
Freeman, V., and L.Martin 1998 Understanding trends in functional limitations among older Americans. American Journal of Public Health 88:1457–1462.
Friedman-Jimenez, G., and J.Ortiz 1994 Occupational health. In Hispanic Health in the U.S.: A Growing Challenge, C.Molina and M.Aguirre-Molina, eds. Washington, D.C.: American Public Health Association.
Garber, A. 1989 Pursuing the links between socioeconomic factors and health: Critique, policy implications, and directions for future research. In Pathways to Health: The Role of Social Factors, J.Bunker, D.Gomby, and B.Kehrer, eds. Menlo Park, Calif.: Henry J.Kaiser Family Foundation.
General Accounting Office (GAO) 1983 Siting of Hazardous Waste Landfills and Their Correlation with Racial and Ethnic Status of Surrounding Communities. RCED-83–168. Washington, D.C.: General Accounting Office.
1992 Hispanic Access to Health Care: Significant Gaps Exist. PEMD-92–6. Washington, D.C.: General Accounting Office.
Geronimus, A., J.Bound, T.Waidmann, M.Hillemeire, and P.Burns 1996 Excess mortality among Blacks and Whites in the United States. New England Journal of Medicine 335:1552–1558.
Gibson, R. 1991 Age-by-race differences in the health and functioning of elderly persons. Journal of Aging and Health 3:335–351.
Grossman, D., J.Krieger, J.Sugarman, and R.Forquera 1994 Health status of urban American Indians and Alaskan Natives. Journal of the American Medical Association 271:845–850.
Guralnik, J., K.Land, D.Blazer, G.Fillenbaum, and L.Branch 1993 Educational status and active life expectancy among older Blacks and Whites. New England Journal of Medicine 329:110–116.
Haan, M., and T.Camacho 1987 Poverty and health: Prospective evidence from the Alameda County Study. American Journal of Epidemiology 125:989–998.
Haan, M., and G.Kaplan 1985 The contribution of SEP to minority health. In Report of the Secretary’s Task Force on Black and Minority Health: Crosscutting Issues in Minority Health. Washington, D.C.: U.S. Department of Health and Human Services.
Hahn, R., and S.Eberhardt 1995 Life expectancy in four U.S. racial/ethnic populations: 1990. Epidemiology 6(4):350– 355.
Halperin, E. 1988 Desegregation of hospitals and medical societies in North Carolina. New England Journal of Medicine 318:58–63.
Hart, C, G.Smith, and D.Blane 1998 Inequalities in mortality by social class measured at three stages of lifecourse. American Journal of Public Health 88:471–474.
Hayes-Bautista, D. 1992 Latino health indicators and the underclass model: From paradox to new policy models. In Health Policy and the Hispanic, A.Furino, ed. Boulder, Colo.: Westview Press.
Haynes, S., C.Harvey, H.Montes, H.Nickens, and B.Cohen 1990 Patterns of cigarette smoking among Hispanics in the United States: Results from the HHANES 1982–84. American Journal of Public Health 80(Suppl):47–54.
Hazuda, H., S.Haffner, M.Stern, and C.Eifler 1988 Effects of acculturation and socioeconomic status on obesity and diabetes in Mexican Americans. American Journal of Epidemiology 128:1289–1301.
Holland, D., and G.Perrott 1938 The health of the Negro. Milbank Memorial Fund Quarterly 16:5–38.
Hoyert, D., and H.Kung 1997 Asian and Pacific Islander mortality, selected states, 1992. Monthly Vital Statistics Report 46(1 Suppl):1–63.
Hutchins, E., J.Reitman, and D.Klaub 1967 Minorities, manpower, and medicine. Journal of Medical Education 42:809–821.
Idler, E., and Y.Benyamini 1997 Self-rated health and mortality: A review of twenty-seven community studies. Journal of Health and Social Behavior 38:21–37.
Indian Health Service (IHS) 1991 Trends in Indian Health, 1991. Rockville, Md.: Indian Health Service, U.S. Department of Health and Human Services.
1996 Trends in Indian Health, 1996. Rockville, Md.: Indian Health Service, U.S. Department of Health and Human Services.
Institute of Medicine (IOM) 1999 Toward Environmental Justice: Research, Education, and Health Policy Needs. Washington, D.C.: National Academy Press.
Jackson, J., T.Brown, D.Williams, M.Torres, S.Sellers, and K.Brown 1996 Racism and the physical and mental health status of African Americans: A thirteen year national panel study. Ethnicity and Disease 6:132–147.
James, S. 1994 John Henryism and the health of African Americans. Culture, Medicine and Psychiatry 18:163–182.
James, W., M.Nelson, A.Ralph, and S.Leather 1997 The contribution of nutrition to inequalities in health. British Medical Journal 314:1545–1549.
Jarvik, G. 1997 Genetic predictors of common disease: Apolipoprotein E genotype as a paradigm. Annals of Epidemiology 7:357–362.
John, R. 1996 Demography of American Indian elders: Social, economic, and health status. In Changing Numbers, Changing Needs: American Indian Demography and Public Health, G.Sandefur, R.Rindfuss, and B.Cohen, eds. Washington, D.C.: National Academy Press.
Johnson, B., and S.Coulberson 1993 Environmental epidemiologic issues and minority health. Annals of Epidemiology 3:175–180.
Kahn, K.L., M.L.Pearson, E.R.Harrison, K.A.Desmond, W.H.Rogers, L.V.Rubenstein, R.H.Brook, and E.B.Keeler 1994 Health care for Black and poor hospitalized Medicare patients. Journal of the American Medical Association 271(15):1169–1174.
Karoly, L. 1996 Anatomy of the U.S. income distribution: Two decades of change. Oxford Review of Economic Policy 12(1):77–96.
Kaufman, J., and R.Cooper 1995 In search of the hypothesis. Public Health Reports 110:662–666.
Kaufman, J., R.Cooper, and D.McGee 1997 Socioeconomic status and health in Blacks and Whites: The problem of residual confounding and the resiliency of race. Epidemiology 8:621–628.
Kaufman, J., A.Long, Y.Liao, R.Cooper, and D.McGee 1998 The relation between income and mortality in U.S. Blacks and Whites. Epidemiology 9:147–155.
Keil, J., S.Sutherland, R.Knapp, and H.Tyroler 1992 Does equal socioeconomic status in Black and White men mean equal risk of mortality? American Journal of Public Health 82:1133–1136.
Kennedy, B., I.Kawachi, K.Lochner, C.Jones, and D.Prothrow-Stith 1997 (Dis)respect and Black mortality. Ethnicity and Disease 7:207–214.
Kessler, R., and H.Neighbors 1986 A new perspective on the relationships among race, social class, and psychological distress. Journal of Health and Social Behavior 27:107–115.
Kington, R., D.Carlisle, D.McCaffrey, H.Myers, and W.Allen 1998 Racial differences in functional status among elderly U.S. migrants from the south. Social Sciences and Medicine 47:831–840.
Kington, R., and J.Smith 1997 Socioeconomic status and racial and ethnic differences in functional status associated with chronic diseases. American Journal of Public Health 87:805–810.
Kipen, H., D.Wartenberg, P.Scully, and M.Greenberg 1991 Are non-Whites at greater risk for occupational cancer? American Journal of Indian Medicine 19:76–74.
Kochanek, K., J.Maurer, and H.Rosenberg 1994 Why did Black life expectancy decline from 1984 through 1989 in the United States? American Journal of Public Health 84:938–944.
Komaromy M., K.Grumbach, M.Drake, K.Vranizan, N.Lurie, D.Keane, and A.Bindman 1996 The role of Black and Hispanic physicians in providing health care for underserved populations. New England Journal of Medicine 334:1305–1310.
Kovar, M., and G.Poe 1985 The National Health Interview Survey design, 1973–1984, and procedures, 1975– 1993. Vital and Health Statistics. Series 1, No. 8. DHHS Pub. No. (PHS) 85–1320. National Center for Health Statistics. Washington, D.C.: U.S. Government Printing Office.
Krieger, N. 1987 Shades of difference: Theoretical underpinnings of the medical controversy on Black/White differences in the United States, 1830–1870. International Journal of Health Services 17:259–278.
Krieger, N., and S.Sidney 1996 Racial discrimination and blood pressure: The CARDIA study of young Black and White adults. American Journal of Public Health 86:1370–1378.
Krieger, N., D.Williams, and N.Moss 1997 Measuring social class in U.S. public health research: Concepts, methodologies and guidelines. Annual Review of Public Health 18:341–378.
Kumanyika, S. 1993 Special issues regarding obesity in minority populations. Annals of Internal Medicine 119:650–654.
Kunitz, S. 1996 The history and politics of U.S. health care policy for American Indians and Alaskan Natives. American Journal of Public Health 86(10):1464–1473.
Kuo, J., and K.Porter 1998 Health status of Asian Americans: United States, 1992–1994. Advance Data from Vital and Health Statistics; no. 298. Hyattsville, Md.: National Center for Health Statistics.
The Lancet 1995 The unequal, the achievable, and the champion. The Lancet 345:1061–1062.
Lantz, P., J.House, J.Lepkowski, D.Williams, R.Mero, and J.Chen 1998 Socioeconomic factors, health behaviors, and mortality. Results from a nationally representative prospective study of U.S. adults. Journal of the American Medical Association 279:1703–1708.
LaVeist, T. 1994 Beyond dummy variables and sample selection: What health services researchers ought to know about race as a variable. Health Services Research 29(1):1–16.
Lavizzo-Mourey, R., and E.Mackenzie 1996 Cultural competence: Essential measurements of quality for managed care organizations. Annals of Internal Medicine 124(10):919–921.
Lewontin, R. 1972 The apportionment of human diversity. Evolutionary Biology 6:381–398.
Liao, Y., R.Cooper, G.Cao, R.Durazo-Arvizu, J.Kaufman, A.Luke, and D.McGee 1998 Mortality patterns among adult Hispanics: Findings from the NHIS 1986–1990. American Journal of Public Health 88:227–232.
Libby, D., Z.Zhou, and D.Kindig 1997 Will minority physician supply meet U.S. needs? Health Affairs 16:205–214.
Lillie-Blanton, M., P.Parson, H.Gayle, and A.Dievler 1996 Racial differences in health: Not just Black and White, but shades of gray. Annual Review of Public Health 17:411–448.
Lind, M. 1998 The Beige and the Black. New York Times Magazine, August 16:38–39.
Loomis, D., and D.Richardson 1998 Race and risk of fatal injury at work. American Journal of Public Health 88:40–44.
Lynch, J., G.Kaplan, and S.Shema 1997 Cumulative impact of sustained economic hardship on physical, cognitive, psychological, and social functioning. New England Journal of Medicine 337(26):1889– 1895.
MacDorman, M, and J.Atkinson 1998 Infant mortality statistics from the linked birth/infant death data set—1995 period data. Monthly Vital Statistics Report 46(6/Suppl 2). Hyattsville, Md.: National Center for Health Statistics.
Malcolm S., G.VanHorne, Y.George, and C.Gaddy 1998 Losing Ground: Science and Engineering Graduate Education of Black and Hispanic Americans. Washington, D.C.: American Association for the Advancement of Science.
Markides, K., and J.Coreil 1986 The health of Hispanics in the southwestern United States: An epidemiologic paradox. Public Health Reports 3:253–265.
Marmot, M. 1998 Improvement of social environment to improve health. The Lancet 351:57–60.
Marmot, M., H.Bosma, H.Hemingway, E.Brunner, and S.Stansfield 1997 Contribution of job control and other risk factors to social variations in coronary heart disease incidence. The Lancet 350:235–239.
Marmot, M., M.Kogevinas, and M.Elston 1987 Social/economic status and disease. Annual Review of Public Health 8:111–135.
May, P. 1996 Overview of alcohol abuse epidemiology for American Indian populations. In Changing Numbers, Changing Needs: American Indian Demography and Public Health, G.Sandefur, R.Rindfuss, and B.Cohen, eds. Washington, D.C.: National Academy Press.
McBean, A., and M.Gornick 1994 Differences by race in the rates of procedures performed in hospitals for Medicare beneficiaries. Health Care Financing Review 15:77–90.
McCord, C., and H.Freeman 1990 Excess mortality in Harlem. New England Journal of Medicine 322:173–179.
McDonough, P., G.Duncan, D.Williams, and J.House 1997 Income dynamics and adult mortality in the United States, 1972 through 1989. American Journal of Public Health 87:1476–1483.
McGarvey, S. 1991 Obesity in Samoans and a perspective in its etiology in Polynesians. American Journal of Clinical Nutrition 53:1586–15945.
McKeown, T. 1979 The Role of Medicine: Dream, Mirage or Nemesis? Princeton: Princeton University Press.
McKinlay, J., and S.McKinlay 1977 The questionable contribution of medical measures to the decline of mortality in the United States in the twentieth century . Milbank Memorial Fund Quarterly (Summer):405–428.
Medicaid Access Study Group 1994 Access of Medicaid recipients to outpatient care. New England Journal of Medicine 330:1426–1430.
Menchik, P. 1993 Economic status as a determinant of mortality among Black and White and older men—Does poverty kill? Population Studies 47:427–436.
Morbidity and Mortality Weekly Report (MMWR) 1994 Prevalence of risk factors for chronic disease by education level in racial/ethnic populations—United States, 1991–1992. Morbidity and Mortality Weekly Report 43(48):894–899.
Moy E., and B.Bartman 1995 Physician race and care of minority and medically indigent patients. Journal of the American Medical Association 273:1515–1520.
Mustard, C., and N.Roos 1994 The relationship of prenatal care and pregnancy complications to birthweight in Winnipeg, Canada. American Journal of Public Health 84(9):1450–1457.
Myers, H., M.Kagawa-Singer, S.Kumanyika, B.Lex, and K.Markides 1995 Behavioral risk factors related to chronic diseases in ethnic minorities. Health Psychology 14:613–621.
Nagi, S. 1976 An epidemiology of disability among adults in the United States. Milbank Memorial Fund Quarterly (Fall):439–467.
National Center for Health Statistics (NCHS) 1998 Health, United States 1998 with Socioeconomic Status and Health Chartbook. Hyattsville, Md.: Public Health Service.
National Research Council (NRC) 1989 Diet and Health: Implications for Reducing Chronic Disease Risk. Committee on Diet and Health. Washington, D.C.: National Academy Press.
Neel, J. 1997 Are genetic factors involved in racial and ethnic differences in late-life health? In Racial and Ethnic Differences in the Health of Older Americans, L.Martin and B. Soldo, eds. Washington, D.C.: National Academy Press.
Nickens, H. 1991 Minorities in medicine. The Association of Black Cardiologists Newsletter 17:1,3.
Nickens, H., T.Ready, and R.Petersdorf 1994 Project 3000 by 2000: Racial and ethnic diversity in U.S. medical schools. New England Journal of Medicine 331:472–476.
Norris, J., L.Harmack, S.Carmichael, T.Pouane, P.Wakimoto, and G.Block 1997 U.S. trends in nutrient intake: The 1987 and 1992 National Health Interview Surveys. American Journal of Public Health 87:740–746.
Northridge, M., and P.Shepard 1997 Comment: Environmental racism and public health. American Journal of Public Health 87:730–732.
O’Brien, T., W.Flanders, P.Decoufle, C.Boyle, F.DeStefano, and S.Teutsch 1989 Are racial differences in the prevalence of diabetes in adults explained by differences in obesity? Journal of the American Medical Association 262:1485–1488.
Office of Management and Budget (OMB) 1997 Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity. Federal Register 62(210):58781–58790.
Olshanksy, S., and A.Ault 1986 The fourth stage of the epidemiologic transition: The age of delayed degenerative diseases. Milbank Memorial Fund Quarterly 64(3):355–391.
Otten, M., S.Teutsch, D.Williamson, and J.Marks 1990 The effect of known risk factors on the excess mortality of Black adults in the United States. Journal of the American Medical Association 263:845–850.
Palinkas, L., and S.Pickwell 1995 Acculturation as a risk factor for chronic disease among Cambodian refugees in the United States. Social Sciences and Medicine 40(12):1643–1653.
Pappas, G., W.Hadden, L.Kozak, and G.Fischer 1997 Potentially avoidable hospitalizations: Inequalities in rates between U.S. socioeconomic groups. American Journal of Public Health 87(5):811–816.
Pappas, G., S.Queen, W.Hadden, and G.Fisher 1993 The increasing disparity in mortality between socioeconomic groups in the United States, 1960 and 1986. New England Journal of Medicine 329:103–109.
Peck, M. 1994 The importance of childhood socio-economic group for adult health. Social Sciences and Medicine 39(4):553–562.
Perez-Stable, E., B.Herrera, P.Jacob, and N.Benowitz 1998 Nicotine metabolism and intake in Black and White smokers. Journal of the American Medical Association 280:152–156.
Piani, A., and C.Schoenborn 1993 Health promotion and disease prevention, United States, 1990. National Center for Health Statistics. Vital Health Statistics 10(185).
Popkin, B. 1997 Correction and revision of conclusions: Dietary trends in the United States. New England Journal of Medicine 337:1846–1848.
Popkin, B., A.Siega-Riz, P.Haines 1996 A comparison of dietary trends among racial and socioeconomic groups in the United States. New England Journal of Medicine 335:716–720.
Popkin, B., and J.Udry 1998 Adolescent obesity increases significantly in second and third generation U.S. immigrants: The National Longitudinal Study of Adolescent Health. Journal of Nutrition 128:701–706.
Preston, S., I.Elo, I.Rosenwaike, and M.Hill 1996 African-American mortality at older ages: Results of a matching study. Demography 33:193–209.
Preston, S., and P.Taubman 1994 Socioeconomic differences in adult mortality and health status. In Demography of Aging, L.Martin and S.Preston, eds. Washington, D.C.: National Academy Press.
Reid, J. 1986 Immigration and the future U.S. Black population. Population Today 14(2):6–8.
Ren, X., and B.Amick 1996 Race and self-assessed health status: The role of socioeconomic factors in the U.S.A. Journal of Epidemiology and Community Health 50:269–273.
Robinson, J. 1984 Racial inequality and the probability of occupation-related injury or illness. Milbank Memorial Fund Quarterly 62(4):567–590.
1987 Trends in racial inequality and exposure to work-related hazards, 1968–1989. Milbank Memorial Fund Quarterly 65(Suppl. 2):404–420.
1989 Exposure to occupational hazards among Hispanics, Blacks, and non-Hispanic Whites in California. American Journal of Public Health 79:629–630.
Rogers, R. 1992 Living and dying in the U.S.A.—Sociodemographic determinants of death among Blacks and Whites. Demography 29:287–303.
Rosenbaum, M., R.Leibel, and J.Hirsch 1997a Obesity. New England Journal of Medicine 337:396–407.
Rosenbaum, S., R.Serrano, M.Magar, and G.Stern 1997b Civil rights in a changing health care system. Health Affairs 16:90–105.
Schneider, D., M.Greenberg, and L.Lu 1997 Region of birth and mortality from circulatory diseases among Black Americans. American Journal of Public Health 87:800–804.
Schneider, R., F.Staggers, C.Alexander, M.Rainforth, K.Kondwani, S.Smith, and C.King 1995 A randomized controlled trial of stress reduction for hypertension in older African Americans. Hypertension 26(5):820–827.
Schoenbaum, M., and T.Waidmann 1997 Race, socioeconomic status, and health: Accounting for race differences in health. Journal of Gerontology 52B(Special Issue):61–73.
Schulman, K., J.Berlin, W.Harless, J.Kerner, S.Sistrunk, B.Gersh, R.Dibe, C.Taleghani, J. Burke, S.Williams, J.Eisenberg, and J.Escarce
1999 The effect of race and sex on physicians’ recommendations for cardiac catheterization. New England Journal of Medicine 340:618–626.
Shea, S., D.Misra, M.Ehrlich, L.Field, and C.Francis 1992 Predisposing factors for severe uncontrolled hypertension in an inner-city minority population. New England Journal of Medicine 327:776–781.
Short, P., A.Monheit, and K.Beauregard 1989 A profile of uninsured Americans. National Medical Expenditures Survey Research Finding 1. DHHS Pub. No. 89–3443. Rockville, Md.: National Center for Health Services Research and Health Care Technology Assessment.
Singer, B., and C.Ryff 1997 Racial and ethnic inequalities in health: Environmental, psychosocial, and physiological pathways. In Intelligence, Genes, and Success: Scientists Respond to the Bell Curve, B.Devlin, S.Fienberg, D.Resnick, and K.Roeder, eds. New York: Springer-Verlag (Copernicus).
Smith, D. 1998 Addressing racial inequities in health care: Civil rights monitoring and report cards. Journal of Health Care Politics, Policy and Law 23(1):75–105.
Smith, J., and R.Kington 1997a Demographic and economic correlates of health in old age. Demography 34:159– 170.
1997b Race, socioeconomic status, and health in late life. In Racial and Ethnic Differences in the Health of Older Americans, L.Martin and B.Soldo, eds. Washington, D.C.: National Academy Press.
Sorlie, P., E.Backlund, N.Johnson, and E.Rogot 1993 Mortality by Hispanic status in the United States. Journal of the American Medical Association 270:2464–2468.
Sorlie, P., E.Rogot, and N.Johnson 1992a Validity of demographic characteristics of death certificate. Epidemiology 3:181– 184.
Sorlie, P., E.Rogot, R.Anderson, N.Johnson, and E.Backlund 1992b Black-White mortality differences by family income. The Lancet 340:346–350.
Strogatz, D., J.Croft, S.James, N.Keenan, S.Browning, J.Garret, and A.Curtis 1997 Social support, stress, and blood pressure in Black adults. Epidemiology 8:482–487.
Tang, M., Y.Stern, K.Marder, K.Bell, B.Gurland, et al. 1998 The apoE-e4 allele and the risk of Alzheimer disease among African Americans, Whites, and Hispanics. Journal of the American Medical Association 279:751–755.
Thamer, M., C.Richard, A.Waldman, A.Casebeer, and N.Ray 1997 Health insurance coverage among foreign-born U.S. residents: The impact of race, ethnicity, and length of residence. American Journal of Public Health 87:96–102.
Thomson, G. 1997 Discrimination in health care. Annals of Internal Medicine 126:910–912.
Tibbits, C. 1937 The socio-economic background of Negro health status. Journal of Negro Education 6:413–428.
Todd, K., N.Samaroo, and J.Hoffman 1993 Ethnicity as a risk factor for inadequate emergency department analgesia. Journal of the American Medical Association 269:1537–1539.
Trask, J. 1917 The significance of the mortality rates of the colored population of the United States. American Journal of Public Health 6:254–260.
United Church of Christ Commission for Racial Justice 1987 Toxic Wastes and Race in the United States: A National Report on the Racial and Socioeconomic Characteristics of Communities with Hazardous Waste Sites. New York: United Church of Christ.
U.S. Bureau of the Census (USBC) 1991 The Hispanic population in the United States, March 1991. Pp. 20–455 in Current Population Reports. Washington, D.C.: U.S. Government Printing Office Bureau of Indian Affairs.
1993 We, the American Asians. Washington, D.C.: U.S. Government Printing Office.
1994 Census of Populations and Housing, 1990: Special Tabulations on Aging (STP 14). Washington, D.C.: U.S. Government Printing Office.
1996 65+ in the United States. Pp. 23–190 in Current Population Reports, Special Studies. Washington, D.C.: U.S. Government Printing Office.
1999 Estimates of the Population of States by Age, Sex, Race, and Hispanic Origin: 1990 to 1996. Washington, D.C.: U.S. Government Printing Office.
U.S. Department of Health and Human Services (USDHHS) 1985 Report of the Secretary’s Task Force on Black and Minority Health. Washington, D.C.: U.S. Department of Health and Human Services.
1988 Nature and Extent of Lead Poisoning in Children in the United States. Washington, D.C.: U.S. Department of Health and Human Services.
1990 Healthy People 2000: National Health Promotion and Disease Prevention Objectives. Washington, D.C.: U.S. Department of Health and Human Services.
1996 Physical Activity and Health: A Report of the Surgeon General. Atlanta, Ga.: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion.
1998 Tobacco Use Among U.S. Racial/Ethnic Minority Groups—African Americans, American Indians and Alaska Natives, Asian Americans and Pacific Islanders, and Hispanics: A Report of the Surgeon General. Atlanta, Ga.: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health.
Valdez, R., H.Morgenstern, E.Brown, R.Wyn, C.Wang, and W.Cumberland 1993 Insuring Hispanics against the costs of illness. Journal of the American Medical Association 269:889–894.
Vistnes, J., and A.Monheit 1997 Health insurance status of the civilian noninstitutionalized population: 1996. MEPS Research Findings No. 1. AHCPR Pub. No. 97–0030. Rockville, Md.: Agency for Health Care Policy and Research .
Wagener, D., and D.Winn 1991 Injuries in working populations: Black-White differences. American Journal of Public Health 81:1408–1414.
Walberg, P., M.McKee, V.Shkolnikov, L.Chenet, and D.Leon 1998 Economic change, crime, and mortality crisis in Russia: Regional analysis. British Medical Journal 317(7154):312–318.
Waldron, I. 1993 Recent trends in sex mortality ratios for adults in developed countries. Social Sciences and Medicine 36(4):451–462.
Wei, M, R.Valdez, B.Mitchell, S.Haffner, M.Stern, and H.Hazuda 1996 Migration status, socioeconomic status, and mortality rates in Mexican Americans and non-Hispanic Whites: The San Antonio Heart Study. Annals of Epidemiology 6:307–313.
Weiner, H. 1992 Perturbing the Organism: The Biology of Stressful Experience. Chicago: University of Chicago Press.
Wild, S., A.Laws, S.Fortmann, A.Varady, and C.Byrne 1995 Mortality from coronary heart disease and stroke for six ethnic groups in California, 1985–1990. Annals of Epidemiology 5:432–439.
Wilkinson, R. 1996 Unhealthy Societies: The Afflictions of Inequality. New York: Routledge.
Williams, D. 1990 Socioeconomic differentials in health: A review and redirection. Social Psychology Quarterly 53(2):81–99.
1994 The concept of race in health services research: 1966 to 1990. Health Services Research 29(3):261–274.
1996 Race/ethnicity and socioeconomic status: Measurement and methodological issues. International Journal of Health Services 26:483–505.
Williams, D., and C.Collins 1995 U.S. socioeconomic and racial differences in health: Patterns and explanations. Annual Review of Sociology 21:349–386.
Williams, D., R.Lavizzo-Mourey, and R.Warren 1993 The concept of race and health status in America. Paper from the CDC-ATSDR Workshop on the Use of Race and Ethnicity in Public Health Surveillance. Public Health Reports 109:26–41.
Wolff, E. 1998 Recent trends in the distribution of household wealth. Journal of Economic Perspectives 12(3):131–150.
Young, T. 1994 The Health of American Indians: Toward a Biocultural Epidemiology. New York: Oxford University Press.
Yu, E., and W.Liu 1992 U.S. national data on Asian Americans and Pacific Islanders: A research agenda for the 1990s. American Journal of Public Health 82:1645–1652.
Zambrana, R., S.Scrimshaw, N.Collins, and C.Dunkel-Schetter 1997 Prenatal health behaviors and psychosocial risk factors in pregnant women of Mexican origin: The role of acculturation. American Journal of Public Health 87:1022– 1026.
Zwerling, C., N.Sprince, R.Wallace, C.Davis, P.Whitten, and S.Heeringa 1996 Risk factors for occupational injuries among older workers: An analysis of the Health and Retirement Study. American Journal of Public Health 86:1306–1309.