NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Research Council (US) Panel on Race, Ethnicity, and Health in Later Life; Anderson NB, Bulatao RA, Cohen B, editors. Critical Perspectives on Racial and Ethnic Differences in Health in Late Life. Washington (DC): National Academies Press (US); 2004.

Cover of Critical Perspectives on Racial and Ethnic Differences in Health in Late Life

Critical Perspectives on Racial and Ethnic Differences in Health in Late Life.

Show details

12Racial/Ethnic Disparities in Health Behaviors: A Challenge to Current Assumptions

Marilyn A. Winkleby and Catherine Cubbin

One of the primary goals of Healthy People 2010 is “eliminating health disparities” among all population subgroups. This replaces the policy in earlier Healthy People Objectives of setting differential health goals by race/ ethnicity, age, gender, and indicators of socioeconomic status (SES). These new objectives acknowledge the need to eliminate, rather than merely reduce, social inequalities in order to achieve a parity of health.

This chapter has two goals that contribute to our understanding of health disparities. First, we examine racial/ethnic disparities in a comprehensive set of health behaviors related to chronic diseases to evaluate the extent to which disparities differ across health behaviors, age groups, and gender. Second, we assess the extent to which racial/ethnic disparities in health behaviors are related to underlying differences in indicators of SES. In addressing these goals, we challenge conventional assumptions about racial/ethnic disparities in health behaviors, especially the assumptions that populations of color have less healthy behaviors than white populations, and that racial/ethnic groups are internally homogeneous. We conclude that for some health behaviors, white populations have less healthy behaviors than do black and/or Hispanic populations, and for other health behaviors, the opposite is true. Furthermore, we conclude that disparities exist within each racial/ethnic group by important sociodemographic indicators, including age, gender, educational attainment, household income, and for Mexican Americans, country of birth and language spoken.

We focus on the following health behaviors and risk factors, all of which are related to chronic diseases: smoking, obesity, physical inactivity, poor diet, high alcohol consumption, and inadequate cancer screening practices. We selected these factors because of their effect on other chronic disease risk factors (hypertension, high cholesterol, diabetes) and important chronic disease outcomes (heart disease, stroke, cancer). While the underlying causes of these behaviors are not yet fully understood, they are all preventable, and change at any age can result in improved health.

In this chapter, we argue that fundamental explanations for racial/ ethnic disparities in health behaviors are largely socioeconomic in nature. Despite a consensus that race and ethnicity are sociopolitical constructs, as opposed to biological categories (Muntaner, Nieto, and O'Campo, 1996; Williams, 1996), some researchers and policy makers have interpreted racial/ethnic disparities in health behaviors, either implicitly or explicitly, as reflecting inherent genetically based differences (for a critique of this approach, see Krieger, 2001). Rather, racial/ethnic disparities may reflect the consequences of a historical pattern of discrimination, by individuals as well as institutions (Geronimus, 1992; Lynch, Kaplan, and Shema, 1997). The consequences of discrimination are expressed through a variety of mechanisms, including differences in population-level SES (Jones, 2000) (e.g., blacks and Hispanics in the United States are far more likely to be poor than whites) and residential environments (e.g., blacks and Hispanics in the United States are far more likely to live in poor communities than whites). Such differences in SES and residential environments have been shown repeatedly to be associated with unhealthy behaviors for whites as well as populations of color (Conference of Socioeconomic Status and Cardiovascular Health and Disease, 1995; Cubbin, Hadden, and Winkleby, 2001; Kaplan and Keil, 1993; Marmot and Elliot, 1992; Winkleby, Kraemer, Ahn, and Varady, 1998). Racial/ethnic disparities in health behaviors may also reflect differences in cultural norms and values. This interpretation may be particularly relevant for groups who have recently immigrated to the United States; for example, foreign-born Mexican Americans may have healthier diets and exercise patterns than those who are born in the United States.

We present data for the three largest racial/ethnic groups in the United States: white non-Hispanics, black non-Hispanics, and Hispanics (with a focus on Mexican Americans when possible). We do not present data on other racial/ethnic groups because data are limited from nationally representative samples across broad age groups. We base our main observations on data from two national data sets, the 1988-1994 Third National Health and Nutrition Examination Survey (NHANES III) and the 2000 Behavioral Risk Factor Surveillance System (BRFSS).

In the first section of this chapter, we (1) present population projections from Census data for selected racial/ethnic and age groups in the United States for the next 50 years; (2) review selected scientific literature on racial/ ethnic disparities in chronic disease outcomes and health behaviors; and (3) evaluate the importance of considering socioeconomic status, residential environments, and acculturation in studies on racial/ethnic disparities in health behaviors. In the second section of this chapter, we present new findings from analyses of racial/ethnic disparities in health behaviors and practices across a broad range of age groups using data from a national sample of white, black, and Hispanic women and men. In the third section we discuss the implications of our findings, and in the final section, we provide conclusions from our analyses.

POPULATION PROJECTIONS

Dramatic changes in the racial/ethnic and age distributions of the U.S. population over the next 50 years will have a significant impact on chronic diseases, most which manifest in later life. Figure 12-1 presents population projections from the U.S. Census for the years 2000, 2010, and 2050 for white, black, and Hispanic adult women and men by age group (U.S. Census Bureau, 2001). Population sizes are given in thousands. There will be large increases in the Hispanic total adult population and to a lesser degree in the black population. In 2000, Hispanics and blacks accounted for 11 percent and 12 percent, respectively, of the three racial/ethnic groups aged 18 and over shown in Figure 12-1. By 2050, Hispanics and blacks will account for 25 percent and 15 percent, respectively. There will be even larger changes in the elderly population. By 2050, the elderly white population (aged 75 and older) will increase twofold, the elderly black population will increase fivefold, and the elderly Hispanic population will increase ninefold. For example, while the elderly Hispanic population in 2000 numbered under 1 million (792,000), it is projected to increase to more than 7 million people (7,055,000) by 2050. Given these projections, ethnic minority populations will bear an increased share of chronic diseases.

FIGURE 12-1. Population projections in thousands for the years 2000, 2010, and 2050 for white, black, and Hispanic adults, 18 and older, by age group.

FIGURE 12-1

Population projections in thousands for the years 2000, 2010, and 2050 for white, black, and Hispanic adults, 18 and older, by age group. Population in thousands. SOURCE: U.S. Census Bureau (2001).

RACIAL/ETHNIC DISPARITIES IN CHRONIC DISEASE INCIDENCE AND MORTALITY

The leading causes of death for all racial/ethnic groups in the United States are from heart disease (30 percent of all deaths), cancer (23 percent), and stroke (7 percent) (Jemal, Thomas, Murray, and Thun, 2002). These chronic diseases account for nearly three-fourths of all deaths among women and men during some of the most productive years of their lives (25 to 64 years of age). Furthermore, they account for more than $300 billion in direct medical costs each year (Institute of Medicine, 1991).

Although the majority of deaths for all racial/ethnic groups occur from chronic diseases, death rates vary considerably across racial/ethnic groups. Black women and men have higher age-standardized death rates from cardiovascular disease (CVD) than white women and men, regardless of income (Singh, Kochanek, and McDonan, 1996). This includes their strikingly higher death rates from stroke. Higher death rates for blacks from CVD begin in early ages and continue until age 65 (Pamuk, Makue, Heck, Reuben, and Lochner, 1998). After age 65, black-white differences in CVD death rates are smaller, with rates converging or crossing over at the oldest ages (Hollman, 1993). Although there have been large declines in U.S. death rates from CVD since the mid-1960s, the declines since the mid-1980s have been slower for blacks than for whites, producing larger black-white disparities (Singh et al., 1996; Tyroler, Wing, and Knowles, 1993).

Blacks also have higher incidence rates for cancer and higher death rates following diagnosis than whites or Hispanics (Jemal et al., 2002). Death rates for all cancer sites are approximately 33 percent higher for blacks than for whites, and more than twice as high as for Hispanics. In the past decade, black men have shown the largest declines in cancer incidence and mortality of all racial/ethnic and gender-specific groups (Jemal et al., 2002), a change most likely related to a combination of risk factor reduction and better treatment and access to care.

In contrast to blacks, past studies have shown that Hispanics have lower age-standardized death rates from all causes, CVD, and cancer than whites (Aguirre-Molina, Molina, and Zambrana, 2001; Elo and Preston, 1997; Goff et al., 1997; National Cancer Institute, 2001; Sorlie, Backlund, Johnson, and Rogot, 1993; Vega and Amaro, 1994). These rates are consistent for women and men, and for all Hispanic groups, including Mexican-Americans, Cuban Americans, and mainland Puerto Ricans. There are several exceptions to the cancer rates——Hispanics have higher incidence and/or mortality rates of stomach, liver, cervical, and gallbladder cancer than whites (Gutierrez-Ramirez, Valdez, and Carter-Pokras, 1994; Sorlie et al., 1993). The overall mortality advantage for Hispanics has been termed the “Hispanic paradox” because Hispanics have higher rates of diabetes and obesity and lower socioeconomic status than whites. However, the Hispanic paradox has recently been called into question. Investigators from the Corpus Christi Heart Project reported a greater incidence of hospitalized myocardial infarction among both Mexican-American women and men than among non-Hispanic whites, concluding that the finding was congruent with the risk factor patterns observed in the Mexican-American population (Goff et al., 1997). More recently, investigators from the San Antonio Heart Study reported greater all-cause, CVD, and coronary heart disease mortality among Mexican Americans than non-Hispanic whites (Hunt et al., 2003). These new findings point out the need for studies of larger populations as well as studies that examine possible bias in mortality rates. For example, studies are needed to determine the extent to which Hispanic mortality rates are underestimated because of factors including selective immigration, return of terminally ill persons to their country of birth, age misreporting at older ages, and record linkage issues (Elo and Preston, 1997; Stephen, Foote, Hendershot, and Schoenborn, 1994).

Are Disparities in Mortality Explained by Differences in Health Behaviors?

Given the racial/ethnic disparities in chronic disease mortality, investigators have explored whether differences in health behaviors and risk factors (e.g., smoking, dietary habits, exercise, alcohol consumption) explain these disparities (Lynch, Kaplan, Cohen, Tuomilehto, and Salonen, 1996; Smith, Neaton, Wentworth, Stamler, and Stamler, 1996). Otten and colleagues, Teutsch, Williamson, and Marks (1990), used data for black and white adults from the NHANES Epidemiologic Follow-Up Survey to evaluate whether black-white differences in all-cause mortality (of which chronic diseases are major contributors) were explained by differences in health behaviors and risk factors (cigarette smoking, systolic blood pressure, cholesterol level, body mass index, alcohol intake, and diabetes mellitus). The black-white mortality ratio for women and men aged 35 to 54 was 2.3 without adjustment for health behaviors and risk factors. This decreased to 1.9 when the six factors were taken into account, and to 1.4 when family income was also considered. The authors concluded that 31 percent of the black-white differences in all-cause mortality could be accounted for by the six health behaviors and risk factors and an additional 38 percent by family income (Otten et al., 1990). The remaining differences in mortality, unexplained by known health behaviors, risk factors, and/or SES, suggest that other mechanisms contribute to racial/ethnic disparities in mortality. However, in this study, as well as in others, adjustment for SES was most likely incomplete.

Racial/Ethnic Disparities in Health Behaviors Among Adults

Many studies have examined racial/ethnic disparities in health behaviors among adults, although most have not used representative samples or had adequate sample sizes to assess differences by gender, age, and/or SES. Studies on smoking show that white women, particularly those with lower SES, are more likely to smoke and to smoke more heavily than black or Hispanic women, especially Mexican-American women (U.S. Department of Health and Human Services, 2001). Although some studies show that white men are more likely to smoke than black and Hispanic men, this finding varies according to the age of the study population, the composition of the Hispanic sample, and the consideration of SES in the analysis (Haynes, Harvey, Montes, Nickens, and Cohen, 1990). Studies that have examined smoking among men from the three main Hispanic populations in the United States (Mexican Americans, mainland Puerto Ricans, and Cuban Americans) have found that Cuban-American men smoke the most and Mexican-American men smoke the least (Rogers, 1991).

In contrast to studies on smoking, other studies consistently show that black women have higher prevalences of excess weight and physical inactivity and poorer diets than white women; differences between black and white men are less consistent and generally of lower magnitude (Burke et al., 1992; DiPietro, Williamson, Caspersen, and Eaker, 1993; Duelberg, 1992; Folsom et al., 1991; Gidding et al., 1996; Kumanyika, Wilson, Guilford-Davenport, 1993). Other studies, with some inconsistencies, show that Mexican-American women have higher prevalences of excess weight and physical inactivity than white women (Balcazar and Cobas, 1993; Diehl and Stern, 1989; Haffner et al., 1986; Kuczmarski, Flegal, Campbell, and Johnson, 1994; Mitchell, Stern, Haffner, Hazuda, and Patterson, 1990; Winkleby, Fortmann, and Rockhill, 1993; Winkleby et al., 1998). Again, findings are less consistent for men (Winkleby et al., 1993; Winkleby, Cubbin, Ann, and Kraemer, 1999a; Elder et al., 1991). In general, these studies show much higher prevalences among those with lower SES.

Most studies of alcohol consumption show few or no differences between black and white populations. Some groups of Hispanic men show heavier alcohol consumption than white men. Mexican-American and Puerto Rican men show heavy use in younger ages, with decreasing use in older ages (Markides, Ray, Stroup-Benham, and Trevino, 1990; Rogers, 1991). As with smoking, alcohol consumption is low for Hispanic women, especially those of Mexican origin; however, rates appear to increase with level of acculturation (Black and Markides, 1993). Heavy alcohol consumption is much more likely in men than in women for all racial/ethnic groups, and has shown inverse associations with SES.

Recent studies of racial/ethnic differences in health behaviors have had the opportunity to include data on representative samples of women and men from the largest racial/ethnic groups in the United States. Given the large sample sizes, these studies have been able to evaluate the influence of SES as well as age. One of the largest national surveys to include multiple racial/ethnic groups, with clinical examination, is NHANES III (National Center for Health Statistics, 1994b). This national survey was conducted from 1988 to 1994 at 89 sites to assess the health and nutrition status of the U.S. population aged 2 months and older. It is noteworthy because it included an oversampling of black and Mexican-American women and men who represent a wide range of SES levels.

In a previous analysis, Winkleby and Cubbin used NHANES III data to examine racial/ethnic differences in health behaviors among 3,229 black, 3,025 Mexican-American, and 3,775 white (non-Hispanic) women and men, ages 25 to 64 (Winkleby et al., 1999a). This analysis evaluated differences in three factors related to chronic disease health behaviors: smoking, obesity, and leisure-time physical inactivity. The results showed that race/ethnicity was independently associated with health behaviors after adjustment for educational attainment and family income divided by family size. Black and Mexican-American women had significantly higher odds of obesity and physical inactivity than white women (odds ratios 1.5 to 2.3, p values <0.01). Black men had higher odds of smoking and physical inactivity than white men (odds ratios 1.3 and 1.4 respectively, p values <.05). In contrast, both Mexican-American women and men had lower odds of smoking than white women and men (odds ratios 0.19 and 0.37 respectively, p values <0.001). The magnitude of the racial/ethnic differences was large for many comparisons (Winkleby et al., 1998). For example, black women were, on average, 16.8 pounds heavier than white women of comparable education and age. While these analyses of NHANES III data adjusted for age, differences across age groups were not examined.

Racial/Ethnic Disparities in Health Behavior Among Children and Young Adults

A number of studies have examined racial/ethnic differences in health behaviors among children and young adults (Anderson, Crespo, Bartlett, Cheskin, and Pratt, 1998; Belcher et al., 1993; Berenson et al., 1998; The Bogalusa Heart Study, 1995; Dwyer et al., 1998; Folsom et al., 1989; McNutt et al., 1997, Srinivasan, Bao, Wattigney, and Berenson, 1996; Tortolero et al., 1997). Using NHANES III data, Winkleby et al., examined racial/ethnic differences in health behaviors in a sample of 2,769 black, 2,854 Mexican-American, and 2,063 white children and young adults, aged 6 to 24 years (Winkleby, Robinson, Sundquist, and Kraemer, 1999b). The analysis evaluated the age groups at which racial/ethnic differences were first apparent and whether differences remained after accounting for educational attainment of the head of the household and family income divided by family size. Whites, especially those from less educated households, had the highest prevalences of smoking; 77 percent of young white men and 61 percent of young white women, aged 18 to 24 years, from lower educated households were current smokers. In contrast, black and Mexican-American girls had significantly higher levels of body mass index (BMI) and percentages of energy from dietary fat than white girls. The racial/ethnic differences for BMI were evident by 6 to 9 years of age (a difference of approximately 0.5 BMI units) and widened thereafter (a difference of more than 2 BMI units among 18- to 24-year-olds). Black boys had higher levels of dietary fat energy intake than white boys. All racial/ ethnic differences remained significant after adjusting for age and education of the head of the household. Adjusting for family income showed similar results.

Racial/Ethnic Disparities in Health Behaviors Among the Elderly

Few studies have examined racial/ethnic differences in health behaviors among elderly populations, especially using nationally representative samples. The initial studies in this area show that: (1) older white women and men are more likely to have ever smoked, but are also more likely to have quit smoking than older black women and men; and (2) older black women and men are more obese and physically inactive, but are less likely to have high alcohol consumption than older white women and men (National Research Council, 1997).

Sundquist and colleagues used data from NHANES III to examine whether racial/ethnic differences in health behaviors shown for younger women and men in NHANES III persisted for elderly women and men (Sundquist, Winkleby, and Pudaric, 2001). His analysis included 700 black, 628 Mexican-American, and 2,192 white women and men aged 65 to 84 years. The health behaviors examined were cigarette smoking, abdominal obesity, and leisure-time physical inactivity. No significant racial/ethnic differences were found for smoking. However, black women were significantly more likely to be obese and physically inactive than white women after accounting for age and years of education (odds ratios 1.8 and 2.6 respectively). Black men were significantly more likely to be physically inactive than white men (odds ratio 1.9). No significant differences were found between Mexican Americans and whites for the three health behaviors. The racial/ethnic differences documented by this study may be underestimated because of survival bias. The study population represents an age cohort born between 1904 and 1929 who survived to 1988 to 1994, the dates of the NHANES III assessments. Therefore, people in this age cohort who survived to 1988 to 1994 may represent those with healthier behaviors. Underestimation may be especially true for the black-white differences because of the substantially higher rates of early death from chronic diseases and injuries for black compared with white women and men (Corti et al., 1999; Ventura, Peters, Martin, and Maurer, 1997).

Health Behaviors Across Age Groups

Despite the strong associations between age and chronic disease outcomes, few studies have examined racial/ethnic disparities in health behaviors across a broad range of age groups. Some studies indicate that disparities in health behaviors are larger for younger and middle-aged adults than for older adults; however, most studies have lacked sufficient sample sizes within racial/ethnic subgroups to provide definitive answers. Because chronic diseases reflect a progressive process that begins early in the life course and initiation of unhealthy behaviors often occurs in early adolescence, it is important to include young populations in analyses of racial/ ethnic disparities in health behaviors. It is also important to include populations across a broad age range to examine when disparities are first apparent and whether disparities differ across age groups. This can provide insight about causal pathways and the timing, focus, and content of primary, secondary, and tertiary prevention programs and policies (Lowry, Kahn, Collins, and Kolbe, 1996; Smith, Hart, Watt, Hole, and Hawthorne, 1998; Winkleby et al., 1999a).

The Role of Socioeconomic Status

Racial/ethnic disparities in health behaviors most likely result from complex relationships, with SES, residential environments, and cultural characteristics each playing a role. SES, however measured, has shown strong, consistent associations with chronic disease outcomes and health behaviors for decades (Adler, Boyce, Chesney, Folkman, and Syme, 1993; Conference of Socioeconomic Status and Cardiovascular Health and Disease, 1995; Kaplan and Keil, 1993; Marmot and Elliot, 1992). Studies on racial/ethnic disparities in health that assess SES should ideally measure multiple dimensions of SES (e.g., measuring power, status, and economic class) at multiple levels (individual, household, neighborhood, community). However, because of its complexity, measuring SES fully is exceedingly difficult. In addition, there are issues of bias because SES measures such as education and income may not be commensurate across racial/ ethnic groups. Thus it is likely that residual confounding by SES exists in any study that investigates racial/ethnic disparities even after “adjusting” for multiple measures of SES (Braveman, Cubbin, Marchi, Egerter, and Chavez, 2001; Kaufman, Cooper, and McGee, 1997; Winkleby and Cubbin, 2003).

Individual-level education, income, and occupational status have most commonly been used as indicators of SES in studies of health behaviors. Each measure has limitations (Smith and Kington, 1997); some are related to general measurement bias and others are particularly relevant to investigations of how racial/ethnic disparities vary across age groups.

The measurement of education in the United States is compromised because measures of educational attainment (e.g., years of education or credentials) do not account for large inequalities in quality of schooling, especially for those from certain racial/ethnic groups and those from the current generation of elderly people. Furthermore, the same level of educational attainment does not convey the same meaning when examining differences in health behaviors across age groups; for example, a high school degree for an elderly population may confer the same status and prestige as a college degree for a younger population. Finally, the measurement of education is difficult to interpret when sample populations include people who have been educated in countries outside the United States.

Measurement of income and occupation/employment status present additional challenges in studies of racial/ethnic differences in health behaviors. Current income may not reflect earnings over one's lifetime and does not reflect wealth (e.g., investment income), especially among the retired. This is particularly problematic in that racial/ethnic differences in wealth are far greater than differences in income (Eller, 1994). Occupational status is also complicated in analyses across age groups because the same occupational category may reflect different exposures, experiences, and/or status across racial/ethnic groups. In addition, standard occupational categories in the United States combine a broad range of occupations and are based on types of work rather than social class theory, limiting their use as a socioeconomic measure (Krieger, Williams, and Moss, 1997).

THE ROLE OF RESIDENTIAL ENVIRONMENTS

In recent years there has been an increasing interest in contextual studies that combine characteristics of both individuals and residential environments to investigate their joint association on individual-level health outcomes (Diez-Roux, 1998; Pickett and Pearl, 2001). A number of investigators have proposed that racial/ethnic differences in health behaviors may be explained in part by factors beyond the individual, such as neighborhood-level influences.

There is a growing consensus that the residential environment that encompasses the immediate physical surroundings, social relationships, and cultural milieus within which people function and interact may influence both the SES and health of their residents. This includes the built infrastructure; industrial and occupational structure; labor markets; social and economic processes; wealth; social, human, and health services; government; race relations; cultural practices; religious institutions and practices; and beliefs about place and community (Macintyre, Ellaway, and Cummins, 2002; Winkleby and Cubbin, 2003). Differences in these residential environments can translate into differences in access to tobacco, alcohol, healthy food choices, safe places to exercise, and preventive health care, all of which can promote or impede healthy behaviors.

Racial/ethnic groups in the United States are highly segregated, resulting in populations of color being far more likely to live in disadvantaged places. Thus, taking into account the characteristics of residential environments may partly explain racial/ethnic disparities in health, after accounting for differences in individual-level demographic and socioeconomic characteristics.

A growing body of research supports the independent association of neighborhood socioeconomic characteristics on chronic disease morbidity and mortality, risk factors, and health behaviors. Residence in a socioeconomically disadvantaged area has been found to be independently associated with heart disease morbidity (Diez-Roux et al., 1997, 2001; Jones, 2000; Smith et al., 1998) and mortality (LeClere, Rogers, and Peters, 1998; Smith et al., 1998; Winkleby and Cubbin, 2003), and CVD risk factors and health behaviors (Cubbin et al., 2001; Diez-Roux et al., 1997, 1999; Duncan, Jones, and Moon, 1996; Ellaway, Anderson, and Macintyre, 1997; Hart, Ecob, and Smith, 1997; Lee and Cubbin, 2002; Smith et al., 1998; Sundquist, Malmstrom, and Johansson, 1999; Yen and Kaplan, 1998). For example, living in a low-SES neighborhood has been independently associated with lower physical activity (Yen and Kaplan, 1998), higher body mass index (Cubbin et al., 2001; Ellaway et al., 1997; Smith et al., 1998), higher prevalence of smoking (Cubbin et al., 2001; Diez-Roux et al., 1997; Smith et al., 1998; Sundquist et al., 1999), and less healthy dietary habits in adults (Diez-Roux et al., 1999) as well as youth (Lee and Cubbin, 2002).

The Role of Acculturation

The role of cultural characteristics in racial/ethnic disparities in health behaviors has been examined using various indicators of acculturation (Aguirre-Molina et al., 2001; Hazuda, Stern, and Haffner, 1988). Acculturation is defined as the degree to which a person participates in the language, values, and practices of his or her ethnic community compared with those in the dominant culture (Padilla, 1980). Some studies that have examined the influence of acculturation on health behaviors suggest that health behaviors may worsen as populations become more acculturated (Espino, Burge, and Moreno, 1991). Other studies suggest that a healthy immigrant effect may bias studies on acculturation (e.g., those who immigrate to the United States have healthier behaviors than those who do not immigrate) (Jasso, Massey, Rosenzweig, and Smith, 2000). Although we could not assess a healthy immigrant effect, we were able to examine the association between level of acculturation and unhealthy behaviors across a wide range of age groups using data from the NHANES III sample of Mexican-American women and men aged 18 to 74. We stratified the Mexican-American sample by two frequently used indicators of acculturation: country of birth (born in the United States or in Mexico) and primary language spoken at home (English or Spanish). We examined differences in four health behaviors and risk factors related to chronic disease:

  • Cigarette smoking (smoked at least 100 cigarettes in entire life and currently smoking cigarettes everyday).
  • Obesity (body mass index ≥30 units).
  • Low vegetable and/or fruit consumption (<3 servings of vegetables and/or fruits per day in the past month).
  • High alcohol consumption (≥5 alcoholic drinks on ≥1 occasion in the past month and/or an average of ≥2 drinks per day for women and ≥3 drinks per day for men among those who drank alcoholic beverages in the past month). The analysis used predicted values from linear models, adjusted for education (in years) and family income.

In general, we found that women and men born in the United States had notably higher predicted prevalences of unhealthy behaviors than those born in Mexico, after adjustment for education and income (Figure 12-2). The relationships were remarkably consistent across health behaviors and gender, and across most age groups. The largest differences were seen for younger women and older men for smoking, younger women and younger men for vegetable and/or fruit consumption, and younger women for high alcohol consumption; adjusted prevalences were two to three times higher for those born in the United States compared with those born in Mexico. There was little change in these findings when the analysis was not adjusted for education and income. We found similar associations for language spoken; after adjustment for education and income, those who spoke English had higher predicted prevalences of unhealthy behaviors than those who spoke Spanish.

FIGURE 12-2. Unhealthy behaviors among Mexican-American women and men, ages 18-74, by country of birth and age group, adjusted for education and family income.

FIGURE 12-2

Unhealthy behaviors among Mexican-American women and men, ages 18-74, by country of birth and age group, adjusted for education and family income. Percentages were calculated with normalized sample weights. SOURCE: National Health and Nutrition Examination (more...)

Analyses of the BRFSS

In the following analyses, we build on the results of previous studies to further our knowledge of racial/ethnic differences in health behaviors across a wide range of age groups. We use data from the Behavioral Risk Factor Surveillance System, a national cross-sectional survey that has standardized definitions and assessments of health behaviors in women and men aged 18 and over from multiple racial/ethnic groups. This analysis provides a picture of racial/ethnic disparities in a comprehensive set of health behaviors across age groups, with consideration of the role of SES. Data are presented for white, black, and Hispanic women and men aged 18 to 74.

The main questions addressed in these analyses are: (1) To what extent do racial/ethnic disparities differ across health behaviors and age groups? (2) To what extent are racial/ethnic disparities in health behaviors accounted for by differences in individual-level indicators of SES?

Background of the BRFSS

The BRFSS is a continuous, state-based telephone survey conducted by state health departments in collaboration with the Centers for Disease Control and Prevention to assess the health of the civilian adult population ages 18 and older (http://www.cdc.gov/nccdphp/brfss). The survey methodology is based on a stratified sampling design whereby eligible individuals are randomly selected within households that have working telephone numbers. Trained interviewers use Computer Assisted Telephone Interview (CATI) software to complete more than 150,000 surveys each year. The BRFSS includes questions on health behaviors, demographic characteristics, socioeconomic status, health status, and health insurance; these are organized as core questions, rotating modules, and state-added questions. The BRFSS thus provides some of the most recent national estimates of health behaviors and screening known to affect chronic disease. The sample for this analysis included 114,325 white (non-Hispanic), 11,938 black (non-Hispanic), and 13,088 Hispanic women and men aged 18 to 74 years who were interviewed for the 2000 BRFSS and were not missing information on educational attainment or household income. Data were not collected for separate groups of Hispanics in the 2000 BRFSS.

Statistical analysis. We carried out our primary statistical analyses, based on linear models, using SUDAAN, a program that adjusts for the complex BRFSS sample design for calculating variance estimates (Shah, Barnwell, Hunt, Nileen, and LaVange, 1991). All analyses incorporated sampling weights that adjusted for unequal probabilities of selection. We used SAS to calculate weighted prevalences. We used multiple linear regression models to calculate odds ratios, unadjusted and adjusted for educational attainment and household income. The dependent variables in our models were seven health behaviors and risk factors and two cancer screening tests (to be defined). The independent variables in the adjusted models were race/ethnicity (black and Hispanic women and men compared separately to white women and men), educational attainment, and household income (to be defined). Because of multiple outcome variables, we selected a conservative level of statistical significance of P < 0.01, two-tailed.

Defining Race/Ethnicity and SES

In the BRFSS analyses, we employ the term race/ethnicity to refer to the three social groups in our sample: white, black, and Hispanic adults. We recognize that SES is a multidimensional construct, and choose educational attainment and household income as indicators of SES. We consider both measures as indicators of prestige, economic resources, and/or power, the dimensions of social class under Weber's framework (Runciman, 1978). We defined race/ethnicity and SES as follows:

Race/ethnicity: Self-reported race/ethnicity as white (not of Spanish or Hispanic origin), black (not of Spanish or Hispanic origin), and Hispanic (of Spanish or Hispanic origin).

Educational attainment: Highest grade or year of school completed (none or elementary school, some middle or high school, high school graduate, some college or technical school, college graduate).

Household income: Annual household income from all sources (<$15,000, $15,000 to <$25,000, $25,000 to <$50,000, $50,000 to <$75,000, ≥$75,000). Respondents who did not report income (12 percent of the sample) were excluded from the analysis. These respondents were more likely to be women, older, and Hispanic, and to have completed less education compared to those who reported income.

The seven health behaviors and risk factors we analyzed were defined as follows:

Current cigarette smoker: Smoked at least 100 cigarettes in entire life and currently smoking cigarettes every day or some days.

Secondhand smoke exposure: Lived in a home where anyone, including the respondent, had smoked cigarettes, cigars, or pipes in the past 30 days.

Failure to quit smoking: Among those who ever smoked cigarettes, those currently smoking cigarettes.

Obesity: BMI ≥30 units, calculated as weight in kg/height in m2.

No leisure-time physical activity: Questions adapted from the 1985 National Health Interview Survey (National Center for Health Statistics, 1994a). No leisure-time physical activity in the past month, including exercises, sports, or physically active hobbies.

Low vegetable and/or fruit intake: <3 servings of vegetables and/or fruits (green salad, carrots, nonfried potatoes, other vegetables, fruit, fruit juices) per day.

High alcohol consumption: ≥5 alcoholic drinks on ≥1 occasion in the past month and/or an average of ≥2 drinks per day for women and ≥3 drinks per day for men among those who drank alcoholic beverages in the past month.

We chose two screening tests that reflect health behaviors related to breast and cervical cancer, both of which have shown racial/ethnic differences in survival (Ragland, Selvin, and Merrill, 1991).

Pap test screening: No Pap test within the past 2 years.

Mammography screening: No mammogram within the past 2 years (among women aged 45 to 74).

Although the 2000 BRFSS included colorectal cancer screening questions, these questions were asked only in four states and therefore are not included.

Results of BRFSS Analysis

Table 12-1 presents the prevalences of unhealthy behaviors for white, black, and Hispanic women and men by age group. Because the main focus of this analysis is on the relationship of race/ethnicity and SES, these prevalences are presented for background information only.

TABLE 12-1. Prevalences of Unhealthy Behaviors for White, Black, and Hispanic Women and Men by Age Group: BRFSS, 2000.

TABLE 12-1

Prevalences of Unhealthy Behaviors for White, Black, and Hispanic Women and Men by Age Group: BRFSS, 2000.

Table 12-2 presents the associations between race/ethnicity and health behaviors by gender and age group. We initially examined differences by 10-year age groups, but then combined the data into 20-year age groups for 25- to 44-year-olds and 45- to 64-year-olds because no substantial differences existed between these 10-year age groups. Odds ratios are presented for each health behavior for black and Hispanic women and men, with white women and men as the reference group. Odds ratios greater than 1.00 indicate higher odds of unhealthy behaviors for black and/or Hispanic women and men compared with white women and men. To assess the extent to which racial/ethnic disparities in health behaviors were accounted for by differences in SES, the odds ratios are first presented unadjusted, and then adjusted for two indicators of SES—educational attainment and household income.

TABLE 12-2. Odds Ratios for Unhealthy Behaviors for White, Black, and Hispanic Women and Men by Age Group, Unadjusted and Adjusted for Education and Family Income: BRFSS, 2000.

TABLE 12-2

Odds Ratios for Unhealthy Behaviors for White, Black, and Hispanic Women and Men by Age Group, Unadjusted and Adjusted for Education and Family Income: BRFSS, 2000.

Black and Hispanic women and men had substantially lower odds of smoking than white women and men in nearly every age group (Table 12-2). The differences in smoking were large and fairly consistent across age groups, and most remained significant after adjustment for education and income. In general, adjustment for education and income increased the racial/ethnic differences in smoking; in some cases, comparisons that were not significant in the unadjusted analysis became significant after adjustment (e.g., the black/white comparisons for women aged 65 to 74). After adjustment, black and Hispanic women had approximately a quarter to half the odds of smoking than white women. The black/white odds of smoking for men showed similar patterns, although the racial/ ethnic disparities were not generally as large as those for women.

Black and Hispanic women also had lower odds of exposure to secondhand smoke than white women, especially in the younger age groups. Hispanic men compared with white men showed similar differences. After adjustment, no differences were found between black and white men.

Few significant racial/ethnic differences were found for failure to quit smoking after adjustment among those who had ever smoked. There were two exceptions. Hispanic women and men aged 25 to 44 were more likely to quit smoking than white women and men of the same ages. In contrast, black men aged 65 to 74 were much less likely to quit smoking than white men aged 65 to 74. This is reflected in the cigarette smoking prevalence rates of black men compared with white men that are lower or similar across younger ages, but then cross over and become higher from ages 45 on (see Table 12-1).

In contrast to smoking, black women had substantially higher odds of obesity and leisure-time physical inactivity than white women after adjusting for education and income. The odds of obesity were approximately twice as high for black women compared with white women at every age group, after adjustment. Hispanic women aged 25 to 44 had higher odds of physical inactivity than white and black women. The patterns for men after SES adjustment were different. Black men had similar odds of obesity and physical inactivity compared with white men (with the exception of higher odds of obesity for black men aged 25 to 44) and Hispanic men (aged 25-64) had higher odds of physical inactivity than white men. Few significant racial/ethnic disparities in low vegetable and/or fruit intake were found for either women or men.

While few racial/ethnic differences were observed for high alcohol consumption, there were several important exceptions. Black and Hispanic women aged 18 to 24, and black women and men aged 25 to 44, had significantly lower odds of alcohol consumption than white women or men in the same age groups after adjustment for education and income.

Black women aged 25-64 were more likely to report having a recent Pap test than white women after adjustment for education and income. The adjusted results for Hispanic women compared with white women were mixed; Hispanic women were less likely to be screened at ages 18 to 24 and more likely to be screened at ages 45 to 74. Black and Hispanic women aged 45 to 64 and Hispanic women aged 65 to 74 were significantly more likely to report a recent mammogram than white women in the same age groups. However, none of the odds for mammography screening were significant when unadjusted for education and income.

Although not the focus of this chapter, we also examined the separate influence of SES on health behaviors. As shown in many past analyses, we found strong, inverse associations between educational attainment and income at almost every age group for each health behavior, except for high alcohol consumption, where education and income were generally not statistically significant. For all other health behaviors, women and men with lower educations and/or lower incomes had higher odds of unhealthy behaviors compared to those with higher educations and/or incomes, after adjusting for race/ethnicity (data not shown).

In addition to adjusted analyses, stratified analyses offer insight about racial/ethnic disparities in health behaviors. From our previous work (Winkleby et al., 1998, 1999b; Winkleby, Schooler, Kraemer, Lin, and Fortmann, 1995), we were aware of the large differences in unhealthy behaviors by educational attainment and household income, especially within black and white racial/ethnic groups. To examine these relationships across age groups, we stratified the BRFSS data for white, black, and Hispanic women and men by age group and educational attainment (<12 and ≥12 years of education) and annual household income (<$25,000 and ≥$25,000). We are aware that the categorization of education and income into two levels introduces assumptions regarding residual confounding (Kaufman et al., 1997); however, more than two levels of categories for three racial/ethnic groups are cumbersome to present.

We found large differences in unhealthy behaviors by both level of education and income for black and white adults that persisted across age groups. The differences were apparent for all of the seven health behaviors and risk factors and two cancer screening tests, with the exception of alcohol consumption. The differences by education for Hispanic women and men were generally much weaker or not apparent, except for obesity for both women and men, and cancer screening for women. Selected findings for cigarette smoking by level of education and obesity by household income are shown in Figures 12-3 and 12-4. The solid lines represent those having lower educational and lower income levels. For smoking, large disparities are seen by education for both black and white women and men, and to a lesser degree for Hispanic women. For obesity, the largest disparities by income are seen for women.

FIGURE 12-3. Prevalence of cigarette smoking at each age group by level of education for white, black, and Hispanic women and men, ages 18-74.

FIGURE 12-3

Prevalence of cigarette smoking at each age group by level of education for white, black, and Hispanic women and men, ages 18-74. SOURCE: Behavioral Risk Factor Surveillance System (2000).

FIGURE 12-4. Prevalence of obesity at each age group by household income for white, black, and Hispanic women and men, ages 18-74.

FIGURE 12-4

Prevalence of obesity at each age group by household income for white, black, and Hispanic women and men, ages 18-74. SOURCE: Behavioral Risk Factor Surveillance System (2000).

IMPLICATIONS OF OUR FINDINGS

Like the answers to many scientific questions, our findings yield a complex picture. We demonstrate that the pattern of racial/ethnic differences in health behaviors varies by health behavior and for subgroups defined by indicators of SES, gender, and age. In setting the goals for this chapter, we challenged the often-held assumption that populations of color have poorer indicators of health behaviors than white populations. We found that unhealthy behaviors are not limited to any one racial/ethnic group. While whites had unhealthier behaviors than blacks and Hispanics for some behaviors, the reverse was true for other health behaviors. In general, racial/ethnic disparities were substantially influenced by education and income, were stronger for women than for men, and were stronger for younger and middle-aged adults than for older adults. Levels of unhealthy behaviors varied within racial/ethnic groups by level of education, household income, and, for Mexican Americans, by country of birth and language spoken.

Racial/Ethnic Disparities in Smoking

In our stratified analysis of BRFSS data, we found that SES greatly influenced smoking among white and black women and men. White women and men with lower SES (indicated by either educational attainment or income) at almost every age group had the highest prevalences of cigarette smoking compared to any other group. Lower SES black men, and to a lesser degree lower SES black women, had higher prevalences of smoking than their higher SES counterparts. In general, white women and men also had the highest probability of exposure to secondhand smoke. Hispanic women at every age group had lower odds of smoking compared with white women. The stratified analysis showed that lower SES white and black women and men (especially older black men) were the most likely to fail to quit smoking.

These findings are consistent with earlier findings from the National Health Interview Survey (NHIS) (Pamuk et al., 1998), which showed that smoking was highest for women and men with less than 12 years of education. In the NHIS, lower educated women and men, unlike higher educated women and men, showed almost no declines in smoking over the past 20 years, from 1974 to 1995 (Pamuk et al., 1998). This has produced a steeper SES gradient in smoking prevalence over time. Although black women and Hispanic women and men in the NHIS were less likely to smoke than their white counterparts, they also showed a gradient in smoking by SES.

Smoking is a particularly important health behavior because it is the leading cause of preventable death and disability in the United States and is a risk factor for heart disease, lung cancer, and other chronic diseases. It is critical to address the exceptionally high rates of smoking in lower educated white and black adults. Smoking prevention programs are needed that focus on white and black youth before they begin smoking and reach those who are at particular risk, such as those who attend continuation high schools and those who are not in school (Winkleby et al., 2004). Because many youth continue smoking into adulthood and other people begin smoking during adulthood, programs that encourage quit attempts and smoking cessation also must be available later in life, with a special focus on those with few financial resources. Given the influence of the tobacco industry, smoking programs and strategies must address the social context of smoking, such as the restriction of tobacco advertising, promotion of low-cost cigarettes, and sponsorship of sports events (Winkleby et al., 2004). In addition, smoking cessation programs, free or low-cost nicotine patches or gum, and support groups on weight and stress reduction are needed (Davis, 1987; Englander, 1986; Ernster, 1991; Pierce, Choi, Gilpin, Farkes, and Berry, 1998). Finally, the lower smoking among Hispanic women and men, and the more successful patterns of quitting for Hispanic women, should be reinforced, especially given that smoking levels are higher among those with longer “exposure” to the U.S. environment.

Racial/Ethnic Disparities in Obesity, Physical Inactivity, and Diet

We found that black women at every age group (with one exception) had significantly higher odds of obesity and physical inactivity than white women after adjustment for education and income. Black women also showed lower vegetable and/or fruit intake than white women at every age group, although these findings were not statistically significant, except for women aged 25 to 44. Black men aged 25 to 44 had significantly higher odds of obesity than white men. Some age groups of Hispanic women and men showed higher odds of leisure-time physical inactivity than white women and men. While neither Hispanic women nor men showed higher odds of obesity than white women and men after adjustment for income and education, there is some recent evidence that even Hispanic populations who work in blue-collar jobs have high prevalences of overweight; a 1999 random sample of 971 Hispanic farm workers from seven California communities found high levels of obesity among men (28 percent) and women (37 percent) during clinical examinations (Villarejo et al., 2000).

Our obesity findings for black women are consistent with the NHIS, NHANES III, and other surveys that show higher odds of obesity for black compared with white women (Kumanyika et al., 1993; Flegal et al., 1998). Our findings for Hispanic women are inconsistent with NHANES III data (Winkleby et al., 1999a) that show odds ratios for obesity for Mexican-American women that are intermediate between black and white women (odds ratios of 1.92 for black women and 1.48 for Mexican-American women with white women as the reference, p values < 0.01). Unlike our finding of higher odds of obesity for black compared with white men (ages 25 to 44), NHANES showed no higher odds of obesity for black men; however the men in the NHANES analysis were ages 25 to 64 (Winkleby et al., 1999a). These differences may be due to measurement differences (self-report versus clinical exam), definitions of race/ethnicity (Mexican American versus Hispanic), our stratification by age groups in the BRFSS, and/or variables used to adjust for SES.

Our findings for physical inactivity are consistent with past studies that have found that black adults (especially women) engage in less leisure-time physical activity than whites (Burke et al., 1992), that these differences remain after accounting for SES (Duelberg, 1992; Winkleby et al., 1998), and that the largest differences are among those with less than a high school education (Folsom et al., 1991). In the NHIS, black and Hispanic women and men aged 18 and older had higher prevalences of leisure-time inactivity than white women and men, with strong gradients evident by family income (Pamuk et al., 1998).

The health and medical care consequences of the current obesity epidemic in the United States are enormous. Overweight contributes to hypertension, diabetes, and other health conditions that are major risk factors for chronic diseases (Pi-Sunyer, 1993). It is well established that proper body weight, regular physical activity, and a diet high in vegetables and fruits can reduce related risk factors and chronic disease outcomes (National Heart, Lung, and Blood Institute, 1998). Given that more than 54 percent of Americans are overweight (an increase of 8 percent in less than 15 years) (Flegal, Carroll, Kuczmarski, and Johnson, 1998), prevention and treatment efforts must include both populationwide and tailored strategies. Of particular concern is the high prevalence of obesity in women, especially black women with lower SES. Because both ethnic minority status and lower SES are related to obesity, tailored treatment programs and behavioral strategies need to be developed that consider economic resources, literacy, culture, and language (Howard-Pitney, Winkleby, Albright, and Bruce, and Fortmann, 1993; Kumanyika et al., 1993). The rapid increases in obesity, physical inactivity, and poor dietary behaviors in the United States in the past decade point to environmental factors as underlying mechanisms and illustrate the need for broad societal changes. These include the availability of affordable, healthy foods in all neighborhoods; policy changes regarding processed and fast food sales; standards for food portion sizes; and safe, convenient places to walk and exercise (French, Story, and Jeffrey, 2001; Jeffery et al., 2000; Young and Nestle, 2002).

Racial/Ethnic Disparities in Alcohol Consumption

We found few racial/ethnic differences in heavy alcohol use in the BRFSS data. There was one notable exception: younger black women and men and younger Hispanic women had a 60 to 80 percent lower likelihood of high alcohol consumption than younger white women and men. The higher heavy alcohol consumption in younger white women and men points to the need for special outreach programs to address their high alcohol intake and/or binge drinking because this health behavior is linked to chronic diseases, injuries, adverse pregnancy outcomes, and other serious health problems.

The Influence of Acculturation

Within the Mexican-American population, we found that those who were born in the United States exhibited less healthy behaviors than those born in Mexico. Some of the largest differences by country of birth were among younger Mexican Americans for smoking (women), low vegetable and/or fruit consumption (women and men), obesity (women), and high alcohol consumption (women). A possible explanation for these findings is that those born in Mexico may have closer ties to the traditional Mexican culture that promotes family and community social support and healthy lifestyles such as nonsmoking, diets high in vegetables and fruits, and physical activity than those born in the United States (Sundquist and Winkleby, 1999).

Racial/Ethnic Disparities in Cervical Cancer and Breast Cancer Screening

We found that white women aged 25 to 64 were less likely to have received recent Pap screening than black women after adjustment for education and income and that Hispanic women aged 18 to 24 were less likely to have received recent Pap screening than white women. White women were also less likely to have received recent mammography screening than black or Hispanic women, although the difference in one age group was not statistically significant. All differences in mammography screening and several for Pap screening would have been missed if results were not adjusted for education and income. As with other health behaviors, the differences by SES were large.

These results for cancer screening are consistent with previous studies. A study of women aged 15 to 44 from the 1995 National Survey of Family Growth found that black women were more likely that any other racial group to be screened for cervical cancer (Hewitt, Devesa, and Breen, 2002). Another study using NHIS data showed that black women aged 50 and older were as likely or more likely than white women to have a mammogram in the past 2 years after controlling for income. NHIS differences were especially apparent for low-income women, with low-income black women being approximately 60 percent more likely than low-income white women to have had a recent mammogram. Low-income Hispanic women in the NHIS were intermediate between black and white women for mammography screening (Pamuk et al., 1998). Other studies have shown that Hispanic women receive Pap and mammography screening as often as white women when screening is covered as part of their health care insurance (Pérez-Stable, Otero-Sabogal, Sabogal, McPhee, and Hiatt, 1994).

Pap and mammography screening are well established as being effective in reducing breast and cervical cancer mortality, especially when the cancers are detected early (Young, 2002). Although we found that black and Hispanic women were more likely to report recent screening for cervical and breast cancer after adjustment for education and income, they are more likely to be diagnosed at later stages and have poorer survival following diagnosis (Miller et al., 1996). Potential explanations for why black and Hispanic women have higher screening rates but later diagnoses and poorer survival include poorer quality of screening tests, barriers to follow-up treatment including inadequate access to health care, and discrimination in the delivery of health services (Institute of Medicine, 1991).

The Influence of SES

Our results confirm the importance of considering socioeconomic factors when assessing racial/ethnic disparities in health behaviors related to chronic diseases. Adjustment for educational attainment and family income had a significant impact on our results, in some instances reducing disparities and in other instances heightening disparities. This has implications for interpreting past studies of racial/ethnic disparities, especially those that have not adjusted for SES.

Socioeconomic factors are important to consider in investigations of health behaviors in any population, but these factors have particular relevance in studies of racial/ethnic disparities. Researchers often adjust for SES using inadequate measures and then conclude that any remaining racial/ ethnic disparities are due to cultural and/or innate physiologic or genetic differences. The result is that considerable residual confounding by SES likely exists in studies of racial/ethnic disparities (Kaufman et al., 1997). Because SES is so difficult to measure completely, residual confounding by SES is likely to exist in any study of racial/ethnic disparities, even when multiple measures of SES are used. Care needs to be taken in the interpretation of “independent” effects of race or ethnicity, particularly given the overwhelming evidence that race/ethnicity is not genetically based and in light of recent evidence of the effects of discrimination on health (Krieger, 1999). Furthermore, studies need to emphasize that the variable “race/ ethnicity” is likely to capture unmeasured socioeconomic factors, even after adjusting for multiple measures of SES (Braveman et al., 2001). Thus, researchers need to acknowledge the limitations of the socioeconomic measures used to adjust for SES. For example, we were not able to consider a person's past SES, socioeconomic characteristics of her or his residential environment, or other factors such as occupational status or wealth. Based on these limitations, as well general limitations of the measures we used (e.g., no information on the quality of schooling), we believe that residual confounding by SES likely exists; that is, the variable race/ethnicity in our models represents, in part, unmeasured socioeconomic factors.

Based on our conceptual framework and on previous empirical evidence, it is evident that racial/ethnic disparities in health behaviors are complex phenomena and that factors beyond SES are partly responsible for their explanations. For example, while some studies show that Hispanic populations experience better health than would be expected given their socioeconomic characteristics alone (Franzini, Ribble, and Keddie, 2001; Markides and Coreil, 1986), findings from other studies counter these findings (Goff et al., 1997; Hunt et al., 2003). It is likely that acculturation is a primary factor accounting for the Hispanic health advantage in some studies. An important health policy question is whether this advantage will diminish as greater numbers of Mexican Americans (as well as other Hispanic populations) are born in the United States and live here longer.

Limitations of the BRFSS and Opportunities for Future Surveys

Although the BRFSS data have limitations that influence the interpretation of analyses on racial/ethnic differences in health behaviors, the opportunity exists to address these limitations in future surveys. The BRFSS is based on self-reported information on health behaviors collected by telephone interviews. Such data are often subject to bias; examples include the underreporting of weight or cigarette smoking or the overreporting of screening tests. In addition, the BRFSS has limited information on physical activity; it is collected for leisure time only. This misses physical activity during working hours that differentially affects racial/ethnic and socioeconomic groups (e.g., black, Hispanic, and lower SES populations are more likely to work in physically demanding jobs compared with white and higher SES populations). Future surveys should collect a broader array of standardized physical activity questions, including those that encompass activity during both leisure and work time.

A serious limitation of the BRFSS is the aggregation of persons of Hispanic origin into one group, making it impossible to distinguish among different Hispanic subgroups such as Mexican Americans, Puerto Ricans, and Cubans. This is important because of the large differences in prevalences of health behaviors within the Hispanic population (e.g., smoking in Mexican-American versus Cuban men). Future BRFSS surveys should delineate Hispanic subgroups. Emphasis also should be placed on including adequate samples of other major racial/ethnic groups in the United States (e.g., Native Americans and Alaska Natives; Chinese, Japanese, Vietnamese, and Filipinos, and other Asian Americans and Pacific Islanders).

Indicators of SES and acculturation available in the BRFSS are limited and lack specificity. If collected with more refinement, these indicators would allow for a more accurate assessment of the degree to which racial/ ethnic disparities are explained by SES. For example, income measures could be collected in such a way as to allow for categories of income in relation to the federal poverty level or could include estimates of childhood socioeconomic factors and adult wealth as is done in the National Longitudinal Survey of Youth. Education could include parental education, country of education, and name of the town or city where a person's highest education was obtained (to provide an indicator of geographic region of education and a surrogate of quality of education). Acculturation measures could include country of birth, language(s) spoken at home, and length of time lived in the United States.

The Next Generation of Chronic Disease Prevention

Chronic diseases, with heart disease ranking first, cancer ranking second, and stroke ranking third, will remain the leading causes of death in the United States for the next 50 years for all major racial/ethnic groups (Cooper et al., 2000). The National Conference on Cardiovascular Disease Prevention, held in 1999, addressed national trends in health behaviors related to CVD and other chronic diseases (Cooper et al., 2000). The conference leaders concluded that little progress has been made recently in addressing smoking, obesity, and physical inactivity despite widespread efforts to promote a populationwide adoption of healthy lifestyles, primary prevention for high-risk groups, and secondary prevention. Furthermore, the conference leaders stressed that wide racial/ethnic disparities in CVD mortality continue and that SES disparities in CVD mortality may be increasing. Their conclusions are supported by findings from other studies that show that the mortality disparity between lower and higher SES groups has widened (Pappas, Queen, Hadden, and Fisher, 1993).

The next generation of chronic disease prevention and control programs and policies must acknowledge and effectively address the social and historical context within which health behaviors are inextricably linked (Green and Kreuter, 1991; Minkler, 1990; Syme, 2004; Wallack and Winkleby, 1987; Wallerstein and Bernstein, 1994). The responsibility for improving health behaviors has been framed too often from an individual perspective that places the main responsibility for change with the individual. The rationale for this approach has been that once individuals are informed of their risk, they will adopt or modify behaviors to lower that risk (Wallack and Winkleby, 1987). Although an individual approach can be effective for addressing health problems (especially at the secondary and tertiary prevention levels), it has had limited success when used in isolation because it (1) places the burden for change on individuals who often are those with the fewest resources (e.g., socioeconomically disadvantaged); (2) can lead to increases in social disparities in health if those with the most resources and power (i.e., white and higher SES populations) are more able to take advantage of health-promoting programs, information, and policies to change their behaviors; (3) deflects attention away from important factors in the social and physical environment that influence choices regarding health-related behaviors; and (4) does not provide reinforcement of positive health behaviors from the environment in which a person lives and works.

In summary, we support a broad health policy agenda for the prevention of chronic diseases that integrates a focus on race/ethnicity, SES, and the social environment (Anderson, 1995; Williams and Collins, 1995). This is critical given that health behaviors are shaped by the communities in which people live (Syme, 2004). A broad focus on socioeconomic inequalities acknowledges the strong influence of SES on chronic disease outcomes, ensures the inclusion of all low-SES populations in health initiatives and guidelines, and achieves more equitable access to resources. Finally, it creates a more valid scientific ground for research on racial/ethnic disparities in health behaviors that goes beyond individual-level measures, and furthers an understanding that social, economic, and political factors are fundamental causes of health (Link and Phelan, 1995).

CONCLUSIONS

In this chapter we examined racial/ethnic disparities in a comprehensive set of health behaviors to assess the extent to which disparities varied across health behaviors, age groups, and gender, and to evaluate the contribution of indicators of SES to racial/ethnic disparities. We used data from national surveys that have large representative samples that allowed for a stratification of data across a wide range of age groups. We included women and men from the three largest ethnic groups in the United States, delineating Mexican Americans when possible. We focused on smoking, obesity, physical inactivity, poor diet, high alcohol consumption, and cancer screening practices, all of which are related to chronic diseases. Our findings highlight many disparities in health behaviors, none of which are restricted to any gender or age group. Furthermore, the disparities were greatly influenced by education and income.

The main conclusions from our BRFSS and NHANES III analyses are:

  • For some health behaviors, white populations have higher levels of unhealthy behaviors than black and/or Hispanic populations (particularly for smoking, secondhand smoke exposure, and inadequate Pap and mammogram screening), and for other health behaviors, the opposite is true (particularly for physical inactivity and obesity, with disparities being larger for blacks than for Hispanics). These disparities remain after adjustment for education and income.
  • Health behaviors also differ within racial/ethnic groups by important sociodemographic indicators, including age, educational attainment, household income, country of birth, and language spoken. These differences have implications for the timing, focus, and content of primary, secondary, and tertiary prevention programs and policies.
  • In general, racial/ethnic disparities in health behaviors are stronger for women than for men, in large part because of the greater disparities for women than for men for smoking, secondhand smoke exposure, physical inactivity, and obesity.
  • Racial/ethnic disparities in health behaviors tend to be stronger for younger and middle-aged adults than for older adults. This is apparent for smoking, secondhand smoke exposure, physical inactivity, high alcohol consumption, and inadequate mammography screening.
  • Both white and black adults with lower SES (as measured by either educational attainment or household income) have considerably less healthy behaviors than those with higher SES for all seven health behaviors, with the exception of high alcohol consumption. These differences show the importance of considering SES when planning and implementing health promotion and disease prevention programs.
  • Hispanic adults have different patterns of results than white and black adults. Few differences in health behaviors are evident between Hispanics and whites after adjustment for education and income. In addition, few differences are evident for Hispanics when stratified by education or income, except for obesity, physical inactivity, and mammography screening. However, large differences in health behaviors exist for Mexican Americans by country of birth; adults who are born in the United States and/or who speak English have higher predicted prevalences of unhealthy behaviors than those who are born in Mexico and/or who speak Spanish.

ACKNOWLEDGMENTS

This work was cofunded by the National Institute of Environmental Sciences and the National Heart, Lung, and Blood Institute: Grant RO1 HL67731 to Dr. Marilyn Winkleby. We thank Dr. David Ahn, Dr. Ying-Chih Chuang, and Dr. Michaela Kiernan for their valuable comments on an earlier draft, and Alana Koehler for her technical assistance in preparing the tables and figures.

REFERENCES

  1. Adler NE, Boyce T, Chesney M, Folkman S, Syme L. Socioeconomic inequalities in health: No easy solution. Journal of the American Medical Association. 1993;269:3140–3145. [PubMed: 8505817]
  2. Aguirre-Molina M, Molina CW, Zambrana RE. Health issues in the Latin community. San Francisco: Jossey-Bass; 2001.
  3. Anderson NB. Behavioral and sociocultural perspectives on ethnicity and health. Health Psychology. 1995;14:589–591. [PubMed: 8654336]
  4. Anderson RE, Crespo CJ, Bartlett SJ, Cheskin LJ, Pratt M. Relationship of physical activity and television watching with body weight and level of fatness among children: Results from the Third National Health and Nutrition Examination Survey. Journal of the American Medical Association. 1998;279:938–942. [PubMed: 9544768]
  5. Balcazar H, Cobas JA. Overweight among Mexican-Americans and its relationship to life style behavioral risk factors. Journal of Community Health. 1993;18:55–67. [PubMed: 8450094]
  6. Belcher JD, Ellison RC, Shepard WE, Bigelow C, Webber CS, Wilmore JH, Parcel GS, Zucker DM, Luepker RV. Lipid and lipoprotein distributions in children by ethnic group, gender, and geographic location: Preliminary findings of the Child and Adolescent Trial for Cardiovascular Health (CATCH) Preventive Medicine. 1993;22:143–153. [PubMed: 8512601]
  7. Berenson GS, Srinivasan SR, Bao W, Newman WP III, Tracy RE, Wattigney WA. Association between multiple cardiovascular risk factors and atherosclerosis in children and young adults. New England Journal of Medicine. 1998;338:1650–1656. [PubMed: 9614255]
  8. Black SA, Markides KS. Acculturation and alcohol consumption in Puerto Rican, Cuban-American, and Mexican-American women in the United States. American Journal of Public Health. 1993;83:890–893. [PMC free article: PMC1694747] [PubMed: 8498630]
  9. The Bogalusa Heart Study 20th Anniversary Symposium. American Journal of the Medical Sciences. 1995;310:S1–S138. [PubMed: 7503110]
  10. Braveman P, Cubbin C, Marchi K, Egerter S, Chavez G. Measuring socioeconomic status/position in studies of racial/ethnic disparities: Maternal and infant health. Public Health Reports. 2001;116:449–463. [PMC free article: PMC1497365] [PubMed: 12042609]
  11. Burke GL, Savage PJ, Manolio TA, Sprafka JM, Wagenknecht LE, Sidney S, Perkins LL, Liu K, Jacobs DR Jr. Correlates of obesity in young black and white women: The CARDIA Study. American Journal of Public Health. 1992;82:1621–1625. [PMC free article: PMC1694535] [PubMed: 1456336]
  12. Conference of Socioeconomic Status and Cardiovascular Health and Disease. Report of the Conference of Socioeconomic Status and Cardiovascular Health and Disease, November 6-7, 1995. Washington, DC: National Heart, Lung, and Blood Institute; 1995.
  13. Cooper R, Cutler J, Desvigne-Nickens P, Fortmann SP, Friedman L, Havlik R, Hogelin G, Marler J, McGovern P, Morosco G, Mosca L, Pearson T, Stamler J, Stryer D, Thom T. Trends and disparities in coronary heart disease, stroke, and other cardiovascular diseases in the United States. Findings of the National Conference on Cardiovascular Disease Prevention. Circulation. 2000;102:3137–3147. [PubMed: 11120707]
  14. Corti MC, Guralnik JM, Ferrucci L, Izmirlian G, Leveille SG, Pahor M, Cohen HJ, Pieper C, Havlik RJ. 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. 1999;89:308–314. [PMC free article: PMC1508599] [PubMed: 10076478]
  15. Cubbin C, Hadden WC, Winkleby MA. Neighborhood context and cardiovascular disease risk factors: The contribution of material deprivation. Ethnicity and Disease. 2001;11:687–700. [PubMed: 11763293]
  16. Davis RM. Current trends in cigarette advertising and marketing. New England Journal of Medicine. 1987;316:725–732. [PubMed: 3547127]
  17. Diehl AK, Stern MP. Special health problems of Mexican Americans: Obesity, gallbladder disease, diabetes mellitus, and cardiovascular disease. Advances in Internal Medicine. 1989;34:79–96. [PubMed: 2644769]
  18. Diez-Roux AV. Bringing context back into epidemiology: Variables and fallacies in multi-level analysis. American Journal of Public Health. 1998;88:216–222. [PMC free article: PMC1508189] [PubMed: 9491010]
  19. Diez-Roux AV, Nieto FJ, Muntaner C, Tyroler HA, Comstock GW, Shahar E, Cooper LS, Watson RL, Szklo M. Neighborhood environments and coronary heart disease: A multilevel analysis. American Journal of Epidemiology. 1997;146:48–63. [PubMed: 9215223]
  20. Diez-Roux AV, Nieto FJ, Caulfield L, Tyroler HA, Watson RL, Szklo M. Neighbourhood differences in diet: The Atherosclerosis Risk in Communities (ARIC) Study. Journal of Epidemiology and Community Health. 1999;53:55–63. [PMC free article: PMC1756776] [PubMed: 10326055]
  21. Diez-Roux AV, Merkin SS, Arnett D, Chambless L, Massing M, Nieto FJ, Sorlie P, Szklo M, Tyroler HA, Watson RL. Neighborhood of residence and incidence of coronary heart disease. New England Journal of Medicine. 2001;345:99–106. [PubMed: 11450679]
  22. DiPietro L, Williamson DF, Caspersen CJ, Eaker E. The descriptive epidemiology of selected physical activities and body weight among adults trying to lose weight: The Behavioral Risk Factor Surveillance System survey, 1989. International Journal of Obesity and Related Metabolic Disorders. 1993;17:69–76. [PubMed: 8384167]
  23. Duelberg SI. Preventive health behavior among black and white women in urban and rural areas. Social Science and Medicine. 1992;34:191–198. [PubMed: 1738872]
  24. Duncan C, Jones K, Moon G. Health-related behaviour in context: A multilevel modelling approach. Social Science and Medicine. 1996;42:817–830. [PubMed: 8778995]
  25. Dwyer JT, Stone EJ, Yang M, Feldman H, Webber LS, Must A, Perry CL, Nader PR, Parcel GS. Predictors of overweight and overfatness in a multiethnic pediatric population. Child and Adolescent Trial for Cardiovascular Health Collaborative Research Group. American Journal of Clinical Nutrition. 1998;67:602–610. [PubMed: 9537606]
  26. Elder JP, Castro FG, de Moor C, Mayer J, Candelaria JI, Campbell N, Talavera G, Ware LM. Differences in cancer-risk-related behaviors in Latino and Anglo adults. Preventive Medicine. 1991;20:751–763. [PubMed: 1766946]
  27. Ellaway A, Anderson A, Macintyre S. Does area of residence affect body size and shape? International Journal of Obesity and Related Metabolic Disorders. 1997;21:304–308. [PubMed: 9130028]
  28. Eller TJ. Household wealth and asset ownership: 1991. Washington, DC: U.S. Bureau of the Census; 1994. (Current Population Reports No. P70-34.
  29. Elo IT, Preston SH. Racial and ethnic differences in mortality at older ages. In: Martin LG, Soldo BJ, editors. Racial and ethnic differences in the health of older Americans. Washington, DC: National Academy Press; 1997. pp. 10–42. Committee on Population, Commission on Behavioral and Social Sciences and Education, National Research Council.
  30. Englander TJ. Cigarette makers shift and strategies. United States Tobacco and Candy Journal. 1986;213:1–46.
  31. Ernster V. How tobacco companies target women. World Smoking and Health. 1991;16:8–11.
  32. Espino DV, Burge SK, Moreno CA. The prevalence of selected chronic diseases among the Mexican-American elderly: Data from the 1982-1984 Hispanic Health and Nutrition Examination Survey. Journal of the American Board of Family Practice. 1991;4:217–222. [PubMed: 1927588]
  33. Flegal KM, Carroll MD, Kuczmarski RJ, Johnson CL. Overweight and obesity in the United States: Prevalence and trends, 1960-1994. International Journal of Obesity and Related Metabolic Disorders. 1998;22:39–47. [PubMed: 9481598]
  34. Folsom AR, Burke GL, Ballew C, Jacobs DR Jr, Haskell WL, Donahue RP, Liu KA, Hilner JE. Relation of body fatness and its distribution to cardiovascular risk factors in young blacks and whites. The role of insulin. American Journal of Epidemiology. 1989;130:911–924. [PubMed: 2683750]
  35. Folsom AR, Cook TC, Sprafka JM, Burke GL, Norsted SW, Jacobs DR Jr. Differences in leisure-time physical activity levels between blacks and whites in population-based samples: The Minnesota Heart Survey. Journal of Behavioral Medicine. 1991;14:1–9. [PubMed: 2038041]
  36. Franzini L, Ribble JC, Keddie AM. Understanding the Hispanic paradox. Ethnicity and Disease. 2001;11:496–518. [PubMed: 11572416]
  37. French SA, Story M, Jeffery RW. Environmental influences on eating and physical activity. Annual Review of Public Health. 2001;22:309–335. [PubMed: 11274524]
  38. Geronimus AT. The weathering hypothesis and the health of African-American women and infants: Evidence and speculations. Ethnicity and Disease. 1992;2:207–221. [PubMed: 1467758]
  39. Gidding SS, Liu K, Bild DE, Flack J, Gardin J, Ruth KJ, Oberman A. Prevalence and identification of abnormal lipoprotein levels in a biracial population aged 23 to 35 years: The CARDIA Study. American Journal of Cardiology. 1996;78:304–308. [PubMed: 8759809]
  40. Goff DC, Nichaman MZ, Chan W, Ramsey DJ, Labarthe DR, Ortiz C. Greater incidence of hospitalized myocardial infarction among Mexican Americans than non-Hispanic whites: The Corpus Christi Heart Project 1988-1992. Circulation. 1997;95:1433–1440. [PubMed: 9118510]
  41. Green LW, Kreuter MW. Health promotion planning: An educational and environmental approach. 2nd ed. Mountain View, CA: Mayfield; 1991.
  42. Gutierrez-Ramirez A, Valdez RB, Carter-Pokras O. Cancer. In: Molina CW, Aguirre-Molina M, editors. Latino health in the U.S.: A growing challenge. Washington, DC: American Public Health Association; 1994. pp. 211–246.
  43. Haffner SM, Stern MP, Hazuda HP, Pugh JA, Patterson JK, Malina R. Upper body and centralized adiposity in Mexican-Americans and non-Hispanic whites: Relationship to body mass index and other behavioral and demographic variables. International Journal of Obesity. 1986;10:493–502. [PubMed: 3804566]
  44. Hart C, Ecob R, Smith GD. People, places and coronary heart disease risk factors: A multilevel analysis of the Scottish Heart Health Study archive. Social Science and Medicine. 1997;45:893–902. [PubMed: 9255922]
  45. Haynes SG, Harvey C, Montes H, Nickens H, Cohen BH. Patterns of cigarette smoking among Hispanics in the United States: Results from HHANES 1982-84. American Journal of Public Health. 1990;80(Suppl.):47–53. [PMC free article: PMC1404508] [PubMed: 9187582]
  46. Hazuda HP, Stern MP, Haffner SM. Acculturation and assimilation among Mexican Americans: Scales and population-based data. Social Science Quarterly. 1988;69:687–706.
  47. Hewitt M, Devesa S, Breen N. Papanicolaou test use among reproductive-age women at high risk for cervical cancer: Analyses of the 1995 National Survey of Family Growth. American Journal of Public Health. 2002;92:666–669. [PMC free article: PMC1447134] [PubMed: 11919069]
  48. Hollman F. U.S. population estimates by age, sex, race, and Hispanic origin: 1980 to 1991. Washington, DC: U.S. Bureau of the Census; 1993. (Current Population Reports, Series P-25, No. 1095)
  49. Howard-Pitney B, Winkleby MA, Albright CL, Bruce B, Fortmann SP. The Stanford Nutrition Action Program: A dietary fat intervention for low literate adults. American Journal of Public Health. 1997;87:1971–1976. [PMC free article: PMC1381239] [PubMed: 9431286]
  50. Hunt KJ, Resendez RG, Williams K, Haffner SM, Stern MP, Hazuda HP. All-cause and cardiovascular mortality among Mexican-American and non-Hispanic white older participants in the San Antonio Heart Study—evidence against the “Hispanic paradox”. American Journal of Epidemiology. 2003;158:1048–1057. [PubMed: 14630600]
  51. Institute of Medicine. Disability in America: Toward a national agenda for prevention. Washington, DC: National Academy Press; 1991.
  52. Jasso G, Massey DS, Rosenzweig MR, Smith JP. The New Immigrant Survey Pilot (NIS-P): Overview and new findings about U.S. legal immigrants at admission. Demography. 2000;37:127–138. [PubMed: 10748994]
  53. Jeffery RW, Drewnowski A, Epstein LH, Stunkard AJ, Wilson GT, Wing RR. Long-term maintenance of weight loss: Current status. Health Psychology. 2000;19:5–16. [PubMed: 10709944]
  54. Jemal A, Thomas A, Murray T, Thun M. Cancer statistics, 2002. A Cancer Journal for Clinicians. 2002;52:23–47. [PubMed: 11814064]
  55. Jones CP. Levels of racism: a theoretic framework and a gardener's tale. American Journal of Public Health. 2000;90:1212–1215. [PMC free article: PMC1446334] [PubMed: 10936998]
  56. Kaplan GA, Keil JE. Socioeconomic factors and cardiovascular disease: A review of the literature. Circulation. 1993;88:1973–1998. [PubMed: 8403348]
  57. Kaufman JS, Cooper RS, McGee DL. Socioeconomic status and health in blacks and whites: The problem of residual confounding and the resiliency of race. Epidemiology. 1997;8:621–628. [PubMed: 9345660]
  58. Krieger N. Embodying inequality: A review of concepts, measures, and methods for studying health consequences of discrimination. International Journal of Health Services. 1999;29:295–352. [PubMed: 10379455]
  59. Krieger N. A glossary for social epidemiology. Journal of Epidemiology and Community Health. 2001;55:693–700. [PMC free article: PMC1731785] [PubMed: 11553651]
  60. Krieger N, Williams DR, Moss NE. Measuring social class in U.S. public health research: Concepts, methodologies, and guidelines. Annual Review of Public Health. 1997;18:341–378. [PubMed: 9143723]
  61. Kuczmarski RJ, Flegal KM, Campbell SM, Johnson CL. Increasing prevalence of overweight among U.S. adults: The National Health and Nutrition Examination Surveys, 1960 to 1991. Journal of the American Medical Association. 1994;272:205–211. [PubMed: 8022039]
  62. Kumanyika S, Wilson JF, Guilford-Davenport M. Weight-related attitudes and behaviors of black women. Journal of the American Dietetic Association. 1993;93:416–422. [PubMed: 8454809]
  63. Kumanyika SK. Special issues regarding obesity in minority populations. Annals of Internal Medicine. 1993;119:650–654. [PubMed: 8363191]
  64. LeClere FB, Rogers RG, Peters K. Neighborhood social context and racial differences in women's heart disease mortality. Journal of Health and Social Behavior. 1998;39:91–107. [PubMed: 9642901]
  65. Lee RE, Cubbin C. Neighborhood context and youth cardiovascular health behaviors. American Journal of Public Health. 2002;92:428–436. [PMC free article: PMC1447094] [PubMed: 11867325]
  66. Link BG, Phelan J. Social conditions as fundamental causes of disease. Journal of Health and Social Behavior. 1995;35(Suppl.):80–94. [PubMed: 7560851]
  67. Lowry R, Kann L, Collins JL, Kolbe LJ. The effect of socioeconomic status on chronic disease risk behaviors among US adolescents. Journal of the American Medical Association. 1996;276:792–797. [PubMed: 8769588]
  68. Lynch JW, Kaplan GA, Cohen RD, Tuomilehto J, Salonen JT. Do cardiovascular risk factors explain the relation between socioeconomic status, risk of all-cause mortality, and acute myocardial infarction? American Journal of Epidemiology. 1996;144:934–942. [PubMed: 8916504]
  69. Lynch JW, Kaplan GA, Shema SJ. Cumulative impact of sustained economic hardship on physical, cognitive, psychological, and social functioning. New England Journal of Medicine. 1997;337:1889–1895. [PubMed: 9407157]
  70. Macintyre S, Maciver S, Sooman A. Area, class and health: Should we be focusing on places or people? Journal of Society and Politics. 1993;22:213–234.
  71. Macintyre S, Ellaway A, Cummins S. Place effects on health: How can we conceptualise, operationalise, and measure them? Social Science and Medicine. 2002;55:125–139. [PubMed: 12137182]
  72. Markides KS, Coreil J. The health of Hispanics in the southwestern United States: An epidemiologic paradox. Public Health Reports. 1986;101:253–265. [PMC free article: PMC1477704] [PubMed: 3086917]
  73. Markides KS, Ray LA, Stroup-Benham CA, Trevino F. Acculturation and alcohol consumption in the Mexican American population of the southwestern United States: Findings from HHANES 1982-84. American Journal of Public Health. 1990;80(Suppl.):42–46. [PMC free article: PMC1404517] [PubMed: 9187581]
  74. Marmot M, Elliot P. Coronary heart disease epidemiology from aetiology to public health. New York: Oxford University Press; 1992.
  75. Martin LG, Soldo BJ. Racial and ethnic differences in the health of older Americans. Washington, DC: National Academy Press; 1997.
  76. McNutt SW, Hu Y, Schreiber GB, Crawford PB, Obarzanek E, Mellin L. A longitudinal study of the dietary practices of black and white girls 9 and 10 years old at enrollment: The NHLBI Growth and Health Study. Journal of Adolescent Health. 1997;20:27–37. [PubMed: 9007656]
  77. Miller BA, Kolonel LN, Bernstein L, Young JL Jr, Swanson GM, West DW, Key CR, Liff JM, Glover CS, Alexander GA, Coyle L, Hankey BF, Gloeckler Ries LA, Kosary CL, Harras A, Percy C, Edwards BK. Racial/ethnic patterns of cancer in the United States 1988-1992. Bethesda, MD: National Cancer Institute; 1996. (NIH Pub. No. 96-4104)
  78. Minkler M. Improving health through community organization. In: Glanz K, Lewis FM, Rimer BK, editors. Health behavior and health education: Theory, research and practice. San Francisco: Jossey-Bass; 1990. pp. 257–287.
  79. Mitchell BD, Stern MP, Haffner SM, Hazuda HP, Patterson JK. Risk factors for cardiovascular mortality in Mexican Americans and non-Hispanic whites: The San Antonio Heart Study. American Journal of Epidemiology. 1990;131:423–433. [PubMed: 2301352]
  80. Muntaner C, Nieto FJ, O'Campo P. The Bell Curve: On race, social class, and epidemiologic research. American Journal of Epidemiology. 1996;144:531–535. [PubMed: 8797512]
  81. National Cancer Institute. SEER program public-use data tapes 1973-1998, August 2000 submission. Bethesda, MD: National Institutes of Health; 2001.
  82. National Center for Health Statistics. National Health Interview Survey/Multiple Cause of Death Public Use Data 1986-1990. Diskette and documentation. Hyattsville, MD: Centers for Disease Control and Prevention; 1994a.
  83. National Center for Health Statistics. Plan and operation of the Third National Health and Nutrition Examination Survey, 1988-1994, series 1: Programs and collection procedures. Vital Health Statistics. 1994b;32:1–407. [PubMed: 7975354]
  84. National Heart, Lung, and Blood Institute. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: The evidence report (Rep. No. 98-4083). Bethesda, MD: National Institutes of Health; 1998. [PubMed: 9813653]
  85. National Research Council. Racial and ethnic differences in the health of older Americans. Martin LG, Soldo BJ, editors. Washington, DC: National Academy Press; 1997. Committee on Population, Commission on Behavioral and Social Sciences and Education. [PubMed: 23115810]
  86. Otten MW Jr, Teutsch SM, Williamson DF, Marks JS. The effect of known risk factors on the excess mortality of black adults in the United States. Journal of the American Medical Association. 1990;263:845–850. [PubMed: 2296146]
  87. Padilla AM, editor. Acculturation: Theory, models, and some new findings. Boulder, CO: Westview Press; 1980. (AAAS Pub. No. SS(NS)-39 ed.). American Association for the Advancement of Science.
  88. Pamuk E, Makuc D, Heck K, Reuben C, Lochner K. Socioeconomic status and health chartbook: Health, United States, 1998. Hyattsville, MD: National Center for Health Statistics; 1998. (Rep. No. 71-641496)
  89. Pappas G, Queen S, Hadden W, Fisher G. The increasing disparity in mortality between socioeconomic groups in the United States, 1960 and 1986. New England Journal of Medicine. 1993;329:103–109. [PubMed: 8510686]
  90. Pérez-Stable EJ, Otero-Sabogal R, Sabogal F, McPhee SJ, Hiatt RA. Self-reported use of cancer screening tests among Latinos and Anglos in a prepaid health plan. Archives of Internal Medicine. 1994;154:1073–1081. [PubMed: 8185420]
  91. Pi-Sunyer FX. Medical hazards of obesity. Annals of Internal Medicine. 1993;119:655–660. [PubMed: 8363192]
  92. Pickett KE, Pearl M. Multilevel analyses of neighborhood socioeconomic context and health outcomes: A critical review. Journal of Epidemiology and Community Health. 2001;55:111–122. [PMC free article: PMC1731829] [PubMed: 11154250]
  93. Pierce JP, Choi WS, Gilpin EA, Farkas AJ, Berry CC. Tobacco industry promotion of cigarettes and adolescent smoking. Journal of the American Medical Association. 1998;279:511–515. [PubMed: 9480360]
  94. Ragland K, Selvin S, Merrill D. Black-white differences in stage-specific cancer survival: Analysis of seven selected sites. American Journal of Epidemiology. 1991;133:672–682. [PubMed: 2018022]
  95. Rogers RG. Health-related lifestyles among Mexican Americans, Puerto Ricans, and Cubans in the United States. In: Rapsenwaike I, editor. Mortality of Hispanic patients. New York: Greenwood Press; 1991. pp. 145–167.
  96. Runciman WG, editor. Weber: Selections in translation. Cambridge, England: Cambridge University Press; 1978.
  97. Shah BV, Barnwell BG, Hunt PN, Nileen P, LaVange LM. SUDAAN user's manual, release 5.50. Research Triangle Park, NC: Research Triangle Institute; 1991.
  98. Singh GK, Kochanek KD, McDonan MF. Advance report of final mortality statistics, 1994. Hyattsville, MD: National Center for Health Statistics; 1996.
  99. Smith JP, Kington RS. Race, socioeconomic status, and health in late life. In: Martin LG, Soldo BJ, editors. Racial and ethnic differences in the health of older Americans. Washington, DC: National Academy Press; 1997. pp. 106–162. Committee on Population, Commission on Behavioral and Social Sciences and Education, National Research Council.
  100. Smith GD, Neaton JD, Wentworth D, Stamler R, Stamler J. Socioeconomic differentials in mortality risk among men screened for the Multiple Risk Factor Intervention Trial: I. White men. American Journal of Public Health. 1996;86:486–496. [PMC free article: PMC1380548] [PubMed: 8604778]
  101. Smith GD, Hart C, Watt G, Hole D, Hawthorne V. Individual social class, area-based deprivation, cardiovascular disease risk factors, and mortality: The Renfrew and Paisley Study. Journal of Epidemiology and Community Health. 1998;52:399–405. [PMC free article: PMC1756721] [PubMed: 9764262]
  102. Sorlie PD, Backlund E, Johnson NJ, Rogot E. Mortality by Hispanic status in the United States. Journal of the American Medical Association. 1993;270:2464–2468. [PubMed: 8031341]
  103. Srinivasan SR, Bao W, Wattigney WA, Berenson GS. Adolescent overweight is associated with adult overweight and related multiple cardiovascular risk factors: The Bogalusa Heart Study. Metabolism: Clinical and Experimental. 1996;45:235–240. [PubMed: 8596496]
  104. Stephen EH, Foote K, Hendershot GE, Schoenborn CA. Health of the foreign-born population: United States, 1989-90. Advance data. Vital Health Statistics. 1994;241:1–12. [PubMed: 10132138]
  105. Sundquist J, Winkleby MA. Cardiovascular risk factors in Mexican American adults: A transcultural analysis of NHANES III, 1988-1994. American Journal of Public Health. 1999;89:723–730. [PMC free article: PMC1508740] [PubMed: 10224985]
  106. Sundquist J, Malmstrom M, Johansson SE. Cardiovascular risk factors and the neighbourhood environment: A multilevel analysis. International Journal of Epidemiology. 1999;28:841–845. [PubMed: 10597980]
  107. Sundquist J, Winkleby MA, Pudaric S. Cardiovascular disease risk factors among older black, Mexican American, and white women and men: An analysis of NHANES III, 1988-1994. Journal of the American Geriatrics Society. 2001;49:109–116. [PubMed: 11207863]
  108. Syme SL. Social determinants of health: The community as an empowered partner. 2004. Available: http://www​.cdc.gov/pcd​/issues/2004/jan/syme.htm. [PMC free article: PMC544525] [PubMed: 15634364]
  109. Tortolero SR, Goff DC Jr, Nichaman MZ, Labarthe DR, Grunbaum JA, Hanis CL. Cardiovascular risk factors in Mexican-American and non-Hispanic white children: The Corpus Christi Child Heart Study. Circulation. 1997;96:418–423. [PubMed: 9244206]
  110. Tyroler HA, Wing S, Knowles MG. Increasing inequality in coronary heart disease mortality in relation to educational achievement: Profile of places of residence, United States, 1962 to 1987. Annals of Epidemiology. 1993;3(Suppl.):S51–54.
  111. U.S. Census Bureau. Population Projections of the United States by Age, Sex, Race, Hispanic Origin, and Nativity: 1999 to 2100. 2001. [Accessed January 26, 2002]. Available: http://www​.census.gov​/population/projections​/nation/detail/np-d1-a.txt.
  112. U.S. Department of Health and Human Services. Women and smoking: Report of the Surgeon General. Washington, DC: National Center for Chronic Disease Prevention and Health Promotion; 2001.
  113. Vega WA, Amaro H. Latino outlook: Good health, uncertain prognosis. Annual Reviews in Public Health. 1994;15:39–67. [PubMed: 8054092]
  114. Ventura SJ, Peters KD, Martin JA, Maurer JD. Births and deaths: United States, 1996. Monthly Vital Statistics Report. 1997;46:1–40. [PubMed: 9404390]
  115. Villarejo D, Lighthall D, Williams D III, Souter A, Mines R, Bade B, Samuels S, McCurdy S. Suffering in silence: A report on the health of California's agricultural workers. Davis, CA: California Institute for Rural Studies; 2000.
  116. Wallack L, Winkleby M. Primary prevention: A new look at basic concepts. Social Science and Medicine. 1987;25:923–930. [PubMed: 3317892]
  117. Wallerstein N, Bernstein E. Introduction to community empowerment, participatory education, and health. Health Education Quarterly. 1994;21:141–148. [PubMed: 8021144]
  118. Williams DR. Race/ethnicity and socioeconomic status: Measurement and methodological issues. International Journal of Health Services. 1996;26:483–505. [PubMed: 8840198]
  119. Williams DR, Collins C. U.S. socioeconomic and racial differences in health: Patterns and explanations. Annual Review of Sociology. 1995;21:349–386.
  120. Winkleby MA, Cubbin C. Influence of individual and neighborhood socioeconomic status on mortality among Black, Mexican-American, and White women and men in the U.S. Journal of Epidemiology and Community Health. 2003;57:444–452. [PMC free article: PMC1732485] [PubMed: 12775792]
  121. Winkleby MA, Fortmann SP, Rockhill B. Health-related risk factors in a sample of Hispanics and whites matched on sociodemographic characteristics: The Stanford Five-City Project. American Journal of Epidemiology. 1993;137:1365–1375. [PubMed: 8333418]
  122. Winkleby MA, Schooler C, Kraemer HC, Lin J, Fortmann SP. Hispanic versus white smoking patterns by sex and level of education. American Journal of Epidemiology. 1995;142:410–418. [PubMed: 7625406]
  123. Winkleby MA, Kraemer HC, Ahn DK, Varady AN. Ethnic and socioeconomic differences in cardiovascular disease risk factors: Findings for women from the Third National Health and Nutrition Examination Survey, 1988-1994. Journal of the American Medical Association. 1998;280:356–362. [PubMed: 9686553]
  124. Winkleby MA, Cubbin C, Ahn DK, Kraemer HC. Pathways by which SES and ethnicity influence cardiovascular disease risk factors. Annals of the New York Academy of Sciences. 1999a;896:191–209. [PubMed: 10681898]
  125. Winkleby MA, Robinson TN, Sundquist J, Kraemer HC. Ethnic variation in cardiovascular risk factors among children and young adults: Findings from the Third National Health and Nutrition Examination Survey, 1988-1994. Journal of the American Medical Association. 1999b;281:1006–1013. [PubMed: 10086435]
  126. Winkleby MA, Feighery E, Dunn M, Kole S, Ahn D, Killen J. Effects of an advocacy intervention to reduce smoking among teenagers. Archives of Pediatrics and Adolescent Medicine. 2004;158:269–275. [PubMed: 14993087]
  127. Yen IH, Kaplan GA. Poverty area residence and changes in physical activity level: Evidence from the Alameda County Study. American Journal of Public Health. 1998;88:1709–1712. [PMC free article: PMC1508581] [PubMed: 9807543]
  128. Young LR, Nestle M. The contribution of expanding portion sizes to the U.S. obesity epidemic. American Journal of Public Health. 2002;92:246–249. [PMC free article: PMC1447051] [PubMed: 11818300]
  129. Young RC. Cancer statistics, 2002: Progress or cause for concern? CA: A Cancer Journal for Clinicians. 2002;52:6–7. [PubMed: 11814066]
Copyright © 2004, National Academy of Sciences.
Bookshelf ID: NBK25518

Views

  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (9.9M)

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...