CHAPTER 109: RACE, ETHNICITY, AND CARDIOVASCULAR DISEASE

Race and ethnicity are social, political, historical, and cultural constructs.^1,2 As such, they are flawed as biological concepts and inappropriate or, at best, imprecise when used as proxy for genetic or other biological phenomena.^1,2,3,4,5 Consistent with this premise, racial and ethnic differences observed in the incidence, prevalence, morbidity, and mortality of cardiovascular diseases (CVD) should not be construed as necessarily resulting from genetic or other biological differences. Additionally, race and ethnicity are often confounded by socioeconomic status (SES). Thus, observed racial and ethnic differences that are not appropriately adjusted for SES and related factors could lead to erroneous conclusions with potentially adverse clinical and public health implications. Despite these challenges, data on race and ethnicity, when properly collected, analyzed, and interpreted, can be invaluable in clinical and public health practice.^6,7,8 In particular, the appropriate collection and use of data on race and ethnicity are crucial in the quest to eliminate racial and ethnic disparities in cardiovascular health and advance health equity.^9,10

In this chapter, we first address the historical origins and definitions of race and ethnicity, and how the terms have been used within the context of clinical and public health practice in the United States. The established racial and ethnic differences in the magnitude, distribution, outcomes, and trends of leading cardiovascular causes of morbidity and mortality are then presented with emphasis on their trends over the past 25 years. The role that research has played in advancing knowledge on race, ethnicity, and CVD is summarized. We identify the critical gaps in knowledge about race, ethnicity, and CVD that present strategic opportunities for future cardiovascular research to advance health equity.

Historical and anthropologic data suggest that the ideology of race was well established in the United States by the 18th century.³ At that time, the three major labels that were used to define, group, and rank people were “European whites,” “Native Americans” or “Indians,” and “Negroes” from Africa.³ These were purely social, legal, and political constructs that had no basis in biological evidence. As Richard Cooper eloquently stated, although race is devoid of any aspects of the “dominant empiricism of modern biology,” clinical medicine and public health practice in the United States have embraced racial/ethnic categories as “fundamental structural elements."¹¹ This use of race/ethnic categories in clinical medicine and public health has remained contentious and controversial, especially when race is viewed or interpreted as a biological construct.¹²

Historically, the biological concept of race has been associated with hierarchical ranking, biological determinism, eugenics, and justification for genocide, colonialism, slavery and other social inequities.¹³ It should not be a surprise that the use of racial classifications in medicine is met with skepticism and fear. A major problem with using race as an identifier is the lack of clear definition.¹⁴ Racial admixture and not knowing family ancestry further erode the utility of race classification. Race as a surrogate for biological characteristics and geographic origin is, therefore, imprecise. Culture, language, religion, and ethnicity have strong sociocultural components and are often poor indicators of ancestry.¹³ Finding better ways to measure ancestry and diversity is more likely to yield greater conceptual precision and could enable identification of useful genetic determinants of complex diseases.^13,15 Importantly, race and ethnicity have been shown to introduce provider bias about an individual's ability or willingness to accept care and may adversely impact disease management.^16,17 Despite these limitations, race and ethnicity are associated with differences in disease prevalence, expression, and outcome, as discussed in other sections of this chapter.

In October 1977, the US Government's Office of Management and Budget (OMB) announced government-wide standards for the collection of federal data on race and ethnicity. This has since been revised to account for the changing demographics of the United States.¹⁸ These standards are required for federal data collection and routinely used in medical and clinical research.^19,20 The revised standards contain five minimum categories for race and two categories for ethnicity, as shown in Table 109–1.¹⁹

Population-based cohort studies²¹ have advanced the understanding of CVDs in racial and ethnic minority populations. As shown in Table 109–2, these cohort studies include the Jackson Heart Study, an African American cohort; Strong Heat Study, a Native American and Alaskan cohort; Hispanic Community Health Study/Study of Latinos (HCHS/SOL), a cohort of Central Americans, Cubans, Dominicans, Mexicans, Puerto Ricans, and South Americans; Coronary Artery Disease Risk in Young Adults (CARDIA), a biracial cohort of young adults; Atherosclerosis Risk in Communities (ARIC), a biracial cohort; and Multi-Ethnic Study of Atherosclerosis (MESA), a multiethnic cohort. These cohort studies have advanced and refined our understanding of disease progression and outcomes in population subgroups and remain a continuing resource for further research in racial and ethnic minority populations and research training in cardiovascular health in racial and ethnic minorities. Examples of intervention studies are shown in Table 109–3. The results from many of these intervention studies have played a crucial role in the improvements in outcomes for hypertension, coronary artery disease, heart failure, and other CVDs. These improvements have not impacted all populations equally. In the next section of this chapter, we review the progress made and persisting challenges in racial and ethnic disparities in CVD.

Racial and ethnic differences in cardiovascular risk factors and CVD have been well described in the United States for decades. For example, using national survey data on risk factor prevalence and indexes of morbidity, mortality, and overall quality of life in adults 18 years of age or older, Mensah et al²² assessed the magnitude of disparities in CVD in the United States in 2005. That assessment concluded that disparities in CVD were pervasive. In particular, racial and ethnic disparities were common in all risk factors examined. For example, hypertension prevalence was highest among blacks (48.7%) regardless of sex or educational status. Hypercholesterolemia was highest among white and Mexican-American men (49.2%) and white women (56.9%) regardless of educational status. In men, the highest prevalence of obesity was found in Mexican Americans (29.2%) who had completed a high school education.²² Black women (47.7%) with or without a high school education had a high prevalence of obesity among Medicare enrollees; congestive heart failure hospitalization was higher in blacks, Hispanics, and American Indians/Alaska Natives than among whites; and stroke hospitalization was highest in blacks. Age-adjusted mortality from CVD at all ages tended to be highest in blacks. Overall life expectancy was higher in whites (77.2 years) than in blacks (71.8 years) by approximately 5 years. Although disparities by race and ethnicity were prominent, so were differences by sex, educational attainment, SES, and geographic location. Nevertheless, racial and ethnic disparities in CVD and risk factors were pervasive even in 2005.²²

Progress in Reducing Disparities

In the decade since that assessment, some progress has been made in the quest to improve healthcare quality and reduce or eliminate health disparities.²³ Although most health disparities have remained unchanged and some have widened, there is evidence to demonstrate that, in some areas, disparities in the burden of CVD and quality of healthcare have improved. For example, in federally funded research on equity and quality of care in US hospitals, Trivedi et al²⁴ assessed performance rates for 17 objective quality measures covering three conditions (six measures for acute myocardial infarction, four for heart failure, and seven for pneumonia). The performance rates were adjusted for patient- and hospital-level covariates and compared among non-Hispanic white, non-Hispanic black, and Hispanic patients who received care between 2005 and 2010 in acute care hospitals throughout the United States. The adjusted performance rates for the 17 quality measures improved by 3.4 to 57.6 percentage points between 2005 and 2010 for white, black, and Hispanic adults (P < .001 for all comparisons).²⁴ Figure 109–1 shows changes in within-hospital and between-hospital differences from 2005 to 2010. The three measures (in black-white comparisons, panel A) and six measures (in Hispanic-white comparisons, panel B) shown are those with adjusted rate differences that exceeded 5 percentage points in 2005. As the figure demonstrates, significant improvement in performance on quality measures for white, black, and Hispanic adults hospitalized for acute myocardial infarction, heart failure, or pneumonia was accompanied by significant reduction in racial and ethnic disparities in performance rates both within and among US hospitals.²⁴ As another sign of progress, Cohen et al²⁵ reported that among 443 hospitals participating in the Get With the Guidelines–Coronary Artery Disease national quality monitoring and improvement program between January 2002 and June 2007, racial/ethnic differences in individual and composite care were reduced or eliminated. Individual core performance measures included appropriate use of angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, use of β-blockers at discharge, appropriate use of lipid-lowering drugs, use of smoking cessation counseling, use of aspirin within 24 hours of admission, and use of aspirin at discharge. The composite performance measure of defect-free care was defined as the proportion of patients who received all interventions for which they were eligible.

Persisting Disparities in Health Behaviors and Cardiovascular Risk Factors

Although examples of successful reduction or elimination of racial and ethnic disparities in CVD exist, as demonstrated in the studies described above,^23,24,25 there are many more instances showing persistence or widening of disparities. For example, Chang et al²⁶ used underlying cause of death data and population estimates from the National Vital Statistics System to calculate age-adjusted death rates for CVD and nine other leading causes of death during 1999 to 2010 in order to explore temporal trends in disparities in death rates by sex and race/ethnicity in the United States during this time period. Substantial declines in age-adjusted mortality for CVD were noted in all race/ethnic groups, as shown in Table 109–4.²⁶ Among the 10 leading causes of death, age-adjusted death rates by sex and race/ethnicity declined from 1999 to 2010 for six causes and increased for four causes; however, sex and racial/ethnic disparities between groups persisted for each year and cause of death.²⁶ For heart disease in 2010, the race/ethnic group with the largest disparity relative to the most favorable group, Asian/Pacific Islander (men 127.3; women 81.4), was the non-Hispanic blacks (men 286.3; women 189.1). Death rates reported were age-adjusted per 100 000 persons based on the 2000 US Census standard population.²⁶ The primary drivers of these persisting disparities included those attributed to cardiovascular health behaviors and risk factors.

In their estimates of age-adjusted mortality from individuals from 3099 counties in the United States, Vaughan et al²⁷ observed county-level race-sex disparities in heart disease mortality trends. For example, from 1973 to 2010, they observed continual strong declines in 73.2%, 44.6%, 15.5%, and 17.3% of white men, white women, black men, and black women, respectively (Fig. 109–2).²⁷ Additionally, during the period of 1998 to 2010, delayed strong declines occurred in 15.4%, 42.0%, 75.5%, and 76.6% of counties for white men, white women, black men, and black women, respectively (see Fig. 109–2).²⁷ The authors concluded that over the nearly four-decade period of the analysis, patterns of decline in age-adjusted mortality for heart disease differed by race and geography, possibly reflecting disparities in national and local drivers of these declines.²⁷ Better understanding of racial and geographic disparities in the diffusion of heart disease prevention and treatment may allow us to make progress toward achieving racial and geographic equity in heart disease survival.

Not surprisingly, hypertension remains a major contributor to the persisting large disparity in age-adjusted mortality from heart disease. The age-adjusted hypertension-related mortality in non-Hispanic blacks is significantly higher than the rate seen in non-Hispanic whites and Hispanics.^28,29 The disparity is more dramatic when examined by sex (Fig. 109–3).²⁸ For example, in 2013, the death rates per 100,000 population were 51.6 for non-Hispanic black men, but 18.9 for non-Hispanic white men, and 20.0 for Hispanic males.²⁸ The corresponding rates for women were 36.5 for non-Hispanic black women, 15.8 for non-Hispanic white women, and 15.3 for Hispanic women.²⁸

Prominent racial and ethnic differences are also seen in the prevalence of other CVD risk factors. For example, although overweight and obesity are common in US adults, with a 2011 to 2012 prevalence approaching 70% in adult men, there is a greater prevalence in non-Hispanic whites (73%) and Hispanics (80%) than in non-Hispanic blacks (69%).²⁸ On the other hand, in women, the prevalence of obesity and overweight is much higher in non-Hispanic blacks (82%) than in Hispanics (76%) or non-Hispanic whites.²⁸ Among children and adolescents aged 2 to 19 years, the prevalence of overweight and obesity in the 2011 to 2012 National Health and Nutrition Examination Survey (NHANES) was 31.8%.²⁸ In general, among all children aged 2 to 19 years, the prevalence of overweight and obesity tends to be higher in non-Hispanic black, non-Hispanic white, and Hispanic children than in non-Hispanic Asian children.²⁸ Additionally, within the Hispanic population, important differences in the prevalence of overweight have been demonstrated. For example, Mexican NHANES participants have a higher mean age-adjusted prevalence of being overweight compared with Puerto Ricans and Cubans.³⁰

The 2011 to 2012 NHANES data have also shown that dyslipidemia, assessed by a total cholesterol of 240 mg/dL or greater, was more prevalent in Hispanic adults (14.2% overall) and non-Hispanic white adults (13.5% overall) than in non-Hispanic black adults (9.8% overall).²⁸ Age-adjusted prevalence estimates from the 2014 National Health Interview Survey (NHIS) in adults demonstrated a greater prevalence of physical inactivity among Hispanics (40.1%) and non-Hispanic blacks (38.3%) than in non-Hispanic whites (26.3%). Finally, the prevalence of current smoking or tobacco use varies significantly by race and ethnicity. Data from the 2014 NHIS in adults aged 18 years and older showed a greater prevalence of current smoking in non-Hispanic black men (21.4%), non-Hispanic white men (19.9%), and Alaska Native men (18.6%) than in Asian men (13.8%) and Hispanic men (13.8%).²⁸ Similarly, non-Hispanic black women (13.4%), non-Hispanic white women (18.3%), and American Indian or Alaska Native women (21.6%) had a greater prevalence of current smoking than Asian women (5.5%) and Hispanic women (7.4%).²⁸

Racial and ethnic differences have also been demonstrated for estimated heart age, the predicted age of a person's vascular system based on their cardiovascular risk factor profile.³¹ Using data from Behavioral Risk Factor Surveillance System applied to the sex-specific Framingham risk score models, Yang et al³¹ compared heart age to chronological age and demonstrated that, on average, predicted heart ages for adult men and women were 7.8 and 5.4 years older than their chronological ages, respectively. Importantly, the prevalence of excess heart age of 5 years or greater (measured as the difference between heart age and chronological age) was greater in non-Hispanic black men (58.7 years) and women (58.9 years) than in other racial/ethnic groups, including non-Hispanic white men (55.3 years) and women (52.5 years).³¹

Persisting Disparities in Cardiovascular Diseases and Conditions

The American Heart Association's Heart Disease and Stroke Statistics provides a comprehensive annual update that includes data on race/ethnicity and health behaviors, risk factors, and specific cardiovascular conditions and diseases as well as their clinical and public health outcomes.²⁸ For example, the 2016 report shows that age-adjusted prevalence estimates for coronary heart disease (CHD) from the 2014 NHIS data are higher in native Hawaiians/Pacific Islanders (6.9%), non-Hispanic whites (5.6%), and non-Hispanic blacks (5.6%) than in Asians (3.3%) and Hispanics (4.9%).²⁸ The reported age-adjusted prevalence for American Indians and Alaska Natives of 6.0% is known to be statistically unreliable. The age-adjusted CHD mortality per 100,000 population in men were higher in non-Hispanic blacks (155.1) and non-Hispanic whites (141.8) than in Hispanics (104.7); the corresponding rates in women were also higher in non-Hispanic blacks (94.7) and non-Hispanic whites (75) than in Hispanics (61.3).²⁸

There are racial/ethnic disparities in the risk for developing congestive heart failure (CHF).^28,32 In the Multi-Ethnic Study of Atherosclerosis (MESA), a cohort study of 6814 participants of four ethnicities—white (38.5%), African American (27.8%), Hispanic (21.9%), and Chinese American (11.8%)—African Americans had the highest incidence rate of CHF, followed by Hispanic, white, and Chinese American participants (incidence rates: 4.6, 3.5, 2.4, and 1.0 per 1000 person-years, respectively).³² In addition, African Americans had the highest proportion of incident CHF not preceded by clinical myocardial infarction (75%) compared with other ethnic groups.³² Importantly, however, differences in the prevalence of hypertension, diabetes mellitus, and socioeconomic factors explained the higher risk of incident CHF among African Americans.³² Data from the Atherosclerosis Risk in Communities (ARIC) cohort showed that incident heart failure per 1000 person-years was higher in African American men (9.1) and women (8.1) than in white men (6.0) and women (3.4); however, these racial differences in heart failure incidence were largely explained by risk factor differences.³³ In the ARIC cohort, heart failure case fatality rates 5 years after an incident heart failure hospitalization were significantly higher in African American men (51.8%) and women (46.1%) than in white men (41.2%) and women (35.8%).³³

Additional information on racial and/or ethnic differences in prevalence and mortality of total CVD; prevalence, mortality, and distribution of new and recurrent strokes; mortality of myocardial infarction; and the distribution of physician-diagnosed and undiagnosed diabetes and related mortality are presented for adult men and women in Tables 109–5 and 109–6, respectively. Similarly, the race/ethnic distribution of congenital cardiovascular defects and the distribution of major cardiovascular risk factors in children and youth are presented in Table 109–7. Figure 109–4 shows additional data on racial and ethnic variations in the prevalence of congenital cardiovascular malformations.³⁴

In contrast to the excess CVD mortality observed in non-Hispanic blacks in the general US population, Kovesdy et al³⁵ reported that, in a large cohort of more than 3 million contemporary US veterans with an estimated glomerular filtration rate ≥ 60 mL/min/1.73 m², black veterans had lower all-cause mortality, lower incidence of CHD, and a similar incidence of ischemic stroke (Fig. 109–5). These data were derived from multivariable-adjusted Cox models in a nationwide cohort of 547,441 black and more than 2.5 million white veterans receiving care from the US Veterans Health Administration (VHA).³⁵ The findings also do not apply to women because the cohort consisted predominantly of men. Compared with the observations in the general US population, the findings could not be explained on the basis of differences in demographic factors, comorbidity, or socioeconomic characteristics; however, the open access to healthcare and the provision of comprehensive healthcare based on a veteran status within the US VHA may have contributed.³⁵ Kuller and Neaton³⁶ and the VHA study authors and others^35,37 caution that the findings may also be attributable to selection for military service and the subsequent use of VHA services among blacks and whites, and thus, the findings may not be generalizable to the overall US population.

Many factors operating at multiple levels account for the racial and ethnic differences in the distribution of health behaviors, risk factors, disease burden, and associated morbidity and mortality. These factors operate at the levels of individual patients, healthcare providers, health systems, health policies, and the communities within which patients live.

Patient-Level Factors and Social Determinants of Health

Some of the most often cited patient-level factors are behavioral and lifestyle choices, cultural beliefs and practices, patient preferences, mistrust, and nonadherence to health provider recommendations for healthful behaviors and practices. However, these factors do not exist in isolation and are often influenced by the social determinants of health or, as defined by the World Health Organization, “the circumstances in which people are born, grow, live, work, and age, and the systems put in place to deal with illness."³⁸ Important within this category are conditions of residential or neighborhood environment and policies at the local, state, and national levels that also impact health and the provision of health-promoting services and resources.

A scientific statement from the American Heart Association addressed socioeconomic position—a broader term that encompasses wealth and income, education, employment/occupational status, and other factors. The statement also addressed the “psychological, behavioral, and biological mechanisms through which social determinants precipitate and perpetuate CVD."³⁹ All of these factors also contribute to the racial and ethnic disparities discussed in this chapter.

Health System Factors and Disparities in Cardiovascular Disease

In addition to these social determinants, there are important health system factors such as access to healthcare in general and availability of high-quality healthcare that play important roles in disparities. Within the health system are also myriad financial, organizational, cultural, and linguistic barriers.³⁹ The amount of contact patients have with the healthcare system, access to specialty care and referral, and availability of hospitals and healthcare facilities with advanced technologies in diagnostic and therapeutic interventions represent health systems factors that may affect disparate CVD outcomes.^40,41 For example, in an assessment of hospital emergency department crowding as measured by ambulance diversion, Shen and Hsia⁴¹ demonstrated that hospitals serving a high volume of black patients spent more hours in diversion status compared with other hospitals and that patients who experienced the highest level of diversion had a low probability of admission to hospitals with capabilities for cardiac catheterization, cardiac intensive care, and coronary artery bypass surgery.⁴¹ Most importantly, patients who faced diversion had a 9.6% higher mortality at 1 year.⁴¹ Independent of the health system factors are individual provider-level factors that also contribute to racial and ethnic differences in cardiovascular health.

Provider-Level Factors and Racial and Ethnic Disparities

Individual provider-level factors are also important in racial and ethnic disparities in CVD. In a landmark report on confronting unequal treatment and related racial and ethnic disparities in health, the Institute of Medicine found that “bias, stereotyping, prejudice, and clinical uncertainty on the part of healthcare providers may contribute to racial and ethnic disparities in healthcare” and may be important factors in the persisting racial and ethnic disparities in health and health care in the United States.⁴² Although the Institute of Medicine found no direct evidence that racism, bias, or prejudice among healthcare professionals affects the quality of care for racial and ethnic minority patients, it concluded that the evidence was strong, albeit circumstantial, and urged further research in this arena.⁴²

In fact, several studies have demonstrated that healthcare providers’ diagnostic, treatment, and referral decisions are influenced by the race/ethnicity of patients and that these influences are complex and may be modified by both patient and provider sex.⁴² For example, Schulman et al¹⁶ developed a computerized survey instrument in which actors portrayed patients (Fig. 109–6) with particular characteristics in scripted interviews about their symptoms in order to assess physicians’ recommendations for cardiac catheterization in the evaluation of chest pain. Their findings suggested that the race and sex of a patient independently influenced how physicians managed chest pain. Although the study found no difference in the referral rate for black men and white men, analysis of race-sex interactions revealed that black women were approximately 60% less likely to be referred for cardiac catheterization than white men.¹⁶ In another study assessing implicit bias, Green et al⁴³ also demonstrated that healthcare providers’ unconscious biases may contribute to racial/ethnic disparities in use of clinical interventions such as thrombolytic therapy for acute myocardial infarction.

A recent American Heart Association scientific statement shows that although clinicians demonstrate little explicit or intentional bias, there is evidence to suggest that they exhibit substantial bias in their implicit (unconscious) attitudes and the statement presents three ways in which bias may operate in this setting.⁴⁴ The study by Schulman et al¹⁶ is cited as an example of how providers’ implicit bias, manifesting as stereotypical assumptions, may influence clinicians’ decisions regarding referral for diagnostic or therapeutic interventions leading to substandard quality of care. Two other ways in which implicit bias may operate include (1) the generation of lower quality clinical interactions and communication between (more biased) clinicians and racial and ethnic minority patients and (2) stereotype threat wherein “individuals, often unconsciously, fear being judged negatively according to racial stereotypes."⁴⁴ The scientific statement calls attention to the need for developing effective interventions to address these phenomena and help improve patient-provider communication as well as patient satisfaction and trust in an effort to improve the delivery of quality healthcare to racial and ethnic minority populations.

In an effort to help advance conceptual thinking and progress in health disparities research within the healthcare system, Kilbourne et al⁴⁵ proposed a framework that organizes the research agenda in three sequential phases: detection, understanding, and reduction or elimination of health disparities. Although work in none of these phases is complete, substantial research investments have been made in the detection, documentation, and follow-up (phase 1) of racial and ethnic disparities in cardiovascular health, especially regarding non-Hispanic black and white persons in the United States.^22,23,28 Further research is needed in American Indians, Alaska Natives, and other race/ethnic groups in whom inadequate data preclude reliable detection of racial and ethnic disparities. In addition, more rigorous definitions of what constitute vulnerable populations within the concept of racial and ethnic disparities are necessary. Methodologic challenges regarding confounders and residual confounding in analyses of racial and ethnic disparities need further work.

Second, further research to clarify the underlying mechanisms of racial and ethnic disparities and improve our understanding (phase 2) of these disparities is necessary. In particular, a deeper understanding of the roles played by stereotyping, provider uncertainty, and implicit bias is needed. Similarly, further research to elucidate other provider- and patient-level factors and how they interact in the clinical and public health settings is needed. Additionally, the role of health systems and healthcare policies and how they impact racial and ethnic disparities need further work. Novel strategies are needed to help disentangle the complexities and interrelationships in the roles played by individual patient behaviors and lifestyles, ancestry, social and environmental determinants, provider and health systems factors, and policies in the health and nonhealth sectors in the real world.

Of all phases of the research framework, it is phase 3 where further research is most sorely needed. With rare exceptions, we lack examples with compelling evidence of sustained elimination of racial and ethnic disparities in health. For example, in an effort to assess the effectiveness of quality improvement strategies in reducing disparities in health and healthcare, McPheeters et al⁴⁶ identified papers published between 1983 and 2011 that reported metrics enabling the impact on a health disparity to be measured. The review focused on disparities in several clinical conditions including heart failure, hypertension, coronary artery disease, and diabetes. Although most studies focused on racial or ethnic disparities, with some targeted interventions demonstrating greater effect in racial minorities, the report concluded that it was unclear whether the interventions were effective at reducing disparities.⁴⁶

Heart failure is an important contributor to health disparities, with blacks suffering a disproportionate burden of disease. Overall, few potentially eligible patients with heart failure with reduced ejection fraction are treated with combination hydralazine plus isosorbide dinitrate (H-ISDN), and among African Americans, less than one-quarter of eligible patients receive guideline-recommended H-ISDN therapy. Yet strong evidence for the H-ISDN recommendation included the landmark African American Heart Failure Trial (A-HeFT), a randomized, placebo-controlled trial that demonstrated 43% and 39% reductions in mortality and heart failure hospitalizations, respectively, in patients receiving guideline therapy plus H-ISDN compared to patients receiving guideline therapy alone. The study population was restricted to self-identified African Americans, which led to a race-based label for the combination pill, BiDil, used in A-HeFT.⁴⁷ African Americans in whom heart failure occurs at an earlier age, with structural and functional changes that predate symptoms by a decade, are disproportionately affected by heart failure. Slow adoption of potentially life-saving H-ISDN therapy is problematic not only for high-risk African Americans, but also for eligible patients who may benefit from H-ISDN and is an example of a race construct that undermines more appropriate biologic and physiologic correlates. Chap. 70 addresses treatment of heart failure.

In an assessment of interventions targeted toward reducing disparities in hypertension control, 39 interventions and several state and national policy initiatives that targeted the reduction of racial and ethnic disparities in blood pressure control were identified. Of the six studies that examined disparities, three reduced and two increased disparities, whereas one had no effect on disparities.⁴⁸ Thus, although several effective interventions exist to improve blood pressure in racial and ethnic minorities, the evidence that they reduce disparities is limited, and many race/ethnic groups remain understudied.⁴⁸ The review concluded with a call for “rigorous evaluation of pragmatic, sustainable, multilevel interventions; institutional support for training implementation researchers, and creating broad partnerships among payers, patients, providers, researchers, policymakers, and community-based organizations” in order to strengthen the evidence and translate it into practice and policy.⁴⁸ Fully embracing implementation research as an important part of the health disparities research agenda will go a long way in identifying effective strategies for reducing and eliminating racial and ethnic disparities in cardiovascular health.^49,50

Race and ethnicity are sociocultural and political constructs that are flawed when used as proxy for genetic or other biological phenomena. As such, racial and ethnic differences observed in the distribution or outcomes of CVDs should not be construed as necessarily reflecting genetic or other biological differences. When properly collected, analyzed, and interpreted, epidemiologic data grouped by race and ethnicity can be invaluable in the quest to eliminate related disparities in cardiovascular health and advance health equity. In the United States, the federal government standards for data collection contain five minimum categories for race and two categories for ethnicity. Using these race/ethnic categories, we have summarized the state of racial and ethnic disparities in CVDs in the United States and shared examples of successful reduction or elimination of disparities and instances showing persistence or widening of disparities. Cardiovascular epidemiologic cohorts of ethnic minority populations represent an important past and future resource for elucidating outcomes in specific minority groups and health disparities. A framework to advance conceptual thinking and progress in health disparities research within the healthcare system was discussed. The need for novel strategies to help disentangle the complexities of racial and ethnic disparities and cardiovascular health outcomes was emphasized. Additionally, advancing implementation research is essential in the quest to reduce and eliminate racial and ethnic disparities in cardiovascular health and will require researchers skilled in implementation science.