Where we live: The impact of neighborhoods and community factors on cardiovascular health in the United States

Abstract

While the prevalence of cardiovascular risk factors has decreased in the United States in recent years, cardiovascular disparities by sex and race persist. Among the factors contributing to these disparities is the physical environment in which individuals live. Neighborhood characteristics, ranging from air pollution exposure to residential segregation, have been found to be related to cardiovascular health (CVH) and stroke risk. Through the use of cross‐sectional, longitudinal, and analytic regression modeling, we are gaining clarity about the relationship between an individual's external environment and CVH. Moreover, differences in CVH vary by sex and/or race within the same neighborhood. The mechanism by which these disparities exist is still being explored. In this review, we examine the literature that has accumulated regarding how external environments and community factors affect individuals and populations by race and sex.

Abbreviations

AAC

abdominal aortic calcification

AHA

American Heart Association

BMI

body mass index

CAC

coronary artery calcium

CBR

cumulative biological risk

CIMT

carotid IMT

CT

computed tomography

CVD

cardiovascular disease

CVH

cardiovascular health

GIS

geographic information systems

IMT

intima‐media thickness

JHS

Jackson Heart Study

MESA

Multi‐Ethnic Study of Atherosclerosis

NO_X

nitrogen oxides

PAD

peripheral artery disease

PM_2.5

particulate matter with diameter < 2.5 μm

SES

socioeconomic status

1. INTRODUCTION

For decades, medical researchers have devoted themselves to documenting, explaining, and offering ideas for mitigating the cardiovascular and stroke disparities associated with race, sex, and socioeconomic status (SES). Yet, despite these investigations, such health disparities persist in the United States.1

One review posits that disparities persist in part because of a misplaced focus of health disparity research in cardiology.2 Specifically, the epidemiology of cardiovascular disease (CVD) has traditionally been focused on individual‐level risk factors arising from behavior and biology. Consequently, CVD prevention has centered on individual choice and appropriate medical care, resulting in intensive, individual counseling and therapy as a means to prevent and treat these risk factors and diseases instead of understanding the role of environment and social context in which these diseases develop. A supplementary approach might modify the environment to create the necessary conditions for large‐scale, more effective change.2 Residential environments with their physical and chemical exposures as well as structural entities (ie, sidewalks, traffic flow, greenery, etc.) can shift a distribution of risk factors within a population (eg, lowering the mean blood pressure in a distribution) as opposed to intervening in high‐risk individuals (eg, targeting individuals with hypertension). The incremental and cumulative effects of environmental modification may result in longer‐lasting and significant change at a individual level. The role of the environment and the “sociological determinants of health,” or factors of living which influence an individual's health, should not be underappreciated.2

In 2015, the American Heart Association (AHA) drew attention to the role of the environment in influencing CVD in a scientific statement.3 The statement notes that despite a linear decrease in mortality because of CVD in the United States, the prevalence of CVD is expected to rise 10% from 2010 to 2030 as a result of an aging population, with a concurrent rise in obesity, hypertension, diabetes mellitus, and physical inactivity over the past quarter century. This concomitant rise suggests that changes in societal and environmental conditions drive this increase. The statement considers how social determinants of health, such as SES, race/ethnicity, social support, culture and language, access to care, and residential environment contribute to the development of CVD through psychological, behavioral, and biological mechanisms. In discussing the interplay of these factors, the AHA aims to raise awareness of the social factors affecting incidence, treatment, and outcomes in CVD and how death and disability could be reduced.3

This review examines the complex relationship between individuals, their environments, and CVD by examining the findings of two ongoing, large‐scale longitudinal studies. Using sophisticated regression analysis techniques, various teams have drawn data from these ongoing prospective studies to gain insights into how neighborhood characteristics influence individual health, and how individuals' health varies based on characteristics, such as race/ethnicity, sex, and SES even within the same neighborhood.

2. BACKGROUND: LONGITUDINAL STUDIES

2.1. The multi‐ethnic study of atherosclerosis

The multi‐ethnic study of atherosclerosis (MESA) is a prospective, longitudinal study of men and women in Los Angeles County, California; St. Paul, Minnesota; Chicago, Illinois; Forsyth County, North Carolina; Baltimore, Maryland; and New York, New York. Participants were aged 45 to 84 years and free of clinical CVD at the time of baseline measurements, between August 2000 and July 2002.4 Here, we highlight recent studies that draw from this ongoing longitudinal study.

2.1.1. Hypertension and neighborhood characteristics

One study analyzed 3382 MESA participants, to see how measures of neighborhood physical and social environments were related to hypertension over a median follow‐up time of 7.2 years, during which there were 1335 incident cases of hypertension.5 This population sample's average age was 59.1 years, with a racial composition of: 46.3% white; 22.5% non‐white Hispanic; 20.6% black; and 13.4% Chinese American. Neighborhood physical data (eg, density of favorable food stores, physical activity environment) and social data (eg, social cohesion, safety) were collected by geographic information systems (GIS) and sequential participant surveys. After adjusting for individual characteristics (ie, age, sex, income, race, etc.), survey‐based healthy food availability and walking environment were found to be predictors of hypertension risk, whereas GIS‐based food environment and physical activity environment scores were not. After accounting for all neighborhood variables and adjusting for individual characteristics, only survey‐based healthy food scores continued to be significantly associated with hypertension, with each standard‐deviation increment in healthy food availability, correlating with a 12% lower rate of hypertension.5

2.1.2. Hypertension and neighborhood stressors

A second study performed cross‐sectional analysis of MESA data to determine whether individual and neighborhood‐level chronic stressors contribute to disparities in hypertension prevalence.6 The study's authors postulated that neighborhood‐level stressors, such as safety and social cohesion, might correlate with racial and ethnic differences in hypertension prevalence. The prevalence of hypertension within the population being analyzed was 59.5% for African‐Americans, 43.9% for Hispanics, and 42.0% for whites. Chronic stressors affected whites the least, African‐Americans moderately, and Hispanics the most. However, African‐Americans reported more perceived major and everyday discrimination vs whites, whereas Hispanics reported levels of major and everyday discrimination similar to whites. Indeed, after adjusting for age and sex, higher levels of perceived major discrimination and neighborhood‐level stressors were significantly and positively associated with hypertension. Age and sex‐adjusted relative prevalence of hypertension compared to whites was 1.30 for African‐Americans and 1.16 for Hispanics. After adjustment for chronic neighborhood stressors, these relative frequencies were reduced to 1.17 for African‐Americans and to 1.09 for Hispanics, suggesting that neighborhood chronic stressors have a causal role in racial and ethnic differences in hypertension prevalence.6

2.1.3. Carotid artery health and neighborhood air pollution

In two separate MESA data analyses, researchers examined air pollution exposures, defined as fine particulate matter with diameter < 2.5 μm (PM_2.5) and nitrogen oxides (NO_X) (the sum of nitric oxide, nitrogen dioxide, nitrous acid, and nitric acid) at the six MESA sites.

The first of these analyses sought to measure relationships between air pollution, race/ethnicity, and residential segregation (5921 participants—40% white; 28% black; 21% Hispanic; 11% Chinese American).7 Independent of race/ethnicity, living in majority white (> 60%) neighborhoods was associated with 5% lower PM_2.5 and 18% lower NO_X exposure than neighborhoods with <25% white population. Living in majority Hispanic (> 60%) neighborhoods was associated with 8% higher PM_2.5 and 31% higher NO_X exposures than that in neighborhoods with <25% Hispanic population.

The second analysis measured the relationship between ambient air pollution, race/ethnicity, and carotid intima‐media thickness (CIMT) (6347 participants—Chinese, Hispanic, Caucasian American, and African‐American).8 Smaller CIMT values indicate lower cardiovascular health. PM_2.5 exposure was found to be inversely correlated with CIMT (ie, more particulates, narrower artery).8 Both PM_2.5 and NO_X exposure varied by race/ethnicity—in descending order: Chinese, Hispanic, African‐Americans, lowest for Caucasian Americans. Adjusting for differing pollution exposure, the association between Chinese ethnicity and lower CIMT compared to Caucasian Americans was even stronger. PM_2.5 concentrations were not found to be associated with CIMT differences in African‐Americans and Hispanics compared with Caucasian Americans. Increased exposure to NO_X did not increase differences in CIMT by race/ethnicity.

2.1.4. Coronary artery calcium and neighborhood characteristics

Relationships between coronary artery calcium (CAC) and social and physical neighborhood characteristics have also been analyzed (5950 MESA participants; 12 years of follow‐up).9 Mean baseline CAC was lower in neighborhoods with a higher density of healthy food stores and higher availability of healthy food. Lower density of healthy food stores, lower availability of healthy food, and fewer recreational facilities were associated with the fastest increase in mean annual CAC. Increases in healthy food store density were associated with individual decreases in CAC. Increases in neighborhood healthy food stores were associated with a drop of 19.99 Agatston units of CAC for each SD increase in neighborhood exposure. The association of healthy food density with CAC was stronger in women than in men, but the effect was not modified by race or ethnicity. These results suggest that food context is one mechanism by which neighborhood factors affect the incidence of coronary heart disease, and that this association is stronger for women than for men.

2.1.5. Cardiovascular health and neighborhood characteristics

The relationship between overall cardiovascular health (CVH) and neighborhood environment was explored in a cross‐sectional analysis of 5805 MESA participants (52% female; mean age 61.7 years). CVH was scored by physical activity, diet, blood glucose, blood pressure, body mass index (BMI), cholesterol, and smoking status. Within this sample, 21.1% of participants had “ideal” CVH scores, 33.2% had “intermediate” CVH scores, and 45.8% had “poor” CVH scores.

After adjusting for individual demographic characteristics and neighborhood SES, all considered neighborhood characteristics except social cohesion were related to CVH. Ideal CVH was significantly associated with favorable food stores, physical activity resources, and walking/physical activity environment. Specifically, for every 1‐SD increase in favorable food stores, participants were 22% more likely to have ideal CVH rather than poor CVH. Similar increased likelihoods were found in neighborhoods that offered physical activity resources (19%), or a walking/physical activity environment (20%).

There were also significant differences between men and women. Among men, the odds of having ideal CVH were 36% higher for each 1‐SD increase in favorable food stores, but among women there was no association between ideal CVH score and favorable food stores. Another difference was that among women the odds of ideal CVH were 41% greater with increased neighborhood SES, and yet there was no such association among men. Both sexes had significantly higher odds of having ideal CVH when greater physical activity resources were available. Race/ethnicity and neighborhood SES did not affect the association between neighborhood characteristics and CVH. This is the first study to report associations between neighborhood characteristics and overall CVH; however, it leaves unclear the reasons for the differences found between the sexes.10

2.1.6. Cardiovascular disease and neighborhood racial/ethnic segregation

Although housing discrimination on the basis of race/ethnicity has been outlawed in the United States for more than 50 years, segregation persists—the average white American lives a neighborhood that is 75% white, whereas the average black or Hispanic individual (approximately 13% and 16% of the population, respectively) lives in a neighborhood that is only 35% white.

The relationship between CVD and neighborhood racial/ethnic segregation has been examined in a group of 5227 MESA participants (2345 non‐Hispanic white, 1595 non‐Hispanic black, and 1289 Hispanic).11 For blacks, higher neighborhood segregation scores were significantly associated with increased CVD incidence. Specifically, each unit increase in black segregation was associated with a 12% increase in CVD hazard, even after adjusting for age, sex, and study site. This association did not diminish after further adjustments for individual SES, physical neighborhood characteristics, social environment, or traditional CVD risk factors. By contrast, in whites, higher segregation scores were associated with lower CVD incidence. This relationship lost statistical significance after adjusting for neighborhood characteristics, suggesting that neighborhood characteristics may in part explain why segregation affects cardiovascular risk. There was no evidence for a link between segregation and CVD incidence in Hispanics.11

2.2. The Jackson heart study

The Jackson heart study (JHS) is an ongoing population‐based, longitudinal cohort study of African‐Americans residing in the three counties (Hinds, Madison, and Rankin) that make up the Jackson, Mississippi, metropolitan statistical area. Data collection began in 2000. A previous review gave a 10‐year perspective on JHS's accomplishments and summarized results that were available by the time of its publication in 2011.12 Below are some additional results from studies published after 2011.

2.2.1. CVD incidence, neighborhood disadvantage, and social environment

The cross‐sectional association between CVD incidence (stroke/coronary heart disease) and neighborhood disadvantage and social environment, was examined in 4096 African‐American JHS enrolees.13 “Neighborhood disadvantage” included eight indicators from the 2000 U.S. Census, and “social environment” included measures of social cohesion, violence, and disorder. Over a median follow‐up time of 8.38 years, 232 CVD events were recorded.

Following adjustment for individual sociodemographic characteristics, African‐American women living in neighborhoods with higher levels of disadvantage, violence, and disorder were found to have a higher risk for CVD. Each 1‐SD increase in levels of disadvantage, violence, and disorder increased CVD risk by 25%, 13%, and 20%, respectively. Social cohesion was not significantly associated with CVD in women. Even after accounting for traditional biological and behavioral risk factors, these associations remained statistically significant. Surprisingly, neither social cohesion nor greater levels of violence and disorder were significantly associated with risk of CVD in men.13

2.2.2. Social cohesion may moderate link between neighborhood disadvantage and health risk

Social cohesion has been found to moderate the association between neighborhood disadvantage and cumulative biological risk (CBR). In a study of 4408 JHS participants (mean age of 54.5 years; 2799 women; 1609 men),14 social cohesion was classified as being “high” or “low,” with CBR representing eight biomarkers in four physiological systems (metabolic, cardiovascular, neuroendocrine, and inflammatory). Sociodemographic factors were controlled for and health behaviors (alcohol consumption, physical activity, cigarette smoking, and percentage of calories from dietary fat) were included in this analysis. The study's analysis found that the association between neighborhood disadvantage and CBR was strongest for men living in neighborhoods with low social cohesion—they had the highest CBR scores. Interestingly, men living in more affluent neighborhoods with higher levels of social cohesion had higher CBR scores than those living in affluent neighborhoods with low levels of social cohesion. This suggests that low social cohesion may amplify the effect of neighborhood disadvantage on CBR, but that high social cohesion does not necessarily insulate from how neighborhood advantage lowers CBR. For women, this study found no evidence that social cohesion modified the association between neighborhood disadvantage and CBR. Of the eight CBR biomarkers, only cortisol was significant for influencing CBR, possibly through a stress‐mediated pathway.14

A separate study found that family income and educational attainment did not affect the association between neighborhood disadvantage and CBR. Living in disadvantaged neighborhoods was associated with higher levels of CBR, after adjustment for sociodemographic characteristics (eg, family income, educational attainment). Adjusting for health behaviors attenuated this association, suggesting that behavioral risk factors may be one pathway through which neighborhood disadvantage influences cumulative risk of disease.15

2.2.3. Patterns of atherosclerosis

To investigate whether patterns of atherosclerosis in predominantly white, urban populations was generalizable to black, rural, populations, one study looked at CIMT, peripheral artery disease (PAD), coronary artery calcification (CAC), and abdominal aortic calcification (AAC) in 4800 JHS participants. Living fewer than 150 m from major roads increased CIMT by 6.67% after adjustment for confounding variables. Similarly, the prevalence of PAD was 1.17 times greater for individuals living fewer than 150 m from major roads compared to those living more than 300 m from such roads, although this increase was statistically insignificant. The prevalence of CAC and AAC was not elevated in individuals living less than 150 m from major roads.16

2.2.4. Abdominal adiposity and neighborhood safety

In a study of 2881 JHS participants who underwent computed tomography (CT) abdominal scanning, Pham do et al. investigated the association between abdominal fat distribution, BMI and waist circumference, and perceived neighborhood safety. Among men, the investigators found no association between perceived neighborhood safety and any of the abdominal fat distribution or anthropometric measures. Among premenopausal women, however, those who considered their neighborhood most unsafe had significantly higher median visceral fat volume, total abdominal fat, and higher BMI and waist circumference than women who strongly agreed that their neighborhood was safe.17

2.2.5. Sex and Cardiometabolic risk factors

Clark et al. explored the relationship of neighborhood disadvantage and neighborhood safety with cardiometabolic risk factors in 3909 JHS participants.18 These risk factors included metabolic syndrome, the constituents of metabolic syndrome (serum triglycerides, fasting plasma glucose, blood pressure, waist circumference, and high‐density lipoprotein cholesterol), insulin resistance, and cardiovascular inflammation. The socioeconomic disadvantage index they used included percentage of adults below poverty in the neighborhood, unemployment rate, percentage of college‐educated residents in the neighborhood, and percentage of households not owning a vehicle. Neighborhood safety was assessed by participant perception. Women in the most disadvantaged neighborhoods had significantly higher prevalence of metabolic syndrome and elevated fasting glucose levels compared to women in the most advantaged neighborhoods. The association between neighborhood socioeconomic disadvantage and metabolic syndrome among women remained unchanged even after adjusting for perceived neighborhood safety, age, and health behaviors. Among women, perceived neighborhood safety was associated with elevated waist circumference and elevated fasting glucose, but not metabolic syndrome. Men in the most disadvantaged neighborhoods had a higher prevalence of elevated fasting glucose and lower high‐density lipoprotein cholesterol than men in the most advantaged neighborhoods. Men who perceived their neighborhoods as unsafe also were more likely to have elevated fasting glucose and elevated insulin resistance.

3. DISCUSSION

Efforts to prevent and treat CVD have largely been focused on identifying high‐risk individuals, and then using specific behavioral and medical intervention to prevent further progression of disease. This has resulted in extensive research being targeted toward specific races/ethnicities that is, because of biologic or behavioral characteristics, are perceived as being at higher risk for developing CVD. However, in the past decade, there has been a shift toward trying to understanding CVD in the context of an individual's environment. Neighborhoods, through their physical structure, resources, and psychosocial atmosphere, can dramatically affect their inhabitants' CVH. However, this effect may be different based on sex and/or race/ethnicity.

We are gaining greater clarity about how physical characteristics of a neighborhood, such as its food environment and physical activity resources affect levels of CVD. A higher density of healthy food stores and higher availability of healthy food were associated with lower mean baseline CAC, while lower density and availability of healthy food stores and fewer recreational facilities were associated with faster progression of CAC.9 More broadly, the presence of favorable foods in a neighborhood was associated significantly with ideal CVH, even after adjusting for SES.10 Physical activity resources and walking/physical activity environment also significantly affected CVH.10 That an increase in healthy food store density was associated with decreases in CAC within individuals suggests that a healthy food environment, in particular, along with adequate physical activity resources, can favorably modify CVH.9

However, although a neighborhood's food environment and physical activity resources are important, we cannot ignore the role of the individual—whether they are engaging in unhealthy behaviors or these healthy aspects of the environment. In a different study investigating how neighborhood physical and social environments relate to hypertension, it was found that, after adjusting for individual characteristics, survey‐based healthy food availability and walking environment were better predictors of hypertension risk than GIS‐based food environment and physical activity environments. This finding suggests that individual perception and subsequent utilization of such resources is essential to CVH.5 Individuals living in disadvantaged neighborhoods, as defined by socioeconomic indicators from the 2000 U.S. Census, are associated with higher CBR after adjustment for sociodemographic characteristics. However, adjustment for health behaviors (ie, diet, cigarette smoking, alcohol consumption, etc.) attenuated this risk, a finding that highlights how behavioral risk factors influence health.15

Beyond providing an environment and resources necessary for good health, neighborhoods also represent a complex and dynamic community that can promote or hinder individuals’ CVD health. The prevalence of hypertension, for example, varies based on race and is significantly associated with perceived major discrimination and neighborhood‐level stressors, particularly for African‐Americans and Hispanics.6 CBR has also been shown to be affected by social cohesion, with lower social cohesion increasing the effect of neighborhood disadvantage on CBR, especially for men; this relationship was not significant for women. The proposed mechanism for stress‐induced disease lies in the hypothalamic‐pituitary axis, producing chronically elevated levels of cortisol,14 leading to poor health outcomes that result from negative factors within the community.

Interestingly, these studies have found that the environment can have different effects on men and women who share the same environment. Health food density more strongly affected women's CAC compared to men's, although ideal CVH score was not associated with favorable food stores among women.9, 10 Instead, for women, odds of ideal CVH grew with increased neighborhood SES, whereas men's grew with increased favorable food stores.10 The differential responses to environment by sex are particularly striking when considering the effect of various social stimuli. Low social cohesion, for instance, mediated the relationship between neighborhood disadvantage and CBR in men, but not women.14 Conversely, neighborhood violence and disorder increased CVD risk in women, but non‐significantly lowered risk of CVD in men.13 Similarly, perceived neighborhood safety was not associated with abdominal fat distribution for men, yet premenopausal women who considered their neighborhoods unsafe had higher visceral/abdominal fat, higher BMI, and higher waist circumference.17 Metabolically, men and women responded differently in disadvantaged neighborhoods, with women having elevated fasting glucose and waist circumference and men having elevated fasting glucose and insulin resistance.18 How a neighborhood can have different effects on a man vs a woman have been attributed to the ways the sexes interact with their environment. For example, women may spend more time in a neighborhood and thus be “more exposed” to the negative factors in that environment. Chronically unemployed African‐American men may in fact be even “more exposed” than women to a negative neighborhood. Moreover, men and women may cope with poor environments differently.14

Race/ethnicity may also have a role to play in the incidence of CVD, but it may be secondary to the composition of the neighborhood in which people dwell. Jones et al. found that, independent of the individual race/ethnicity of the residents, residential segregation was significantly associated with varying amounts of exposure to air pollution.7 People living in majority white neighborhoods had lower rates of exposure to air pollution, defined as PM_2.5 and NO_X, while those living in majority Hispanic neighborhoods had higher levels of exposure.7 A separate study correlated air pollution exposure with CIMT and PAD.16 Moreover, higher segregation resulted in varying CVD outcomes by race. In blacks, higher neighborhood segregation was significantly associated with higher CVD incidence even after adjustment for individual demographics, SES, traditional CVD risk factors, and neighborhood characteristics. However, in whites, higher segregation was associated with lower CVD incidence, but this relationship was attenuated when neighborhood characteristics were taken into account.11 These studies suggest that the racial composition of neighborhoods on CVD incidence is significant. Kershaw et al. proposed poverty and segregation as driving factors in health disparities, because both limit individuals' access to adequate healthcare.11

The advantages of examining how neighborhood environments affect their inhabitants through prospective, longitudinal trials like MESA and JHS are evident: large sample sizes combined with years of follow‐up yield robust data that researchers can interpret. However, like all observational studies, the results are not without their weaknesses. Because of the lack of randomized, controlled trials, true causality is impossible to determine. Nonetheless, the associations drawn from these studies are strong and reliable. The specific and sophisticated statistical techniques used in investigating various characteristics and relationships within individuals and across their communities can help to strip away confounding variables, leaving only the crux of associations. While the exact mechanisms by which neighborhood environments are affect their inhabitants’ CVH are still being explored, we have relative confidence about the strength of the associations found and the conclusions drawn, while leaving room for further growth in understanding.

The accumulating body of research offers potential building blocks for innovative solutions to old problems, yet, it also is raising provocative questions. It is becoming evidently clear, as noted by Diez Roux et al., that “transportation and urban planning policies are, in fact health policy.”2 Structural modifications to already existing environments as well as an awareness of the promoters of good health including healthy food stores, physical activity environments, and neighborhood safety will enhance CVH. Social programs may boost communities and strengthen relationships among individuals in a community to enhance psychosocial well‐being and overall social cohesion. A sound understanding of these structural and psychosocial determinants of CVH, along with traditional behavioral modification and health education, will serve to effect population‐level change. Further research must strive to learn why and how neighborhood characteristics seem to act differentially on men and women and on different races and ethnicities. Answering these questions will get us closer to understanding and eliminating CVD disparities. The future of this research program seems bright; however, because techniques are now available to study gene and protein expression as people are being exposed to their neighborhoods' stimuli. Addressing the macro‐environment and understanding the micro‐biological differences among different sexes and races/ethnicities will bring us ever closer to eliminating CVD.

4. CONCLUSION

The use of cross‐sectional, longitudinal, and analytic regression modeling in analyzing the data that has emerged from prospective, observational cohort studies offer valuable, new information about the well‐researched topics of cardiovascular risk. The inclusion and consideration of various physical entities, such as healthy food stores and walking environments, and social factors, like perceived discrimination and social cohesion, into the cardiology literature can help to broaden therapeutic approaches to preventing and treating CVD. These variables impact men, women, and races differently is notable in that it suggests that one solution will not benefit all people equally. Instead, a multi‐modal approach addressing both structural and psychosocial issues would be more effective in changing the CVH of individuals in communities where we live.

CONFLICTS OF INTERESTS

Yang‐Yu Xiao reports no conflicts of interests and Garth Graham reports salary from Aetna Inc.

Xiao Y‐YK, Graham G. Where we live: The impact of neighborhoods and community factors on cardiovascular health in the United States. Clin Cardiol. 2019;42:184–189. 10.1002/clc.23107