Abstract
Objectives.
Using data from black and white adults enrolled in a community-based, multi-city cohort assembled in the mid-1980s, we examined whether reported experiences of interpersonal racial and gender discrimination differentially impacted on future cardiovascular health (CVH) depending on gendered race and the setting in which the interactions were reported to have occurred.
Methods.
Discrimination in eight possible settings were assessed using the Experiences of Discrimination scale at year 7; CVH two decades later was examined using a modified Life’s Simple 7 score, with higher scores indicating better health. Separate multivariable linear regressions evaluated the associations between reports of racial and gender discrimination and CVH score in each possible setting stratified by gendered race.
Results.
Mean (SD) CVH scores at year 30 were 7.8(1.9), 8.1(1.8), 8.9(2. 0), and 8.8(1.8) among black women, black men, white women, and white men, respectively. For black women, reporting both racial and gender discrimination while receiving medical care was associated with lower CVH score. Among black men, reporting both forms of discrimination while getting a job, at work, at school, and while receiving medical care was associated with lower CVH score. Among whites, reported discrimination while obtaining housing and by the police or courts (women), and in public and at work (men), was associated with a lower CVH score.
Conclusions.
The setting in which discrimination is reported may be an important indicator of whether discriminatory experiences are negatively associated with CVH, providing insight on distinct effect pathways among black and white women and men.
Keywords: Gendered race, interpersonal discrimination, cardiovascular health, intersectionality, identity pathology, social context
Introduction
Social determinants of health have recently taken center in public health endeavors to address glaring health inequities between marginalized and dominant-status groups in the United States [1]. This shift has resulted in part from a growing recognition of the downstream effects of structural factors such as institutional racism in shaping access to health-impacting resources [2]. As an increasing body of research has been dedicated to clarifying the effects of social exposures on disease outcomes, the complexity of these relationships has become apparent. Much of the difficulty in establishing clear evidence for pathways from structural to health inequity stems from conceptualizing and operationalizing appropriate surrogates of structural inequity [3]. Interpersonal discrimination is among the most frequently studied measures of inequitable social conditions in epidemiologic research [3]. The case of interpersonal discrimination and cardiovascular disease among black and white women and men provides an informative opportunity to examine some challenges in identifying and intervening on the social causes of unequal health outcomes.
Common approaches to examining the relationship between interpersonal discrimination and cardiovascular disease
Interpersonal discrimination has been defined as “encounters between individuals in which one person acts in an adversely discriminatory way toward another person” [3]. When these behaviors are based on characteristics such as race or gender, they are considered racial or gender discrimination. Typically, interactions encompass what have been recently termed microagressions (e.g. being followed in a store, receiving poor customer service, belittling remarks, skepticism of capability to complete a job or task, sexual harassment, police harassment, etc.) as well as overt expressions of racism or sexism [3]. Structural discrimination, on the other hand, describes the set of social institutions, laws, cultural practices, and norms that systematically disadvantage particular groups while advantaging others [3].
Interpersonal discrimination, particularly racial discrimination, has long been the focus of efforts to explain persistent disparities in cardiovascular disease morbidity and mortality between black and white women and men [2–7]. Often conceptualized as a proxy for structural discrimination, or, alternatively, as a mechanism through which structural discrimination acts on health, interpersonal discrimination provides an accessible method for investigating social determinants of health [3]. The underlying assumption for the majority of studies examining interpersonal discrimination appears to be that the stress associated with experiencing discriminatory interactions has a detrimental effect on cardiovascular health (CVH), directly through chronic activation of the stress response system, or indirectly through promoting poor health behaviors, which in turns increases risk for cardiovascular morbidity and mortality [3,6,8].
The resulting stress stemming from perceiving discrimination on the basis of classification in a social group, particularly race or gender, has been shown to exert unique physiological and psychological impacts [3,8]. As such, populations more likely to encounter these experiences (e.g. black persons compared with white persons in the case of racial discrimination) will exhibit poorer health behaviors, experience higher rates of cardiometabolic dysfunction, and necessarily have a greater burden of disease. Admittedly, researchers have emphasized relevant differences in the effects of exposure depending on the basis of discrimination [3,9] (racial versus weight, for example), the frequency of discrimination [3], demographic characteristics such as the age or gender of the individual to whom the discrimination is directed [10], and how individuals respond to stress [11,12]. Even still, few have theoretically considered the nature of these differences and whether the reasons for these differences have implications for the exposure-disease relationship; even fewer have taken these potential implications into account during analysis.
The large, conflicted body of literature on the topic provides testament to the oversimplification of, and, perhaps, misguided focus on, interpersonal discrimination as an explanation for the excess burden of cardiovascular disease among black persons [3]. Some studies have cross-sectionally linked racial and weight discrimination to higher blood pressure [13,14], increased rate of cigarette smoking [15], and changes in waist circumference and BMI [16,17], while others have failed to find positive associations with cardiovascular disease risk factors among black women and men [5,18,19]. Two of three prospective analyses have linked reported overall and racial discrimination to incident cardiovascular disease among black and white women and men while adjusting for race/ethnicity and/or sex [20,21]. The third found no association with the same outcome, and a negative association with all-cause mortality, in examining only black women and men [7]. While more longitudinal studies are necessary to make any conclusive claims, these conflicted findings underlie a necessity for clarifying the relationship between structural inequity, interpersonal discrimination, and cardiovascular disease disparities. The current evidence calls into question to what degree disparate cardiovascular disease outcomes between black and white women and men can be attributed to interpersonal discrimination.
Sources of complexity in measuring the cardiovascular health impacts of interpersonal discrimination
Evidence suggests that the complex, multifaceted nature of the interpersonal discrimination experience operates within distinct social groups to differentially influence CVH in a manner not frequently captured in epidemiologic studies [3,5,6,10]. Inconsistencies in the literature may be attributable, in part, to an inadequate conceptualization, measurement, and analysis of interpersonal discrimination in relation to CVH across demographically diverse populations.
One such population designation is “gendered race”, a term increasingly used to describe inseparable racialized and gendered identity characteristics (among many other health-impacting social group categorizations) that are internalized in the context of social inequity [10,11]. Identity beliefs (e.g. gender-dependent racial identity) associated with social group membership can lead to variation in the overlap between the actual occurrence, perception, and reporting of discrimination, as described in the emerging Identity Pathology (IP) framework [22] and other critical race theories [23]. The degree of social group identification and preference for social hierarchy have been shown to vary with gendered race [10,21,24]; both correlate with likelihood of perceiving discrimination independent of objective occurrences of discriminatory treatment [10,23–26]. In this way, the degree of measurement error (alignment of conceptualization and operationalization of discrimination) becomes dependent on gendered race. Additionally, identity beliefs diversify susceptibility to the health consequences of adverse psychosocial exposures such as interpersonal discrimination through influencing the stressfulness of the experiences and access to positive coping resources across different gendered race groups [12]. When conceptualizing individuals as possessing intersectional identities, the process of measuring the health effects of highly subjective psychosocial exposures such as interpersonal discrimination becomes all the more difficult in ways often overlooked in common epidemiologic methods.
Beyond internal psychosocial characteristics that complicate the measurement of interpersonal discrimination, Intersectionality Theory [27] recognizes the unique impact of multiple intersecting domains of oppression that act simultaneously to shape health-impacting resources for individuals located at different social junctures. While evidence has demonstrated a distinct effect of reported racial discrimination compared with other forms of interpersonal discrimination [3,5,9], such conceptualizations of the social experience outlined by Intersectionality Theory, along with tenets of the IP model [22], requires the interrogation of what is actually being captured when asking individuals to distinguish the basis of the discrimination directed towards them. With quantitative methods limiting the ability to conduct such inquiries, a more accurate risk profile may be obtained in quantitative studies by using theory-driven approaches to assessing the health impact of reporting discrimination on the basis of multiple social group characteristics at once. A recent analysis grounded in IP theory, for example demonstrated how simultaneously reported racial and gender discrimination is not only differentially associated with CVH across gendered race groups, but is also associated with CVH in ways distinct from reported racial or gender discrimination alone that follow a pattern outlined by the framework [22].
Previous findings also suggest that the magnitude of stress discriminatory experiences cause and whether responses to these experiences exacerbate or reduce the risk of cardiovascular disease depends on the context in which they occur [8,24,28]. Therefore, in addition to the challenge of capturing variation in the subjective identity characteristics that might render interpersonal discrimination detrimental to CVH, as well as the complex psychological processes by which individuals attribute discriminatory experiences, it is also necessary to consider how the setting in which discrimination is reported reflects access to both internal and external psychosocial resources that may independently relate to CVH differently for different gendered race groups. Everyday experiences of discriminatory treatment not only encompass individual acts but also the complex relation of acts that will be specific to specific social contexts, as argued by some critical race theorists [23]. The particular relationship between individual and context bears important implications for the physiological impact of perceived discriminatory interactions.
Further, the context of reported discrimination, such as at school, at work, by the police or courts, or while seeking healthcare, may provide insight into distinct effect pathways operating among different gendered race groups. While discrimination may act directly on CVH through repeated activation of the stress response system for some, others may be more susceptible to the indirect effects of interpersonal discrimination such as barriers to quality health care [2,3,8]. The IP framework [22] argues that this susceptibility to direct versus indirect effects of discrimination on CVH are primarily a function of an individual’s cumulative social experiences and will therefore manifest differentially across gendered race groups. To the best of our knowledge, no studies have previously examined the importance of social setting in determining whether reporting interpersonal discrimination will affect CVH, and whether these relationships depend further on an individual’s gendered race.
The benefits of a life-course approach to the study of social determinants of cardiovascular disease
The complex pathogenesis of the set of syndromes captured under the umbrella of cardiovascular disease renders these outcomes ideal for the study of social determinants of health. Conditions that arise in utero and behaviors begun in early childhood can affect risk for the development of cardiovascular disease decades later [1,2,4]. Because of the cumulative nature of cardiovascular pathologies, a life-course approach enables a more comprehensive understanding of the relative contribution of various risk factors to the progression of disease. For psychosocial risk factors such as interpersonal discrimination whose relevant etiologic periods are difficult to determine, examining longitudinal health effects can provide insight into the pathways over which these exposures act to shape cardiovascular health.
Current study
The objective of this study was to engage an emerging theory, the Identity Pathology framework, to challenge common assumptions about the role of interpersonal discrimination in contributing to social group disparities in CVH. To do so, we explored two sources of complexity in the relationship of interpersonal discrimination with CVH: complexities arising from gendered race effects and those arising from social setting, approaching the measurement of interpersonal discrimination from an intersectional perspective. Specifically, using data from the Coronary Artery Risk Development in Young Adults (CARDIA) study, we explored variation in the relationship between simultaneously reported racial and gender discrimination and future CVH across a number of social settings in a large, community-based population of black and white women and men.
Methods
Study design and participants
The CARDIA study is an ongoing population-based prospective cohort study of risk factors for for cardiovascular disease conducted in four U.S. centers (Birmingham, AL; Minneapolis, MN; Chicago, IL; and Oakland, CA). A total of 5,114 self-reported, non-Hispanic black or white persons, aged 18 – 30 years at the baseline examination (1985–1986), were recruited primarily from random-digit dialing of community lists and random selection from a health-care plan [28,29]. The goal of recruitment was to balance gender and race; those aged 18 – 25 years and aged 25 – 30 years; and those attaining a high school education or less versus more education, at each of the four centers. The institutional review board at each center approved the CARDIA study protocol and informed consent was obtained from each participant. Following the initial in-person examination, participants were re-examined at years 2, 5, 7, 10, 15, 20, 25, and 30 post-baseline. Individuals missing a CVH score (n=787) or data on discrimination (n=1,089) were excluded, leaving an analytic sample of 3,758 participants.
Self-reported interpersonal discrimination
In contrast to the ways in which interpersonal discrimination is commonly conceptualized in epidemiologic literature [15–18], we do not consider racial and gender discrimination as separate exposures with the capacity to interact. We consider simultaneously reported experiences of racial and gender discrimination as an imperfect measure of the way interpersonal discrimination is experienced by individuals who are themselves simultaneously racialized and gendered. Any gender discrimination is inherently racialized, and any racial discrimination inherently gendered [10,11]. In efforts to capture this framing, we examined simultaneously reported racial and gender discrimination in each setting.
Discrimination was assessed in CARDIA 7, 15, and 20 years after the baseline, in-person clinical examination using the valid and reliable Experiences of Discrimination Scale.13 Because preliminary analyses of CARDIA data showed the prevalence of reported race and gender discrimination is comparable at years 7, 15, and 25 within each gendered race group, we used discrimination reported at year 7 only rather than treating reported discrimination as time-varying. We decided on year 7 both because the age of the cohort at this time point allowed that those who would experience discrimination most likely would have done so and in order to establish temporal feasibility between the exposure and outcome. Participants reported having ever experienced discrimination, been prevented from doing something, or been hassled or made to feel inferior (yes/no) in any of the following situations because of their 1) gender and 2) race or color: at school; getting a job; at work; at home; getting medical care; getting housing; by the police or courts; or on the street or in a public setting. At year 7, the racial discrimination scale did not include “at home” whereas for the gender discrimination scale “getting housing” or “by the police or courts” were not included. The combined racial and gender discrimination exposure was created as a binary variable, categorized as “yes” or “no”, respectively, if participants answered “yes” or “no” to both racial and gender discrimination in each setting (excluding the settings not included on both scales).
Cardiovascular health
Cardiovascular health (CVH), as distinct from CVD, refers to a multidimensional measure recently adopted by the American Heart Association (AHA) as an important area of focus for CVD prevention efforts [30]. The primary outcome of this study was a CVH score measured at year 30, based on the AHA’s Life’s Simple 7 cardiovascular health scoring method [30]. Because no dietary measures were included in the present study, the composite score consisted of six rather than the standard seven measures. We have previously published using this method [22], and despite this modification, we believe that the included markers sufficiently represent risk for poorer cardiovascular health as each condition included in the composite score has been previously shown to individually correlate highly with poor cardiovascular outcomes. These measures included: two behavioral factors (smoking status and physical activity) and four cardiometabolic factors (hypercholesterolemia, hypertension, obesity, and diabetes defined per AHA and National Heart, Lung, and Blood Institute guidelines) [30]. The total CVH score was calculated by adding the points assigned to each factor’s specification. Smoking status was operationalized as self-report of never (2 points), former (1 point), or current (0 points). Physical activity was defined according to Centers for Disease Control and Prevention (CDC) guidelines [30] as ≥75 minutes/week vigorous physical activity (VPA) or ≥150 minutes/week of moderate physical activity (MPA) (2 points); <75,>0 minutes/week VPA or <150,>0 minutes/week MPA) (1 point); and none (0 points). Hypercholesterolemia was operationalized as total cholesterol <200 mg/dL (2 points); 200–239 mg/dL (1 point); and ≥240 mg/dL (0 points); systolic hypertension as the average of three systolic readings <120 mm/Hg (2 points); 120–139 mm/Hg (1 point); and ≥140 mm/Hg (0 points); obesity as a body mass index (BMI) of <25 kg/m2 (2 points); 25–29.9 (1 point); ≥30 (0 points); and diabetes as a fasting blood glucose of <100 mg/dL (2 points); 100–125.9 (1 point); and ≥126 (0 points). Total scores ranged from 0–12, with higher points indicating healthier status. Although right-truncated, the CVH score met assumptions of normality and was treated as a continuous variable.
Covariates
Potential confounders for this analysis included age and the study center in which the baseline exam was conducted. Other potentially relevant sociodemographic variables such as annual family income, marital status, and education level were not included as potential confounders because they were conceptualized as potential mediators of the association between reported interpersonal discrimination and CVH. Due to previous research indicating racial differences in the associations of socioeconomic status with both prevalence of reported discrimination and cardiovascular disease risk factors [8,13,24], however, we also conducted sensitivity analyses among white women and men which included years of education as a covariate in regression models. Information on race and gender were collected at baseline, and a race-by-gender variable was used to operationalize gendered race. Since less than 5% of the study population reported changes in their geographic location between baseline and year 7, we used the study center in which the baseline examination was conducted to characterize participants’ geographic location.
Statistical analysis
Descriptive statistics including age, study center, education (included for descriptive purposed only), type and level of discrimination, and CVH scores, were calculated for each gendered race group using Pearson’s chi-square test for categorial variables and t-tests for continuous variables. Multivariable linear regression analyses were performed to evaluate the associations between level of perceived discrimination at year 7 and CVH scores at year 30 (or the last follow-up) for those reporting both racial discrimination and gender discrimination, separately for each setting and further stratified by gendered race group. To address potential bias from attrition and gendered race group compositional effects, we also conducted sensitivity analyses weighting effect estimates. Weights were calculated using combined inverse probability of missing at year 30 (age, study center, and years of education) and group propensity score matching (age, study center, and years of education) for each gendered race group relative to white men. Finally, we adjusted for multiple testing using the Holm method, which has been argued as preferable to the Bonferroni method [32]. Analyses were conducted using Stata 14.0.
Results
Study population
The mean age of the study sample of 3,758 participants at year 30 was 55 years, and 28% were black women, 20% were black men, 29% were white women, and 26% were white men. On average, white women and men completed 16 years of education by year 30 compared with 14 years among black women and men (Table 1).
Table 1.
Characteristics for CARDIA Participants by Gendered Race: Year 30 (2015–16)
Black women n=1,039 | Black men n=743 | White women n=1,045 | White men n=931 | |
---|---|---|---|---|
Age, mean (SD) | 54.6 (3.8) | 54.3 (3.7) | 55.6 (3.4) | 55.5 (3.3) |
Education, mean yrs (SD) | 14.6 (2.4) | 14.0 (2.4) | 16.1 (2.4) | 16.0 (2.6) |
Study Center, % | ||||
Birmingham | 24.4 | 25.7 | 17.0 | 22.6 |
Chicago | 22.0 | 20.2 | 21.6 | 23.0 |
Minneapolis | 17.7 | 24.5 | 32.9 | 33.5 |
Oakland | 35.8 | 29.6 | 28.4 | 20.9 |
CVH score, mean (SD) | 7.8 (1.9) | 8.1 (1.8) | 8.9 (2.0) | 8.8 (1.8) |
Body mass index, kg/m2 % | ||||
<25 | 13.8 | 20.1 | 38.7 | 22.2 |
25–29.9 | 24.3 | 35.0 | 26.7 | 40.7 |
≥30 | 62.0 | 45.0 | 34.6 | 37.1 |
Total cholesterol, mg/dL % | ||||
<200 | 64.1 | 71.6 | 52.7 | 65.7 |
200–239 | 26.3 | 22.2 | 34.5 | 25.9 |
≥240 | 9.6 | 6.2 | 12.7 | 8.4 |
Systolic blood pressure, mm/Hg % | ||||
<120 | 43.6 | 41.0 | 68.9 | 55.6 |
120–139 | 39.0 | 43.3 | 24.3 | 37.1 |
≥140 | 17.4 | 15.6 | 6.8 | 7.3 |
Fasting blood glucose, mg/dL % | ||||
<100 | 67.9 | 61.5 | 78.0 | 61.0 |
100–125.9 | 22.3 | 27.2 | 17.5 | 30.7 |
≥126 | 9.7 | 11.3 | 4.5 | 8.3 |
Smoking status, % | ||||
Never | 64.6 | 56.8 | 57.1 | 64.3 |
Former | 16.0 | 16.4 | 31.4 | 23.0 |
Current | 19.4 | 26.8 | 11.5 | 12.7 |
Physical activity, min/wk % | ||||
≥75 VPA or ≥150 MPA | 48.9 | 73.1 | 70.4 | 81.8 |
<75, >0 VPA or <150, >0 MPA | 43.1 | 22.6 | 26.7 | 16.8 |
0 VPA and 0 MPA | 8.0 | 4.3 | 2.9 | 1.4 |
Cardiovascular health scores were calculated based on data collected in year 30 or the last follow-up after year 7, using six components with a total possible 12 points: body mass index, total cholesterol, systolic blood pressure, fasting glucose, smoking status, and physical activity. Higher scores indicate better health.
Baseline characteristics
The proportion of participants who reported interpersonal racial and gender discrimination in each setting varied within and across gendered race groups (Table 2). For all groups, discrimination on the street or in public was most frequently reported, and experiences of discrimination while receiving medical care were the least frequently reported. Exposure was most prevalent among black women in every setting, with the exception of discrimination while seeking a job; in this setting, the proportion of black men reporting discrimination was higher than the proportion of the other comparison groups. Mean CVH scores at year 7 were highest among white women, and were not statistically different across settings among black women and men. White women who reported racial and gender discrimination while seeking housing (9.7, SD=1.8) and by the police or courts (9.2, SD=1.6) had the lowest mean CVH scores at year 7 of their gendered race group. White men who reported racial and gender discrimination in the street or in public had the highest year 7 CVH scores of their gendered race group (9.9, SD=1.6), while white men who reported being discriminated against while receiving medical care had the lowest scores of their group (8.8, SD=1.8). All comparisons were statically significant at p=0.05.
Table 2.
Racial and Gender Discrimination and Mean Cardiovascular Health Score across Settings by Gendered Race: CARDIA, Year 7 (1992–93)
Black women | Black men | White women | White men | |||||
---|---|---|---|---|---|---|---|---|
| ||||||||
Settinga | Disc., %b | Mean (SD) CVH Scorec | Disc., % | Mean (SD) CVH Score | Disc., % | Mean (SD) CVH Score | Disc., % | Mean (SD) CVH Score |
In public/on the street | 47.4 | 9.3 (1.7) | 40.1 | 9.5 (1.6) | 21.1 | 10.3 (1.5) | 8.3 | 9.9 (1.6) |
Seeking a job | 31.6 | 9.4 (1.7) | 32.1 | 9.3 (1.6) | 3.2 | 10.3 (1.4) | 3.4 | 9.4 (1.6) |
Seeking housing | 29.5 | 9.3 (1.7) | 31.2 | 9.2 (1.7) | 1.3 | 9.7 (1.8) | 1.1 | 9.8 (1.6) |
At work | 38.7 | 9.4 (1.7) | 33.5 | 9.3 (1.6) | 5.7 | 10.3 (1.5) | 3.7 | 9.6 (2.0) |
At school | 18.6 | 9.3 (1.7) | 15.1 | 9.5 (1.5) | 5.7 | 10.5 (1.3) | 2.6 | 9.5 (1.7) |
Receiving medical care | 9.0 | 9.5 (1.6) | 7.4 | 9.3 (1.7) | <1.0 | 10.1 (1.5) | 1.0 | 8.8 (1.8) |
By the police or courts | 26.9 | 9.4 (1.7) | 57.1 | 9.4 (1.7) | 2.1 | 9.2 (1.6) | 3.2 | 9.2 (1.6) |
At home | 13.6 | 9.3 (1.8) | 7.9 | 9.1 (1.6) | 21.6 | 10.1 (1.5) | 5.0 | 9.7 (1.8) |
At year 7, discrimination “at home” was excluded from the race or color scale; “by the police or courts” and “getting housing” were excluded from the gender scale
Racial and gender discrimination simultaneously reported
Cardiovascular health scores are calculated based on data collected in year 7, using six components with a total possible 12 points: body mass index, total
cholesterol, systolic blood pressure, fasting glucose, smoking status, and physical activity. Higher scores indicate better health.
Reported Discrimination and CVH
With a single exception (while seeking medical care, B = −0.5, 95% CI: −0.9, −0.1), multivariable-adjusted CVH score differences at year 30 associated with reported racial and gender discrimination in each of the eight settings examined were either non-significant or were associated with a positive score difference among black women (Table 3). For black men, self-reported discrimination in four of the eight settings was significantly associated with a negative score difference; while seeking a job (B= −0.3, 95% CI: −0.6, 0.0), at work (B= −0.4, 95% CI: −0.7, −0.1), at school (B= −0.4 95% CI: −0.8, 0.0), and while seeking medical care (B = −0.7, 95% CI: −0.9, −0.1). Associations across settings also differed between white women and men. For white women, reported racial discrimination by the police or courts (B = −1.1, 95% CI: −2.0, −0.3) or while seeking housing (B = −1.5, 95% CI: −2.5, −0.4) was associated with lower CVH scores, while among white men, self-reports of racial and gender discrimination in public (B = −0.5, 95% CI: −1.0, −0.1) or at work (B = −1.0, 95% CI: −1.6, −0.3) were associated with a lower CVH score.
Table 3.
Adjusteda Difference in Cardiovascular Health Scoreb at Year 30 of the CARDIA Study across Settings of Simultaneously Reported Racial and Gender Discrimination at Year 7, β (95% CI): CARDIA, 1992–2016
Settingc | Black women | Black men | White women | White men |
---|---|---|---|---|
In public/on the street | +0.2* (0.0, +0.5) | 0.0 (−0.3, +0.3) | +0.1 (−0.2, +0.5) | −0.5* (−1.0, −0.1) |
Getting a job | 0.0 (−0.3, +0.3) | −0.3* (−0.6, 0.0) | −0.6 (−1.3, +0.1) | −0.5 (−1.1, +0.1) |
Getting housing | −0.1 (−0.3, +0.2) | −0.2 (−0.5, +0.1) | −1.5* (−2.5, −0.4) | −0.4 (−1.5, +0.7) |
At work | +0.1 (−0.2, +0.4) | −0.4* (−0.7, −0.1) | −0.4 (−0.9, +0.1) | −1.0* (−1.6, −0.3) |
At school | +0.3* (0.0, +0.6) | −0.4* (−0.8, 0.0) | −0.1 (−0.6, +0.5) | −0.3 (−1.1, +0.4) |
Receiving medical care | −0.5* (−0.9, −0.1) | −0.7* (−0.9, −0.1) | −1.5 (−3.8, +0.7) | −1.1 (−2.7, +0.5) |
By the police or courts | −0.1 (−0.2, +0.4) | −0.1 (−0.4, +0.2) | −1.1* (−2.0, −0.3) | −0.3 (−0.9, +0.4) |
At home | +0.1 (−0.3, +0.4) | −0.1 (−0.6, +0.4) | −0.2 (−0.5, +0.1) | −0.2 (−0.7, +0.3) |
All models are adjusted for age and study center.
Health scores are calculated based on data collected in year 30 or the last follow-up after year 7, using six components with a total possible 12 points: body mass index, total cholesterol, systolic blood pressure, fasting glucose, smoking status, and physical activity. Higher scores indicate better health.
At year 7, discrimination “at home” was excluded from the race or color scale; “by the police or courts” and “getting housing” were excluded from the gender scale.
Denotes statistically significant estimates at p=0.05.
Sensitivity analyses and multiple testing adjustment
The inclusion of years of education in regression models for white women and men did not change effect estimates by more than 10%; these results were therefore not reported. Estimates weighted for attrition and group compositional effects were also not statistically different from unweighted estimates; these results are not reported. Adjusting for multiple comparisons using the Holm method yielded corrected critical p-values of 0.006 for black women, 0.008 for black men, and 0.007 for white women and men. The effect estimates meeting these thresholds within each group are reported in supplementary table 1.
Discussion
Simultaneously reported racial and gender discrimination were differentially associated with CVH depending on gendered race and setting. Among black women, reported instances of interpersonal discrimination were not associated with CVH or were associated with a higher CVH score while the opposite findings were observed among all other gendered race groups. For all groups, reporting discriminatory experiences while receiving medical care had a negative impact on future CVH, although effect estimates did not reach statistical significance among white women and men. While the predicted effect on subclinical or clinical cardiovascular disease of a one- or two-point change in CVH score has not been documented as far as we are able to determine, such changes in CVH could signify the transition from ideal blood pressure to hypertension, from overweight to obese, etc. Because each of the risk factors comprising the CVH score are independently associated with increased risk of cardiovascular disease, we interpret the observed changes as both clinically and theoretically meaningful.
Gendered racial differences in the effects of discrimination on CVH
A number of previous studies have examined the effects of racial discrimination on risk factors for cardiovascular disease including hypertension [33], sedentary behavior [34], cigarette smoking [15], and inflammation [18] in black and white persons. These analyses consistently found that associations observed depended on gendered race. For example, in the study assessing racial discrimination and inflammation, the authors found an association only among white women [16]. Other studies have assessed multiple forms of discrimination on CVD risk [9,20]. Most found that whether multiple forms of discrimination were more detrimental to CVH than a single form depended on the type of discrimination reported and the race and gender of the individuals reporting. Furthermore, although we could not find any investigations that examined whether simultaneous reports of racial and gender discrimination differentially impacted on CVH in different social contexts, studies that have examined associations of reported racial or gender discrimination in occupational [35,36] and healthcare settings [37,38] with CVD-related outcomes have also found differences between black and white women and men.
That the settings in which reported racial and gender discrimination were associated with poorer CVH differed among black and white women and men is consistent with disparate effect pathways for these groups that may be linked to gendered race-specific characteristics. Our findings suggest that for black women, interpersonal experiences of discrimination are more likely to act indirectly on CVH by deterring access to health-influencing resources such as medical care, a mechanism that has been demonstrated in previous research [38]. Black women who reported racial and gender discrimination while receiving medical care were the only individuals of their gendered race group to experience a decline in CVH associated with reported discrimination; reported exposure in other settings was measured as either protective or had no influence on CVH. Rather than yielding a greater vulnerability to the negative health consequences of psychosocial stress as might be intuitively concluded, these findings suggest that black women may more readily adapt to hostile social environments such that the effects of recurrent interpersonal discrimination on the stress response system [12], or on certain health-related behaviors that preempt cardiovascular disease [34], are minimized in comparison to other gendered race groups. We do not suggest that black women are immune to the physiological impacts of the discrimination they report. Rather, we take these findings as evidence that structural barriers, such as reduced access to high-quality medical care, may have a much more compelling effect on the cardiovascular health of black women than stress stemming from encounters with interpersonal discrimination, as has been previously argued [4,8].
The settings in which reported discrimination impacted CVH among black men in this study indicate that members of this group may be more susceptible to the direct physiological impact of perceived subordinate status than black women. This may be because racism targeted at black men has historically been more ostensibly violent [10], or due to other psychosocial and cultural factors influencing the distinct coping methods of these groups [10,12,27]. In either case, we were surprised at the lack of association of CVH with reported discrimination by the police or courts among black men in this study. We consider that the prevalent, highly publicized, severe mistreatment that black men have endured by the police may enable black men to adapt and cope specifically with these encounters in healthier ways as a result of both the duration and validation of these experiences. Although hypervigilance has been considered as detrimental to health [34], in encounters with law enforcement, expectation and recognition of mistreatment may lead black men to connect with resources which ameliorate long-term physiological effects of associated stress. With this exception, the observed patterns in the associations of reported discrimination with CVH indicate that reminders of marginalized status may be experienced as more stressful among black men than black women and therefore may be more likely to act on CVH through direct physiological mechanisms in addition to creating barriers to health and social resources in this group.
The findings of the present study also provide evidence that reported racial and gender discrimination act on CVH both directly and indirectly for white men and women, although we hypothesize that these effects are likely not triggered by the same psychosocial cues operating among black persons [22]. For members of dominant status groups, reported experiences of racial and gender discrimination in settings with a heightened risk of perceiving violations of their group privileges [21] may be inherently more stressful for individuals aware of their racial or gender privilege [21], as is suggested by prior research finding an association of racial privilege with poorer health and health behaviors among whites [39]. Our findings that reports of racial and gender discrimination in public and at work are associated with poorer CVH among white men are consistent with this theory. More research is necessary to better understand why experiences of discrimination in certain settings, such as perceptions of race-based mistreatment while seeking housing, may be more damaging to the CVH of white women than discrimination in other settings.
Setting differences in the effects of discrimination on CVH
In demonstrating how discriminatory experiences in certain settings have unique impacts on CVH among black and white women and men, our findings build on previous work we have conducted on the complexity of interpersonal discrimination and CVH. In a study of overall (not by setting) reporting of racial and/or gender discrimination also using CARDIA data, we found an association of simultaneously reported racial and gender discrimination with poorer CVH only among white men [22]. This discrepancy in our findings offers further evidence that grouping certain experiences of discrimination can mask important differences in effect; when further examined by setting, we see that negative associations are present for white women and black men as well.
One explanation for the patterns we observed in the present study is that interpersonal discrimination may act as an “identity trigger” consistent with claims of the Jedi Public Health framework [40]. The authors suggest that identity triggers, or elements of the social environment that trigger awareness of one’s social status, are one mechanism through which structured inequities act to differentially impact on health and lead to health disparities. The unequal social conditions in which black and white women and men are situated influence the type and saturation of identity triggers each of these groups will encounter, as well as available coping resources [41,42], within and across various social settings [40]. According to this framework, experiences of discrimination pose a setting-specific disease risk for each gendered race group. We suggest further that perceived experiences of interpersonal discrimination can act as identity cues, even in the absence of actual occurrences of discrimination, which might partially explain the associations we found among white women and men. Identity triggers and the perceived coping resources [41] particular to black and white women and men may act to specify conditions under which experiences of interpersonal discrimination will have a measurable impact on CVH.
Hierarchical social conditions create power dynamics between marginalized and dominant status groups which influence how inequity will be experienced on a personal basis by members of both types of groups [3,10,12,27]. Experiences of discrimination based on gendered race that occur in the context of medical care, education, or in interactions with law enforcement, for example, can bring to bear historically structured power imbalances through heightened awareness of one’s stigmatized status in the form of race consciousness [37,43]. Instances of interpersonal discrimination in these specific settings may be uniquely stressful for marginalized persons both because of the likelihood of recurrence and a perceived lack of opportunities for retribution [40–42].
On the other hand, the settings in which awareness of unequal social status might be triggered among dominant group members—whether or not a discriminatory interaction actually occurred—and the resources they believe are available for coping with the accompanying stress, likely differ. These perceptions of social status triggered by interpersonal discrimination lead to between-group differences in the types of social contexts in which experiencing discrimination will contribute to deteriorated CVH. This interplay is consistent with our findings that although a higher percentage of black men reported encountering discriminatory treatment in public or on the street than in any other setting, this setting was the only one in which exposure was not associated with poorer CVH within this group. In contrast, “in public or on the street” was one of the two settings in which white men who reported experiencing racial and gender discrimination experienced declining CVH. Given the historical contexts in which white men’s social status afforded a measure of public and occupational deference [21,27], for some white men instances when this deference is absent or challenged in settings such as on the street or at work may be more likely to be perceived as discriminatory and more stressful than encounters perceived as discriminatory in other settings, an explanation that is consistent with our findings.
Limitations
Our findings need to be considered in the context of several limitations. There is some indication that race consciousness among white persons, which has been interpreted as an awareness of racial privilege [27,39], in the context of medical care is associated with lower medication adherence [37]. Our point estimates for change in CVH score in this study were highest in the setting of receiving medical care for white women and men, indicating that interpersonal discrimination may impact on CVH through influencing health behaviors such as accessing healthcare. However, we cannot draw any conclusions due to a limited sample size. We also caution the generalizability of our findings. Despite CARDIA’s rigorous study design and samples drawn from multiple geographic areas, because of the interdependency of structural and individual-level factors that influence how interpersonal discrimination affects CVH, the results of this study cannot be used to directly infer conclusions about demographically dissimilar populations.
Conclusions
In revealing the importance of gendered race and setting in the health effects of simultaneously reported racial and gender discrimination, this study provides insight into the distinct mechanisms by which exposure to interpersonal discrimination operates to impact on the CVH of black and white women and men. These results underscore a need for a more nuanced approach to the study of the health-effects of interpersonal discrimination, one which better considers sources of effect variability such as context and psychosocial characteristics of the groups under study. The oversimplification of interpersonal discrimination and assumptions about its effect on health in epidemiological studies can lead to continued inconsistencies in the literature and further obfuscation of the mechanisms by which structural-level discrimination operates to cause a greater disease burden among black women and men.
Future research should also empirically evaluate whether experiences that impact on identity beliefs differentially influence the likelihood of reporting discrimination and risk for cardiovascular disease in these groups. The IP framework proposes that such identity beliefs may be better predictors of disease risk than reported experiences of discrimination, but this assertation has yet to be examined. Additional research is also needed to explore the ways in which white women and men conceptualize discrimination. The results of this study suggest that evaluating associations of reported discrimination with health among socially-privileged groups, as well as dually marginalized and privileged groups, can aid in identifying inconsistences in the conceptualization and operationalization of self-reported experiences of discrimination and point to areas of improvement. These lines of investigation would advance our understanding of whether and how interpersonal experiences of racial and gender discrimination contribute to persistent gendered racial disparities in risk of developing cardiovascular disease, but also whether the role of structural inequity in causing these disparities may be better captured through investigating other exposures.
Supplementary Material
Compliance with Ethical Standards
Approval for the use of Human Subjects data in the CARDIA study was granted by the Institutional Review Boards of the following institutions:
University of Alabama at Birmingham
Northwestern University
University of Minnesota
Kaiser Permanante, Research Division
References
- 1.Secretary’s Advisory Committee on Health Promotion and Disease Prevention Objectives for 2020. Healthy People 2020: An Opportunity to Address the Societal Determinants of Health in the United States. July 26, 2010. [Google Scholar]
- 2.Phelan JC and Link BG (2015). Is racism a fundamental cause of inequalities in health? Annual Review of Sociology, 41:311–33. [Google Scholar]
- 3.Krieger N (2014). Discrimination and health inequities. International Journal of Health Services Research, 44(4):643–710. [DOI] [PubMed] [Google Scholar]
- 4.Wyatt SB, Williams DR, Calvin R, Henderson FC, Walker ER, Winters K. Racism and cardiovascular disease in African Americans. Am J Med Sci. 2003;325(6):315–331. [DOI] [PubMed] [Google Scholar]
- 5.Lewis TT, Williams DR, Tamene M, and Clark CR (2014). Self-Reported experiences of discrimination and cardiovascular disease. Current Cardiovascular Risk Reports, 8(1):365. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ferdinand KC and Nasser SA (2017). Disparate Cardiovascular Disease Rates in African Americans: The Role of Stress Related to Self-Reported Racial Discrimination. Mayo Clinic Proceedings, 92(5):689–692. [DOI] [PubMed] [Google Scholar]
- 7.Dunlay SM, Lippmann SJ, Greiner MA, et al. (2017). Perceived discrimination and cardiovascular outcomes in older African Americans: insights from the Jackson Heart Study. Mayo Clinic Proceedings, 92(5):699–709. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Williams DR and Mohammed SA (2013). Racism and Health I: Pathways and Scientific Evidence. American Behavioral Scientist, 57(8). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Udo T and Grilo CM (2017). Cardiovascular disease and percieved weight, racial, and gender discrimnation in U.S. adults. Journal of Psychosomatic Research, 100:83–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Harnois CE and Ifatunji MA (2011). Gendered measures, gendered models: toward an intersectional analysis of interpersonal racial discrimination. Ethnic and Racial Studies, 34(6):1006–1028. [Google Scholar]
- 11.Szymanski Dawn M. and Lewis Jioni A. (2016). Gendered racism, coping, identity centrality, and African American college women’s psychological distress. Psychology of Women Quarterly, 40(2). [Google Scholar]
- 12.Bey GS, Waring ME, Jesdale BM, Person SD (2016). Gendered race modification of the association between chronic stress and depression among Black and White U.S. adults. Am J Orthopsychiatry, 88(2):151–160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Krieger N and Sidney S. (1996). Racial discrimination and blood pressure: The CARDIA Study of young black and white adults. American Journal of Public Health, 86:1370–1378 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sims M, Diez-Roux AV, Dudley A, et al. (2012). Perceived discrimination and hypertension among African Americans in the Jackson Heart Study. Am J Public Health, 102(Suppl 2):S258–S265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Borrell LN, Diez-Roux AV, Jacobs DR, Shea S, Jackson SA, Shrager S, and Blumenthal RS (2010). Perceived racial/ethnic discrimination, smoking and alcohol consumption in the Multi-Ethnic Study of Atherosclerosis (MESA). Preventive Medicine, 51(3–4):307–312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Cunningham TJ, Berkam LF, Kawachi I, et al. (2013). Changes in waist circumference and body mass index in the US CARDIA cohort: Fixed-effect associations with self-reported experiences of racial/ethnic discrimination. Journal of Biosocial Science, 45(2):267–278. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Johnson P, Markham Risica P, Gans KM, Kirtania U, and Kumanyika SA (2012). Association of Perceived Racial Discrimination with Eating Behaviors and Obesity among Participants of the SisterTalk Study. Journal of Black Nurses Association, 23(1): 34–40. [PMC free article] [PubMed] [Google Scholar]
- 18.Cunningham TJ, Seeman TE, Kawachi I, Gortmakera SL, Jacobs DR, Kiefe CI, and Berkman LF (2012). Racial/Ethnic and Gender Differences in the Association between Self-Reported Experiences of Racial/Ethnic Discrimination and Inflammation in the CARDIA Cohort of 4 US Communities. Social Science and Medicine, 75(5): 922–931. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Krieger N, Waterman PD, Kosheleva A, et al. Exposing racial discrimination: implicit & explicit measures--the My Body, My Story study of 1005 US-born black & white community health center members. PLoS One. 2011; 6(11):e27636. [PubMed: 22125618] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Everson-Rose S; Lutsey P; Roetker N; Lewis T; Kershaw K; Alonso A; and Diez Roux A (2015). Perceived Discrimination and Incident Cardiovascular Events: The Multi-Ethnic Study of Atherosclerosis. American Journal of Epidemiology, 182(3). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Pratto F, Sidanius J, and Levin S (2006). Social dominance theory and the dynamics of intergroups relations: Taking stock and looking forward. European Review of Social Psychology, 17:271–320 [Google Scholar]
- 22.Bey GS., Jesdale B, Forrester S, Person SD, & Kiefe C (2019). Intersectional effects of racial and gender discrimination on cardiovascular health vary among black and white women and men in the CARDIA study. SSM - population health, 8, 100446. 10.1016/j.ssmph.2019.100446Essed, [DOI] [PMC free article] [PubMed] [Google Scholar]; P (1991). Understanding Everyday Racism: An Interdisciplinary Theory. CA: Sage Publications. [Google Scholar]
- 23.Chae DH, Lincold KD, and Jackson JS (2011). Discrimination, Attribution, and Racial Group Identification: Implications for Psychological Distress Among Black Americans in the National Survey of American Life (2001–2003). American Journal of Orthopsychiatry, 81(4):498–506. [DOI] [PubMed] [Google Scholar]
- 24.Sellers RM and Shelton JN (2003). The role of racial identity in perceived racial discrimination. Journal of Personality and Social Psychology, 84(5), 1079–1092. [DOI] [PubMed] [Google Scholar]
- 25.Major B; Gramzow RH; McCoy SK; Levin S; Schmader T; and Sidanius J. (2002). Perceiving personal discrimination: The role of group status and legitimizing ideology. Journal of Personality and Social Psychology, 82(3), 269–282. [PubMed] [Google Scholar]
- 26.Crenshaw Kimberle (1989). Demarginalizing the intersection of race and sex: A black feminist critique of antidiscrimination doctrine, feminist theory and antiracist politics. University of Chicago Legal Forum, 1989(1):139–167. [Google Scholar]
- 27.Bey GS, Ulbricht CM, and Person SD (2018). Theories for race and gender differences in management of social identity-related stressors: a systematic review. Journal of Racial and Ethnic Health Disparities, 10.1007/s40615-018-0507-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Cutter GR, Burke GL, Dyer AR, Friedman GD, Hilner JE, Hughes GH, et al. (1991). Cardiovascular risk factors in young adults: The CARDIA baseline monograph. Controlled Clinical Trials, 12(1 Suppl), 1S–77S. [DOI] [PubMed] [Google Scholar]
- 29.Friedman GD, Cutter GR, Donahue RP, Hughes GH, Hulley SB, Jacobs DR Jr., et al. (1988). CARDIA: study design, recruitment, and some characteristics of the examined subjects. Journal of Clinical Epidemiology, 41(11), 1105–1116. [DOI] [PubMed] [Google Scholar]
- 30.Pool LR, Ning H, Lloyd-Jones DM, and Allen NB (2017). Trends in Racial/Ethnic Disparities in Cardiovascular Health Among US Adults From 1999–2012. JAHA, 6(9). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Aickin M and Gensler H (1996). Adjusting for multiple testing when reporting research results: The Bonferroni vs. Holm methods. American Journal of Public Health, 86(5). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Sims M, Diez-Roux AV, Dudley A, et al. (2012). Perceived discrimination and hypertension among African Americans in the Jackson Heart Study. Am J Public Health, 102(Suppl 2):S258–S265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Womack VY, Ning H, Lewis CE, Loucks EB, Puterman E, Reis J et al. (2014). Relationship between perceived discrimination and sedentary behavior in adults. American Journal of Health Behaviors, 38(5):641–649. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Krieger N, Chen JT, Waterman PD, Hartman C, Stoddard AM, Quinn MM, Sorensen G, and Barbeau EM (2008). The inverse hazard law: Blood pressure, sexual harassment, racial discrimination, workplace abuse, and occupational exposures in US low-income black, white, and Latino workers. Social Science & Medicine, 67(12):1970–1981. [DOI] [PubMed] [Google Scholar]
- 35.Rospenda KM, Richman JA, and Shannon CA (2008). Prevalence and mental health correlates of harassment and discrimination in the workplace: results from a national study. Journal of Interpersonal Violence, 42(5):819–843. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Brewer LC and Cooper LA (2014). Race, discrimination, and cardiovascular disease. Virtual Mentor AMA Journal of Ethics, 16(6):455–460. [PMC free article] [PubMed] [Google Scholar]
- 37.Shariff-Marco S, Klassen AC, and Bowie JV (2010). Racial/Ethnic differences in self-reported racism and its association with cancer-related health behaviors. American Journal of Public Health, 100(2):364–374. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Fujishiro K (2009). Is perceived racial privilege associated with health? Findings from the Behavioral Risk Factor Surveillance System. Social Science and Medicine, 68(5):840–844. [DOI] [PubMed] [Google Scholar]
- 39.Geronimus AT, James SA, Destin M, Graham LF, Hatzenbuehler ML, Murphy MC, Pearson JA, Omari A, and Thompson JP (2016). Jedi public health: Co-creating an identity-safe culture to promote health equity. SSM – Population Health, 2:105–116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Mezuk B, Abdou CM, Hudson D, Kershaw KN, Rafferty JA, Lee H, and Jackson JS (2013). “White Box” epidemiology and the social neuroscience of health behaviors: The Environmental Affordances model. Society and Mental Health, 3(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Brenner AB, Diez-Roux AV, Gabreab SY, Schulz AJ, and Sims M (2018). The Epidemiology of Coping in African American Adults in the Jackson Heart Study (JHS). Journal of Racial & Ethnic Health Disparities, 5(5):978–994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Jones CP (2002). Confronting institutionalized racism. Phylon, 50:7–22. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.