1. The Fundamental Causes of Breast Cancer Disparities
Breast cancer disparities persist despite the development and implementation of numerous interventions seeking to eliminate them
[15][1]. There are several comprehensive and wide-ranging reviews of the current state of scientific research on factors that contribute to breast cancer incidence, outcomes, and disparities
[3,16,17][2][3][4]. Overall, this
litrese
rature arch indicates that risk and protective factors are distributed differently by race, with racial minorities more likely than whites to be exposed to many risk factors. Specifically, it points to the need for prevention efforts aimed at
reducing alcohol consumption, tobacco use, and chronic stress, as well as chemical, occupational, and ionizing radiation exposures, and
promoting breastfeeding, healthy diet patterns, physical activity, and healthy vitamin D levels
[3,16,17][2][3][4]. Though
ouresearcher
s' ultimate aim is to highlight the complex interrelationships between the environmental determinants of breast cancer outcomes,
weresearchers will begin with a brief discussion of some individual risk and protective factors that are differentially distributed by race.
WeResearchers will then turn
ourresearchers' attention to the ways in which racial residential segregation ensures the inequitable distribution of these determinants at the population level.
1.1. Risk and Protective Factors for Breast Cancer
Alcohol is a known risk factor for breast cancer, and there is a dose response relationship between the two, with higher consumption corresponding to greater breast cancer risk
[18][5]. Though there are some conflicting results, there is limited evidence that high levels of alcohol consumption are associated with greater breast cancer risk among Black women than for any other racial group, and that higher levels of alcohol consumption increase risk of more aggressive subtypes of breast cancer, which also disproportionately impact Black women
[19,20][6][7].
The carcinogenic properties of tobacco smoke are well known. Though smoking is more strongly associated with cancers of the mouth, throat, and lungs, it is also a risk factor for breast cancer, particularly among women who started smoking at a young age or before carrying a pregnancy to term
[16,17][3][4]. Smoking is a risk factor for luminal breast cancer in particular, and there is some evidence that that risk is especially elevated for Black women
[21][8]. Further, another study found that though race and class were not a predictor of tobacco spending, vulnerable populations may experience worse breast cancer outcomes attributed to tobacco addiction than others
[22][9].
Environmental pollutants and toxic exposures account for between 7% and 19% of the world’s cancer cases
[16,23][3][10]. Exposures to toxic substances in the air, in the water, and through industrial agricultural pesticides contribute significantly to those figures, and there is burgeoning evidence that people of color are disproportionately impacted by those exposures
[24][11]. There is a broad array of chemical compounds in common household products that are of concern for breast cancer preventionists, including: bisphenol A (BPA); heavy metals, such as cadmium; polybrominated diphenyl ethers (PBDEs, AKA flame retardants); phthalates; alkylphenols; per- and polyfluoroalkyls (PFAs); pesticides; herbicides; solvents; aromatic amines; and parabens
[16][3]. There are various mechanisms by which these compounds may increase breast cancer risk, including endocrine disruption, promotion of tumor growth, and adverse epigenetic impacts. There is an emerging body of evidence to suggest that people of color are exposed to carcinogens in consumer products more frequently and in higher doses than whites
[25,26][12][13].
Ionizing radiation is a risk factor for various cancers, including breast cancer, and a common mechanism of exposure is in medical imaging, such as CT scans and X-Rays
[16][3]. One area of potential concern regarding racial disparities in breast cancer incidence is that, though Black women are less likely to receive testing for BRCA1/2 mutations, exposure to ionizing radiation through medical diagnostics was found in at least one study to increase risk of breast cancer by 90% among women with that mutation
[27][14].
Occupational factors may relate with breast cancer incidence and outcomes through numerous possible mechanisms, including: “chemical exposure; stress, including around job security and fair wages, threats or acts of sexual and physical violence, and lack of power to advocate for oneself; challenges with time and accommodation for breastfeeding; light-at-night exposure; and many other issues”
[16][3] (p. 204). These factors are of particular concern for women of color, who likely experience added stressors related to discrimination which may exacerbate their deleterious effects
[28,29][15][16].
Though the biological mechanisms linking breast cancer and diet are not well understood, the extant literature indicates that diet can function both as a risk factor and as a protective factor for breast cancer. Though some conflicting findings complicate the picture, the bulk of relevant studies show a positive association between consumption of red and processed meat and breast cancer
[30,31,32][17][18][19]. A recent meta-analysis also found a positive association between adolescent fat intake and breast cancer risk later in life
[33][20].
On the other hand, the preponderance of scientific evidence suggests that healthy eating habits, such as the Mediterranean diet, have a protective effect against breast cancer, and improve overall health
[16][3]. Due to the relatively lower quality of food sources in low-income communities of color, these diet-related risk factors may contribute significantly to disparities
[34][21].
A recent meta-analysis including 13,907 breast cancer cases from 27 studies on the relationship between breastfeeding and breast cancer found that there was an inverse association between the two, with longer breastfeeding duration related with lower relative risk compared with shorter duration (RR = 0.471, 95% CI, 0.368–0.602)
[35][22]. At the same time, one California based study found that Black women are less likely to exclusively breastfeed their infants than white women
[36][23]. This may be related to the fact that Black women are more likely to be employed in occupations with little to no flexibility or support services for new mothers, such as paid family leave, or experience other structural impediments related to discrimination that make breastfeeding more difficult.
Vigorous physical activity is another likely protective factor against breast cancer
[15,16][1][3]. These effects have been observed in studies that focused on Black women specifically
[37,38][24][25]. There are reasons to believe, however, that women of color may have lower levels of physical activity, and that this disparity in activity can be explained by the fact that they disproportionately experience structural and logistical barriers to physical activity, such as unsafe neighborhoods and a lack of access to greenspace
[39,40,41][26][27][28].
There is some evidence to suggest that vitamin D deficiency may be related to breast cancer disparities. In one study, low levels of vitamin D were associated with an elevated risk of breast cancer of 23% in Black women
[42][29]. Black women are at elevated risk of triple negative breast cancer, a particularly aggressive subtype, and are 10 times more likely to experience vitamin D deficiency than white women. This has led some researchers to suggest that some of that elevated risk may be attributable to vitamin D deficiency
[43][30].
1.2. The Role of Social Determinants
Williams et al. note that though both racial discrimination and racial breast cancer disparities are well documented, the mechanisms that link discrimination to disparities are not well understood
[3][2]. There are numerous plausible ways to begin tying the individual risk and protective factors together with discrimination into a coherent theoretical framework of racial breast cancer disparities.
WResearche
rs briefly outline and assess several possibilities below. Though they are presented separately here, the causal mechanisms are not truly separable. In reality, they are deeply interconnected.
At the individual level, the positive association between low SES and disease does not generally hold for breast cancer, but striking patterns emerge when examining incidence, severity of breast cancer subtype, race/ethnicity, and SES together
[3,44,45][2][31][32]. Discrimination can be understood as a stressor, and there is burgeoning research examining the impact of this specific type of stress as it relates to overall health. For example, experiences of discrimination have been linked to poorer health, and higher levels of biological and behavioral indicators of disease risk, such as inflammation, obesity, and smoking
[46,47][33][34].
Further, there is a growing body of literature demonstrating the neurobiological impacts of stress and trauma in early life, including methylation of the nuclear receptor 3C1, “the gene that codes for the glucocorticoid receptor on the hypothalamic-pituitary-adrenal axis”
[3][2] (p. 2141). Increased methylation in this nuclear receptor “represents a unique record of past adverse psychosocial experience,” and has been shown to be associated with numerous negative health outcomes, including breast cancer
[3][2] (p. 2141). Further, Geronimus et al. put forth the concept of weathering to encapsulate the effect of chronic stressors experienced by Black women in America
[28,29][15][16]. One example of this effect is reflected in a process of accelerated aging. By measuring telomere length, Geronimus and colleagues were able to determine that, compared with white women, Black women in their sample were biologically 7.5 years older on average. Another key concept related to stress across the life course is allostatic load, a biological measure of overall wear and tear on the body due to stress. As Williams et al. put it: “the concept of allostatic load has been used to capture the biological dysregulation across multiple physiological systems that result from the cumulative burden of repeated stressors”
[3][2] (p. 2141). Allostatic load has been found in multiple studies to be higher in people of color than among whites, and to be associated with poor health
[29][16].
Place of residence impacts breast cancer risk significantly. In a recent systematic review of 17 studies focusing on racial residential segregation and cancer disparities, 70% of studies showed a statistically significant association between segregation and health disparities. The authors state: “residing in segregated African-American areas was associated with higher odds of later-stage diagnosis of breast and lung cancers, higher mortality rates and lower survival rates from breast and lung cancers, and higher cumulative cancer risks associated with exposure to ambient air toxins”
[48][35] (p. 1195). Racial residential segregation also ensures that place-based risk factors are inequitably distributed by race. Given that individual and environmental exposures to breast cancer risk factors are impacted by place of residence, it is plausible that the disparities in incidence, care, and outcomes outlined by Landrine et al. are the result of the differential distribution of these exposures between segregated neighborhoods
[48][35]. Again, though these myriad factors are presented separately above, the upstream contributors to breast cancer disparities are not actually separable in reality.
The treatment above is not nearly complete, as it only alludes to a system of racial inequity, discrimination, and disparity that has deep historical roots and implications for nearly every domain in which health and welfare can be measured. Though there are many plausible mechanisms by which racial breast cancer disparities may arise, empirical investigation into those mechanism is in its early stages. For example, questions remain regarding the linkages between socioeconomic factors such as labor market opportunity, educational access, and credit access, indicators of health risk such as high allostatic load, and breast cancer risk. It is clear, however, that the impacts of both historical and ongoing racial discrimination are pervasive and highly interrelated. The high degree of interrelation between possible causes complicates matters for researchers seeking a parsimonious explanatory model, and interventionists seeking a small number of variables on which to focus programs. A clearer understanding of these causes and their interrelationships could aid in efforts to prevent breast cancer, and reduce disparities.
2. How Systems Thinking Can Identify Leverage Points
Arising out of the biological sciences, with applications across a broad range of fields, including economics, bioinformatics, meteorology, and others, systems theory is designed to model complex, interrelated networks of component subsystems. It may therefore prove useful in developing conceptual models to explain the complex and highly interrelated causes of breast cancer disparities, and to clarify why those disparities have thus far proven resistant to intervention. Most discussions of breast cancer disparities and other health inequities are confined to the healthcare or public health systems, and stop short of articulating the broader, more fundamental problem of systemic racism that contributes to inequities across a diverse array of outcomes. In the following section,
wresearche
rs will outline a framework for research and intervention in breast cancer disparities that is explicitly informed by an acknowledgement of the interlinked nature of disparities across multiple domains, including large-scale policy frameworks, local-level social and environmental factors, and the biological sequelae of discrimination
[4][36].
Much like racial breast cancer disparities themselves, the biological and social contributors underlying those disparities do not arise in a vacuum, but out of a long history of systemic racism and discrimination. There are significant disparities between Black and white Americans in the criminal justice system, including policing practices, incarceration, and jury participation, which have been linked not only to explicit legal discrimination, but also to implicit bias
[49][37]. Residential segregation, which persists today despite no longer having the force of law, is a significant driver of negative outcomes across numerous domains, including health, through its impact on educational access, employment opportunities, and exposure to environmental hazards
[7][38].
Systemic racism, and racial residential segregation in particular, also ensure the inequitable distribution of each of the risk and protective factors for breast cancer outlined above. For example, place of residence impacts the quality of available foods, exposure to environmental toxins through residential and occupational sources, likelihood of tobacco and alcohol use, and opportunities for physical activity (through access to safe outdoor spaces)
[15,22,35,40,41,42,50,51,52][1][9][22][27][28][29][39][40][41]. Further, though socioeconomic status, education, and employment status are each associated with both race and health outcomes, racial disparities in health outcomes persist even among Black Americans who are high SES, highly educated, and gainfully employed
[53][42]. This suggests that racism contributes independently to health disparities over and above the effects of these mediators.
There is increasing scholarly and scientific consensus that racism itself is a fundamental cause of health disparities. Yearby argues that the social determinants of health framework articulated in Healthy People 2020 is insufficient precisely because it does not give a primary place to racism as an upstream factor contributing to all other social determinants
[6][43]. Williams et al. provide an overview of the social context of breast cancer disparities among Black women, outlining the burgeoning body of research on the biological effects of discrimination which ultimately lead to elevated breast cancer risk
[3][2]. Reskin gives a broad outline of racial discrimination in American society, and makes a compelling case that each of the subsystems in which disparities exist are connected by the broader system of racism within which they are situated
[4][36]. According to her model, racial disparities in outcomes across numerous domains are the result of a single, integrated system of racial discrimination. She argues that ad hoc, sector-specific interventions to reduce disparities will likely fail to disrupt the systemic equilibrium that maintains inequity due to the systemic property of robustness. This literature indicates that, in order for racial breast cancer disparities to be eliminated, researchers, policymakers, and interventionists must address racism
as a system. In the following section,
wresearche
rs will introduce key concepts from systems theory, and discuss ways in which they can be used as tools to reconceptualize racial breast cancer disparities in light of the system of racial discrimination within which they arise.
3.1. Emergence
2.1. Emergence
Established systems exhibit the property of emergence. In a network of interrelated subsystems, the system as a whole will begin to exert an influence over and above the sum of the additive effects of those component subsystems. As Reskin explains, within the race discrimination system, this means that, in addition to the effects of individual experiences of discrimination and explicitly traceable instances of institutional racial bias, the system of racial discrimination itself exerts influence in maintaining the equilibrium of racial disparity
[4][36]. In other words, the interrelationships, or feedback loops, between factors like Adverse Childhood Experiences (ACEs) and childhood poverty, environmental exposures, and healthcare quality and access, produce effects all of their own, independent of those exerted by each subsystem individually. With regard to breast cancer specifically, it is possible that the interactive effects between tobacco use, poor diet, and low physical activity result in a greater increase in breast cancer risk than the additive effects of each risk factor would indicate. These behavioral risk factors may also interact with the effects of ACEs to further increase breast cancer risk. Given the disproportionate impact of each of these environmentally conditioned risk factors on Black women, breast cancer disparities between Black and white women may be understood as an emergent effect of systemic racism, operationalized through racial residential segregation. Such emergent effects are generally robust to changes in a given subsystem.
3.2. Robustness
2.2. Robustness
Established systems exhibit the property of robustness. This means that systems tend toward a state of equilibrium, and that that state of equilibrium will generally persist even in the face of significant disruptions within individual subsystems. The emergent effects of the system as a whole will ensure that the equilibrium is reasserted. In the case of racial disparities, this may explain the resistance of disparities to targeted interventions
[4][36]. For example, a bias-reduction intervention targeting the healthcare subsystem that does not address risk factors related to chemical exposure, physical activity, or ACEs may not disrupt systemic equilibrium sufficiently to significantly reduce disparities. This is, in part, due to the fact that each of these subsystems is related to the others in complex ways.
3.3. Complexity
2.3. Complexity
In systems theory, complexity is a formal designation which refers to a system in which there are a large number of component subsystems, and a large degree of interrelatedness between those component subsystems
[54][44]. In the study of racial disparities in breast cancer, there are several layers of complexity: (1) the complexity of the problem itself; (2) the complexity of the interconnected web of causes that contribute to the problem; and (3) the complexity of the tasks associated with researching and intervening on the problem. For example, as mentioned above, behavioral risk factors, such as smoking, diet, and physical activity, may interact with one another in producing breast cancer risk. At the same time, they may also interact with social environmental exposures (i.e., violence, ACEs), or residential and occupational exposures to endocrine disrupting substances. Each of these layers of complexity contributes to breast cancer risk and to the intransigence of disparity. As mentioned above, complexity can pose a challenge for researchers and for interventionists, as parsimonious explanatory models and predictable results are difficult to achieve.
3.4. Leverage Points
2.4. Leverage Points
In the context of complex systems, a full representation of all subsystems and all their interrelationships would yield a model insufficiently parsimonious for practical use.
WResearche
rs propose that researchers, policymakers, and interventionists should conduct a critical review of the complex relationships between upstream causes of breast cancer and racial breast cancer disparities to identify systemic leverage points. Drawing on Reskin,
weresearchers define leverage points as systemic nodes (or subsystems) where multiple other subsystems intersect, modifications in which have the potential to exert significantly amplified effects across the entire system
[4][36]. As long as interventions to reduce racial breast cancer disparities confine themselves to only one or two subsystems within the broader context of racial inequity, the systemic quality of robustness will likely result in the reassertion of equilibrium, and the persistence of disparity. In contrast, intervention approaches to racial breast cancer disparities that explicitly address leverage points have the potential to impact numerous subsystems simultaneously and disrupt that equilibrium.