2

The Need for a Systems Approach to Address Structural Racism

The workshop’s keynote speakers both focused their remarks on addressing the need to dismantle structural racism through a systematic approach. David Williams, Florence and Laura Norman professor of public health at the Harvard T.H. Chan School of Public Health, discussed the current disparities that exist between White Americans and people of color and the factors that perpetuate them. Brian Smedley, chief of psychology in the public interest and acting chief diversity officer at the American Psychological Association, argued for the importance of place as a way to frame race and focused his presentation on actionable opportunities to demonstrably improve conditions for health equity. This chapter reviews the systemic racial inequities in the United States, including the negative health effects these disparities have, and then reinforces the concept that “place matters,” describing the various ways that geography of opportunity and segregation have impacted populations of color throughout history in the United States.

SYSTEMIC RACIAL INEQUITIES

Challenges with equity begin early in life, stated Williams. Though there has been progress made over time, many challenges still persist. As an example, he highlighted racial differences in infant mortality over the last few decades. Rates for both races have improved over time, but even today, Black babies are still more than twice as likely to die as White babies (Ely and Driscoll, 2019). Trained as a sociologist, Williams learned that socioeconomic status (SES) is a central determinant of the distribution of valuable resources in society, and he has found large racial and

ethnic differences that paint a broader picture. Household income data from 2018 shows large differences across races: For every dollar made in a White household, Hispanic American households make $0.73 and Black American and Native American households make $0.59 (Semega et al., 2019). This gap is nearly identical to the income gap following the antipoverty programs and civil rights initiatives of the 1960s and 1970s, Williams said. Income only captures the flow of resources into a household, and this does not account for the economic reserves of families, resulting in understating the economic differences across races. For that, he explained, we need to incorporate aspects such as wealth, home equity, savings, stocks, and bonds. The latest data on these differences is even more striking. For every dollar of wealth that White Americans have, Black Americans have $0.10, Hispanic Americans have $0.12, and other races have $0.38 (Dettling et al., 2017). What this means, Williams said, is that right now during the COVID-19 pandemic, we are all experiencing the same storm but in very different boats. When he began his career, he explained, many people thought that income and education would determine wealth, and once those were controlled for, race as a factor would not be influential, but this has not been the case, Williams said. He presented a graph showing infant mortality by race controlling for the mother’s level of education (see Figure 2-1): Across the board, as the mother’s education

level increases, the chance of the baby dying before age 1 declines. He noted, however, the towering excess risk for Black mothers. Even those with the highest education level have infant mortality rates more than twice that of other racial groups.

To determine what other factors are at play influencing these disparities, Williams asked if racism could constitute a critical piece of the puzzle. Specifically, he defined racism as an organized system that categorizes, devalues, and differentially allocates opportunities and resources. This system is typically guided by an ideology in which some population groups are regarded as inferior, leading to the development of negative attitudes and prejudice to certain groups and discrimination by individuals and social institutions.

Residential Segregation

As an example of structural institutional racism, Williams discussed residential segregation by race, which has had pervasive negative effects on health and development. It is a legacy of systemic racism of the past, he said, but the structures that remain in place still produce very unequal outcomes. He compared segregation to a burglar who slips in unnoticed at night and over time steals valuables such as quality schools, safe housing, transportation, and access to health care and healthy foods. Providing empirical evidence to support this statement, Williams noted a study that looked at 171 of the largest cities in the United States and found that because of segregation, there is not one city where White Americans live in equal conditions to Black Americans (Sampson and Wilson, 1995). To measure the impacts of segregation a different way, he shared the Child Opportunity Index, which has 24 items that collectively capture the opportunities available to children in the United States (diversitydatakids.org, 2020). In looking at the 100 largest metropolitan areas, researchers found that 66 percent of all Black children, 58 percent of Latino children, and 53 percent of Native American children are living in very low- or low-opportunity neighborhoods. Conversely, almost two-thirds of White and Asian children are in high- or very high-opportunity neighborhoods, which leads to large racial/ethnic differences in SES. Williams presented another study from a Harvard economist showing that if segregation were eliminated, the differences between Black and White American populations in earnings, high school graduation rate, and unemployment would be erased entirely (Cutler and Glaeser, 1997). In an intergenerational study that looked at boys born at the same household income level, researchers found that Black boys come from families that have lower incomes than White boys in 99 percent of census tracts in America (Chetty et al., 2018). They did find this gap narrowed in low-poverty neighborhoods with low

levels of racial bias and a high presence of fathers, but fewer than 5 percent of Black children grow up in these areas.

These racial inequities in SES are not reflective of a broken system, Williams stated. It is a carefully crafted system that is working exactly as it was designed, successfully implementing social policies, many of which are rooted in racism. He offered several research goals for efforts to redesign this system to more effectively address root causes:

• Routinely capture data on geographic location of study participants.
• Link neighborhood context to biomarkers and other outcomes.
• Enhance understanding of how structural racism, especially in longitudinal studies, is associated with changes in biomarkers and risk and resilience factors in children over time.

Toxic Stress

In addition to structural and systemic racism, Williams also discussed the interpersonal racism that many people of color are subjected to throughout their daily lives. This ongoing discrimination, which he described as small indignities or microaggressions, can include being treated with less courtesy and respect than others, receiving poorer service, or people thinking you are unintelligent, scary, or inferior. These indicators can be predictive of other health implications, he noted, from sleep duration and quality to obesity to the risk of breast cancer. Research also increasingly shows it can have negative effects on children. A systematic review found a number of studies that document discrimination experienced by the mother during pregnancy predicting low birthweight in the baby (Priest et al., 2013). Another study that Williams shared examined Mexican immigrant parents and found that experience of racial bias by both the mother and father led to worse outcomes in children (Gassman-Pines, 2015).

Again highlighting the intergenerational connections, Williams presented a small study performed in Connecticut of mother-child pairs that linked increased maternal discrimination while pregnant to increased inflammation in the child from 4 to 9 years old (Condon, 2019). This goes beyond behavioral influences, he emphasized, and demonstrates that even underlying biological processes can be affected by everyday discrimination. These biological changes and impacts are not limited to the person who directly experiences the discrimination. Williams introduced the concept of epigenetics,¹ noting that stressors in early life and adulthood can be passed on to future generations. While this has not been fully studied in the context

___________________

¹ Epigenetics refers to the study of how behavior and environment can affect gene expression. For more information, visit https://www.cdc.gov/genomics/disease/epigenetics.htm.

of race and ethnicity, he said, there are important established patterns that demand attention. There is evidence suggesting that exposure to childhood abuse (sexual or physical) and intimate partner violence causes epigenetic changes in children. Similarly, he highlighted the Dutch famine study in the 1940s that documented the effects of inadequate nutrition among pregnant women at that time. Recent evidence now shows the grandchildren of those mothers have poorer health, and people prenatally exposed to famine showed epigenetic changes that were still evident six decades later. Similar findings have been seen in studying Holocaust witnesses and epigenetic changes in their offspring. Taken together, Williams said, we have only just begun to understand the full impact that these exposures may have on health. He called for more research to understand the long-term impacts of the historical trauma of slavery, Jim Crow laws, and colonial abuses of indigenous people, as well as the impacts of present-day discrimination.

Consequences of Racism and Toxic Stress

Research has also shown that in addition to higher levels of racial discrimination, Black American and Hispanic American populations encounter higher levels of other economic, psychosocial, physical, and environmental stressors (Williams, 2018). Recent research also shows that added burdens for young people of these races, such as seeing someone from your own race detained, arrested, or shot by police, lead to increased posttraumatic stress disorder symptoms and depressive symptoms. Williams highlighted some of his recent work that found every police shooting of an unarmed Black male was associated with worsening health for the entire Black population in the state in which the shooting occurred for the following 3 months (Bor, 2018). The White population in the study was unaffected. The consequence for all of these collective exposures is often referred to as “accelerated aging” or “biological weathering,” Williams explained. When living in an environment with adverse exposures, your biological age not only measures how long you have been alive, it measures the degree to which you have been compromised physiologically by these exposures. He shared a few disease examples with data from the Centers for Disease Control and Prevention, showing that African Americans² are at higher risk for a broad range of chronic conditions at much younger ages than White Americans. According to 2015 data from the Behavioral Risk Factor Surveillance System (BRFSS), high blood pressure rates are higher for African Americans in every age group (Centers for Disease Control and Prevention, 2015). He noted a similar picture for diabetes. Some studies

___________________

² The BRFSS uses the term African American to include persons having origins in any of the Black racial groups of Africa.

examining telomere length³ have even suggested that among middle-aged adults, African Americans are chronologically the same age but physiologically 8 to 10 years older than White Americans because of this constant exposure to stress and adversity.

Understanding Negative Stereotypes in Culture

To truly address these consequences, Williams advocated for understanding how racism has become embedded within our culture and how the negative stereotypes develop. He described a database of American culture that was developed with millions of words from magazines and articles, trying to assess how often people have heard words paired together (Verhaeghen, Aikman, and Van Gulick, 2011). When the word “Black” was searched for, the words most commonly found paired with it were poor, violent, religious, lazy, cheerful, and dangerous. Conversely, when the word “White” was searched, the most commonly paired words were wealthy, progressive, conventional, stubborn, successful, and educated. This demonstrates the predominantly negative stereotypes given to Black Americans, Williams said, which can lead to unconscious discrimination and implicit bias against different racial groups. As an example, he shared a study of children suspended in preschool, and found that compared to White preschoolers, Black preschoolers are 3.6 times as likely to receive one or more suspensions (Gilliam et al., 2016). In addition to preschool, this bias is seen in medicine and, Williams noted, can even be internalized by those who are the target of bias. When a stigma of inferiority is activated for African Americans under experimental conditions, for example, their exam performance is adversely affected (Steele, 1997).⁴ Even just being asked to write down their race can result in worsened outcomes.

Solutions to Address Racial Inequities

These are all problems of systemic and structural biases, said Williams, but there are things that can be done to address them. He first highlighted ways to reduce prejudice through focusing on identifying and tackling implicit bias. While implicit biases are deeply ingrained habits, they can be replaced through multiple strategies. In one study that implemented a multifaceted intervention across 12 weeks, participants were presented with a bias education and training program with the goal of evoking concern about implicit bias and training people to eliminate these biases.

___________________

³ Telomeres are DNA-protein structures that protect the genome. Each time a gene copies itself, telomeres get shorter. Thus, telomere length can be a sign of aging in cells.

⁴ The study refers to the population as African American.

Following the intervention, non-Black adults were able to increase their awareness of their bias, increase their concerns about the effects of bias, and implement multiple strategies to combat bias (Devine et al., 2012). These substantial reductions in bias remained evident 3 months later, showing that longer-term bias reduction is possible. In addition to reducing biases, Williams also called for structural competence, which goes beyond simple diversity training. Racism that is deeply embedded in our culture does not just influence people at the individual level, he said: it shapes our social institutions and policies. Effectively addressing bias at this level requires identifying and dismantling institutional legacies and social consequences. Policy change across multiple domains will be necessary to reduce the negative impacts of implicit bias, he commented. As a strategy to achieve this, Williams suggested creating communities of opportunity, to “minimize, neutralize, and dismantle the systems of racism that create inequities in health.” To do this, he offered several methods outlined in Box 2-1 and explored in further detail below.

Invest in Early Childhood Education

Williams shared an example of positive outcomes through investment in early childhood education demonstrated by the Perry Preschool Study, a preschool program designed for disadvantaged African American children in the 1960s. One evaluation of the program studied 123 Black American children living in low-income households who participated. Follow-up analysis nearly 40 years later found that those in the program had improved high school graduation rates; higher levels of income, employment, savings, and home ownership; and had fewer arrests for violent, property, and drug crimes (Heckman and Karapakula, 2019). Additional intergenerational

follow-up has shown that the children of those enrolled in the program had improved outcomes such as higher levels of education and employment and lower levels of crime (Muennig et al., 2009). Effects even exist when those in the intervention group lived in worse neighborhoods than the controls. Williams also noted that benefits were seen in siblings as well, not just the children enrolled in the program, which means that the entire family can experience the ripple effect.

Utilize Resilience Resources

Williams also called for utilizing resilience resources to reduce the negative effects of discrimination on health. Even as young as age 20, research has shown considerable biological dysregulation in young African Americans⁵ who experienced discrimination as teenagers (Brody et al., 2014). This included higher cortisol levels, blood pressure, inflammation, and weight, all of which were linked to their experiences and encounters with discrimination. There was a buffering effect to this association, though, Williams added that higher levels of social support or religious engagement can reduce the effects of discriminatory experiences on the mental health of African Americans.

Another example of understanding resilience and its important effects can be seen through a study of First Nations communities in Canada who were experiencing high levels of adolescent suicide. Researchers found that the communities that were challenging the government for increased land rights, self-governance, and control over their own education and health care services had a significantly lower risk of youth suicide (Chandler and Lalonde, 1998). Williams also added that there was a strong dose-response relationship such that the more protests, advocacy opportunities, and empowerment a community had, the lower the prevalence of suicide was. “We need to think of resilience not only as a characteristic of individuals, but as a property of environments and social policy,” he stated.

Reduce Child Poverty

One way to build this resilience is by reducing child poverty, said Williams. He presented a United Nations Children’s Fund report ranking industrialized countries in their rates of childhood poverty (Adamson, 2012). In particular, he highlighted a chart that showed child poverty rates before and after taxes and transfers take place in various countries (see Figure 2-2). For Ireland, 42 percent of children are born into poverty, but after taking into account supportive government taxes and transfers, this

___________________

⁵ The study refers to the population as African American.

number decreases to just 8 percent. By comparison, the United States starts at 25 percent, but after accounting for all tax and transfer programs, the number decreases just one point to 24 percent. This leaves a lot of opportunity for Americans, Williams stated.

Williams also called attention to a recent National Academies of Sciences, Engineering, and Medicine report, A Roadmap to Reducing Child Poverty, which listed several of the report’s recommended strategies to reduce child poverty by 50 percent within 10 years (National Academies of Sciences, Engineering, and Medicine, 2019) by a number of policy initiatives:

• Earned-income tax credits
• Expand housing vouchers and SNAP
• Child allowances
• Child and dependent care tax credit
• Work-based programs
• Government and tax transfers
• Public health insurance

The report reviews the research on linkages between child poverty and child well-being. It also analyzes the poverty-reducing effects of major assistance programs directed at children and families.

Enhance Income and Employment Opportunities for Youth and Adults and Strengthen Families

Communities of opportunity also need better income and employment opportunities for youth and adults, continued Williams. He emphasized the connection between increased economic resources and improved health, even without the presence of an explicit health intervention. One study found, for example, that raising the minimum wage in a state by just one dollar produces decreases in low birthweight of up to 1 to 2 percent and decreases in post-neonatal mortality of 4 percent in the United States (Komro et al., 2016). If all states had done this in 2014, Williams said, there would have been 2,790 fewer low-birthweight babies and 518 fewer post-neonatal deaths that year.

Williams also commented that marriage rates increase when average male earnings increase, but when male unemployment increases, marriage rates decline. Learning from the military, which provides its members access to family housing and day care centers, he said that active-duty service promotes marriage and leads to greater stability in the marriage (Teachman, 2009). These effects are more pronounced for Black families compared to White families and in part is attributed to the well-defined career ladder in the army and the race-blind environment of the military compared to civilian work (Teachman and Tedrow, 2008).

Improve Neighborhoods and Housing Conditions

Finally, Williams introduced purpose-built communities as a strategy to build communities of opportunity, a possible means for addressing all of these challenges simultaneously. In 1995, for example, East Lake Meadows was a high-crime African American⁶ public housing project in Georgia where only 13 percent of adults were gainfully employed and the elementary school was one of the worst in the state. The neighborhood had 18 times the national crime rate, and neighborhood high school kids graduated at rates of only 30 percent. To build a purpose-built community, they used a variety of integrative strategies such as mixed-income housing and cradle-to-college education investment and also focused on community engagement and philanthropy. Today, the schools are top performers, 75 percent of able-bodied individuals are employed, and there has been a 73 percent reduction in crime.

___________________

⁶ The study refers to the population as African American.

Challenges and Barriers

Knowing all of these solutions and seeing improvements in various cities around the country, Williams said the question remains: What is holding us back? One of the first places he called for action is raising the levels of awareness about the extent of the problem. He acknowledged the role of COVID-19 in calling attention to many health inequities but said that we need to develop and mobilize the political will to address them, which will require developing empathy to fuel the political will. He elaborated on this, saying that many have called out the empathy gap between races. Brain-imaging researchers have found consistent racial in-group biases that produce different empathic responses in the brain when viewing suffering of one’s own race compared relative to responses regarding members of another race. There is a stronger empathic response to the pain of one’s own race compared to other-race individuals (Han, 2018). This gap actually begins to develop at quite a young age, Williams explained. Studies have shown no racial bias at age 5, but children start showing a weak bias by age 7, and by age 10, there is evidence of strong, reliable racial bias (Dore et al., 2014). This bias at such a young age poses a real challenge for designing strategies to overcome this empathy gap. In conclusion, he said, “we need to move from a commitment of equality of giving everyone the same thing to a commitment of equity and giving everyone that which they need to thrive.”

SOCIAL DETERMINANTS OF HEALTH: PLACE MATTERS

Smedley called attention to residential segregation, which Williams mentioned previously, and said that it set the stage for many racial inequities in other areas besides health. Segregation has faded as a key objective of racial justice work, he noted, but it is increasingly returning to the focus as a key structural determinant of racial inequality. “In the wake of the COVID-19 pandemic and unrest across communities due to systemic racism, we are at an inflection point where we can choose to move forward with a more equitable nation or we can continue with the status quo,” he said.

Smedley explained that there is a geography of opportunity related to where people live, and it is abundantly clear from research that a disproportionate cluster of health risks and a lack of resources predominate in the spaces where people of color live, all of which are tied to policy decisions. While we have made progress on segregation, Smedley said, there is a considerable distance to go. He used housing as an example, noting that there are laws related to civil rights and fair housing, but many people of color still encounter persistent housing discrimination. Some believe the myth that racial segregation arises because people of the same race prefer to live

together or experience economic limitations, he suggested, but evidence tells a different story. Government at every level has systematically segregated minority groups in access to housing through tactics such as racial zoning, the segregation of public housing, the subsidies offered to developers after World War II to make suburbs available only to White Americans, and the proffering of tax exemptions to institutions to encourage the enforcement of segregation. All of these factors taken together throughout history explain why segregation still exists today, he explained.

Smedley then reviewed the concept of redlining, which he said tilted the economic playing field and deeply disadvantaged populations of color. Redlining was a practice of the former Federal Home Ownership Loan Corporation that designated minority neighborhoods as “risky” for financial investment and color-coded such neighborhoods in red. This stymied investment in these communities, but its impact persists: Three-quarters of neighborhoods that were redlined on government maps 80 years ago continue to struggle economically today. Cities with more redlined neighborhoods also have significant economic inequality (Mitchell and Franco, 2018). Conversely, 91 percent of areas that were classified as the best places to live remain middle- to upper-class areas and 85 percent are predominantly White. This created a legacy of not just segregation, he explained, but deliberate economic disenfranchisement and marginalization, which explains the patterns of deep poverty often associated with segregation. To provide context for racial residential segregation in the United States, Smedley presented data comparing South Africa during the apartheid era in 1991 to several U.S. cities in 2010 (see Figure 2-3). These data feature a measure known as the dissimilarity index, which is an indication of the percentage of White and Black people that need to move in order to achieve integration in a defined geographic area. Many cities were just behind South Africa in the percentage of people that would need to move to achieve integration. Smedley was hopeful that the forthcoming 2020 census data will show progress in desegregating the nation’s cities, but he emphasized that this has been a lasting pattern for generations.

Negative Effects of Segregation on Health and Development

Emphasizing why this matters, Smedley presented several negative effects of segregation on human health and development by negatively shaping social determinants of health. Segregation restricts socioeconomic mobility by creating public schools that are more under-resourced, fewer employment opportunities, and smaller value appreciations on real estate in minority neighborhoods. African Americans⁷ are five times less likely than

___________________

⁷ The population as referred to by the speaker.

White Americans to live near supermarkets and more likely to live near fast food and liquor stores, he said. From the perspective of the physical environment, Black and Hispanic neighborhoods have fewer parks and green spaces, and low-income communities and communities of color are more likely to be exposed to environmental hazards. A United Church of Christ study in 2007 found that 56 percent of residents living near commercial hazardous waste facilities were people of color, even though they were only about one-third of the U.S. population at the time (Bullard et al., 2007).

It also concentrates people of color at all levels of income in poor neighborhoods, Smedley said, and he shared multiple examples from different cities. Poor Black Americans and Hispanic Americans are more likely than poor White Americans to live in medium- and high-poverty census tracts. Some might argue that these concentrations are due to income or educational differences, he said, but even controlling for income, the racial differences are stark (see Figure 2-4). This is a direct result of public policy, he said, as people of color were deliberately sorted into specific neighborhoods that then experienced economic disinvestment.

In metro Cleveland, the same results were found, and there were stark racial disparities between who has access to which neighborhoods. Black and Hispanic kids were much more likely to live in neighborhoods characterized by moderately and highly concentrated poverty than White children, even when comparing children whose families had incomes below the

poverty level: Poor White children are much more likely than poor children of color to live in a low-poverty neighborhood. He presented a report called “History Matters” in which Cuyahoga County PlaceMatters and several other partners reviewed old homeownership loan corporation redlining maps, demonstrating the sanctioned disinvestment in overwhelmingly Black neighborhoods (Reece et al., 2015). Comparing those maps from the 1940s with the maps of today, he pointed out the strong geographic correlation with redlined communities then and poor health outcomes and low life expectancy today. This was also recreated with several other cities around the country, demonstrating the widespread, pervasive effects of such housing policies. To make things worse, there is also often a hidden “poverty tax,” he explained. Residents of poor communities pay more for the exact same consumer products than those in higher-income neighborhoods. This can include anything like furniture, insurance, bank fees, and groceries.

A System in Need of Solutions

Knowing how policies and historical exclusion have impacted people of different races throughout history, Smedley offered possible solutions to help begin to solve these longstanding problems. First, there should be a focus on prevention, particularly where people live, work, play, and study. We also need a sustained, long-term policy agenda, he argued. It took hundreds of years to create these conditions, and they are multifaceted and

layered. They will not be changed overnight, but the science is very clear and should be used to inform policy and practice. Finally, there is a need for multiple strategies across sectors. Smedley suggested exploring both place-based strategies such as investment in communities simultaneously with people-based strategies such as investing in early childhood education or increasing housing mobility options. He reviewed several different types of these strategies, outlined in Box 2-2.

As an example of a people-based strategy, Smedley described a 10-year randomized controlled trial supported by the U.S. Department of Housing and Urban Development to test what happens when low-income families move from high-poverty to low-poverty neighborhoods. Beginning in the 1990s, nearly 5,000 families were tracked. Some were put in the experimental group, receiving both rental assistance and help to find rental units in low-poverty neighborhoods, and some in control groups in which they received assistance but no special help finding units in low-poverty neighborhoods. The study was conducted in five cities: Baltimore, Boston, Chicago, Los Angeles, and New York City. The long-term evaluation of this effort found the following results for families that moved to lower-poverty neighborhoods:

• Parents had lower rates of obesity and depression and reported lower levels of stress.
• Children who moved before age 13 experienced improved college attendance rates and earnings.
• Children were less likely to become single parents later in life.
• Children who were less than 13 when their families moved grew up to have an annual income in their mid-20s, 31 percent higher on average than those in the control group.

While most results were positive, moving to low-poverty areas when children were older than 13 actually had negative long-term impacts on the children, likely because of the disruptions at an important time in their life. Smedley highlighted the importance of social cohesion in many low-income neighborhoods and said its protective effects could be considered in any policies. He also noted that housing mobility is not a panacea, and communities have learned that wholesale dislocation is truly ineffective. Forcibly moving low-income communities has many detrimental consequences, so efforts could be made to strengthen communities and not split them apart. He highlighted the importance of accompanying people-based interventions such as housing mobility with place-based investments to stimulate economic activity and improve community conditions so that people who choose to stay where they are have better opportunities.

He concluded with a few examples further emphasizing that place and race matter for health. In both New Orleans and Bernalillo County, New Mexico, there were vast differences in life expectancy across census tracts—as high as 28 years of difference—even though these communities are just a few miles apart. These differences are not explained by health care access, genetics, or behavior, he said. Instead, differences in investment and other structural inequities across the neighborhoods expose the true root of these differences. Finally, Smedley offered a quote from Mary Bassett, former commissioner of health for New York City: “Antiracism is … a collective healing without which our nation will remain painfully and inequitably divided, corroding opportunity, spirits, and bodies alike” (The Huffington Post, 2017).