The Sick Side of Town

Residential Segregation

Residential Segregation

Residential segregation is another lens through which researchers are exploring the relationship between disparities in neighborhood physical and social environments and disparities in health outcomes. This research often aligns with the interests of environmental justice, a concept and social movement focused on evidence showing that certain communities are subject to a disproportionate burden of pollution and contamination.31

Environmental Justice

Rachel Morello-Frosch’s explicit interest in31 advancing environmental justice is evident in research that shows how historical and contemporary racial inequality and discrimination shapes disparities in heat-related land cover.32 This 2013 study found that racial/ethnic minorities, particularly Hispanics and Asians, were more likely to live in areas with heat risk-related land cover, such as asphalt and concrete, than whites. While only 29 percent of whites lived in areas with no tree canopy and mostly covered with concrete or asphalt, 50 percent of Hispanics and 54 percent of Asians lived in areas with heat risk-related land cover.33 Figure 8 describes these data.

These findings have important implications for health. Extreme heat events are responsible for one in five natural hazard deaths in the United States.34 As climate change brings more extreme heat to cities, heat waves are expected to become more intense, more frequent, and longer lasting.35 Surfaces like asphalt and concrete contribute to high ground temperatures and tend to concentrate in areas called “urban heat islands” while trees lower surface and air temperature. In addition to Jesdale and Morello-Frosch’s findings, several other studies document racial/ethnic disparities in urban tree cover and find that racial/ethnic minorities are more likely to live in neighborhoods with lower tree coverage.36

In a paper focused on advancing theory and methods in residential segregation-based research, Rachel Morello-Frosch and Russ Lopez (2006) write, “examining [environmental justice] issues through the lens of racial residential segregation can offer new insights into the junctures of the political economy of social inequality with discrimination, environmental degradation, and health.” Indeed, research about residential segregation has led to findings that connect it to disparities in adverse birth outcomes, and increased risk of chronic disease, as well as increased risk of exposure to environmental hazards.37

Figure 8 shows the proportion of urban residents living in areas with no tree canopy, high proportions of impervious surface, and both conditions, by race/ethnicity, segregation, housing tenure, and poverty.

Violent injury

Recent research from Haas Institute Diversity and Health Disparities research member and UC Berkeley Epidemiology and Community Health Sciences Associate Professor Amani Nuru-Jeter, along with Joshua Berezin, Sara Gale, Maureen Lahiff, Colette Auerswald, and Harrison Alter, finds that lower levels of neighborhood-level residential segregation are associated with lower levels of violent injury.38

Based in Oakland, the study used violent injury data from the detailed trauma registry at the Alameda County Medical Center’s Highland Campus to geo-code injury locations and link these locations to census block groups and determine the number of violent injuries per block group. The research team calculated residential segregation using a diversity composite measure that ranges from zero to one, with one representing maximum diversity (all groups are represented in equal proportions in the geographic area).39 Figure 9 presents these data.

Results from this study are consistent with a large body of research showing that lower levels of segregation (or higher levels of diversity, as presented in this study) are associated with lower levels of violence, particularly for predominantly Black and predominantly Hispanic neighborhoods. This finding challenges the idea that high neighborhood diversity is a destabilizing force and suggests that it may actually be protective.

Scholars debate the effects of neighborhood diversity on trust. Some, like Robert Putnam, argue that increased diversity leads people to “hunker down” and withdraw from the community—which means that Nuru-Jeter’s results could be interpreted as evidence that lower rates of violence in highly diverse neighborhoods are the result of decreased social interactions. Other experts argue that increased diversity fosters neighborhood trust and is a driver behind lower rates of violence in those neighborhoods. Nuru-Jeter’s research implies that decreasing segregation, or increasing resident diversity, creates positive neighborhood conditions like lower rates of violence, which are good for all residents’ health.

Income Inequality & Mortality

Residential segregation is not only racial, it is also economic. William J. Wilson’s seminal work The Truly Disadvantaged presents evidence of the way inner-city communities became more economically segregated as a result of the decline in low-skilled manufacturing jobs, which contributed to high rates of unemployment and white-flight.40 As economic segregation increased, so did income inequality.

Income inequality is a significant predictor of population health: greater income inequality is associated with poorer population health status.41 But the effects of income inequality cannot be separated from the role of race and ethnicity, two long-established correlates of mortality.42 Amani Nuru-Jeter, Chyvette Williams, and Thomas LaViest combined Wilson’s perspective with Massey and Denton’s argument that residential segregation is fundamentally racial to explain why the association between mortality and income inequality is different for Blacks and whites.43

Figure 9 includes two map diagrams which compares the variables on violent injuries and composite diversity (represented by darkening shades) and higher levels of injury and composite diversity (darker colors).

In a national, cross-sectional ecological study, Nuru-Jeter et al. examined the association between income inequality and mortality among Blacks and whites separately.44 Their data showed that racial segregation explained the income inequality and mortality association for Blacks but not for whites, suggesting that racial segregation is potentially more harmful for Blacks than income inequality. Furthermore, the study showed that the association between income inequality and mortality was protective for whites, meaning white mortality rates decrease with increasing levels of income inequality. Taken together, the study’s findings suggest that racial segregation is bad for Blacks and explains the higher mortality rates associated with income inequality among Blacks.

Neighborhood residents may find it easy to observe how residential segregation unequally distributes trees or the frequency of violence, but they might not be able to see how residential segregation is "getting under the skin." In a study published in 2011, Mahasin Mujahid, with Kiarri Kershaw, Ana Diez Roux, Sarah Burgard, Lynda Lisbeth, and Amy Schultz, found evidence of a strong correlation between residential segregation and hypertension disparities.

Figure 10 includes a graph of the predicted probability of hypertension for Blacks and whites by level of neighborhood poverty at low (10th percentile) and high (90th percentile) levels of segregation.

The research team sought to identify environments where the Black-white hypertension disparities were smallest or nonexistent in order to learn if context perpetuated Black residents’ unequal burden of hypertension. They used a national set of health data to identify people with hypertension and assessed residential segregation at the census tract level. The analysis found that Black people living in low segregation areas were 1.67 times more likely to have hypertension than whites, and Black people living in high segregation areas were 3.57 times more likely than whites to have hypertension.45

This startling difference is evidence of a connection between neighborhood context and a specific measure of health. While a biological mechanism for the relationship between hypertension and place is still unknown, the findings offer valuable insights. It is hypothesized that residential segregation leads to health disparities because it concentrates Black people into high poverty areas, and living in poverty is associated with many neighborhood characteristics that are bad for health. This includes decreased neighborhood safety, limited access to healthy foods and recreational resources, and lower levels of educational attainment. However, Mujahid’s results suggest that neighborhood poverty alone cannot account for the stark Black and white hypertension differences.

In Figure 10, we see how the researchers consider segregation, neighborhood poverty, and race differences in hypertension together. Living in a low segregated area is associated with lower predictability of hypertension for Black people across low to high poverty neighborhoods than living in a high segregated area. The data also illustrate how Black and white differences in hypertension diminish as neighborhood poverty increases because predicted probability of hypertension in whites increases as neighborhood poverty increases. Therefore, when the three factors—segregation, neighborhood poverty, and differences in hypertension—are considered together, race differences in hypertension were greatest in segregated, low-poverty areas.

As the Haas Institute Diversity and Health Disparities research cluster members’ research findings show, residential segregation shapes racial disparities in health outcomes, neighborhood social conditions, and environmental features. As one scholar argues, “racial residential segregation and social inequality are fundamental causes of racial/ ethnic disparities in neighborhood physical and social environments.”46 This research suggests that addressing and reducing residential segregation benefits the public’s health and may reduce racial health disparities.

  • 31. NRDC, 2017
  • 32. Jesdale et al., 2013
  • 33. Jesdale et al., 2013
  • 34. Borden & Cutter, 2008
  • 35. Meehl & Telbaldi, 2004
  • 36. Heynen et al. 2006; Landry & Chakraborty 2009; Lowry et al. 2012; Ogneva-Himmelberger et al. 2009; Perkins & Heynen 2004; Zhang et al. 2008
  • 37. Williams & Collins, 1995; Bailey et al., 2017; Braveman, 2017
  • 38. Berezin et al., 2017
  • 39. Berezin et al., 2017
  • 40. Wilson, 1987
  • 41. Nuru-Jeter et al., 2014; Backlund, E., et al., 2007; Kaplan, G. A., et al., 1996
  • 42. Lynch & Kaplan, 1997; Cooper et al., 2001; Deaton & Lubotsky, 2003
  • 43. Massey & Denton, 1993; Nuru-Jeter et al., 2014
  • 44. Nuru-Jeter et al., 2014
  • 45. Kershaw et al., 2011
  • 46. Casey et al., 2017