Residential Segregation and Health Outcomes: The Role of Health-Promoting Community Organizations in Urban Neighborhoods

Abstract

Research has long established that racial/ethnic minority residential segregation is related to a variety of social problems, including poor health and health care outcomes. Yet, little research in this area has attempted to tease out the mechanisms underlying this association. Furthermore, within this work, few studies address this relationship for racial/ethnic minority groups outside of the Black/White differences. In this dissertation, I argue that community organizations and service providers in neighborhoods provide important local sites which can improve residents' access to key health-related resources. I put this forward as a mechanism which can link segregation to health and health care outcomes, in that racial/ethnic minority neighborhoods are less likely to have a variety of such establishments compared to their White counterparts. I test this mechanism through a series of three empirical studies. In the first, using a nation-wide health data set combined with metropolitan area measures, I examine the association between racial residential segregation and poor self-rated health. I include the three largest racial/ethnic minority groups in the U.S. - Blacks, Latinos, and Asians. I find that the segregation of each group is related to a greater likelihood of poor self-rated health. However, for Black respondents, this is partially accounted for by economic variables at the metropolitan level, and for Asian respondents, is fully accounted for by recent immigration. In the second study, I analyze the distribution of health-related facilities by segregation status using a nation-wide Census data source. I find that racial residential segregation for Blacks, and to a lesser extent Latinos and Asians, is related to a decrease in the density of such facilities, including food sources, fitness facilities, pharmacies, a number of health care organizations, civic society, and social services. The inclusion of socio-economic indicators removes this effect for most of the health care organization outcomes, across all three groups. In the third and final study, I examine this full pathway using the case of the Phoenix urbanized area. Using a 2013-2014 survey of families about their children's health care utilization combined with area-level Census measures, I test where families are able to obtain care for their children, if any location, in light of what is physically proximate to them in space, and how this may be patterned or constrained by the segregation status of their local neighborhood. I find that Latino and Native American segregation is related to a decreased number of health care facilities. Further, the lack of such facilities is related to a decreased likelihood of families being able to utilize a physician's office as their regular source of care, versus a clinic or health care center. However, for those who do not have a regular source of care, or who utilize an emergency room, this seems to be patterned more by economic considerations, chiefly whether or not the child has health insurance. Further, the distribution of health care organizations also accounts for a sizable portion of the effect of segregation on health care utilization choices. As a whole, this dissertation demonstrates that segregation, for a number of racial/ethnic minority groups, is related to a lower density of a number of health-related organizations, both in a nationwide analysis, as well as through an extended case study of the Phoenix area. Moreover, using the Phoenix case, I find that the lack of such facilities is consequential, and that at least for certain health care providers, patterns where families are able to go for their children's health care needs. This suggests that the distribution of health-related organizations and service providers across communities may serve as an important explanatory mechanism to understand the association between segregation and a variety of health and health care outcomes.