Elsevier

Social Science & Medicine

Volume 63, Issue 5, September 2006, Pages 1289-1303
Social Science & Medicine

Racial residential segregation and weight status among US adults

https://doi.org/10.1016/j.socscimed.2006.03.049Get rights and content

Abstract

While the segmentation of residential areas by race is well known to affect the social and economic well-being of the segregated minority group in the United States, the relationship between segregation and health has received less attention. This study examines the association between racial residential segregation, as measured by the isolation index, and individual weight status in US metropolitan areas. Multi-level, nationally representative data are used to consider the central hypothesis that segregation is positively associated with weight status among African Americans, a group that is hyper-segregated and disproportionately affected by unhealthy weight outcomes. Results show that among non-Hispanic blacks, higher racial isolation is positively associated with both a higher body mass index (BMI) and greater odds of being overweight, adjusting for multiple covariates, including measures of individual socioeconomic status. An increase of one standard deviation in the isolation index is associated with a 0.423 unit increase in BMI (p<0.01), and a 14% increase in the odds of being overweight (p<0.01). Among whites, there is no significant association between the isolation index and weight status. These findings suggest that in addition to differences among people, differences among places and, in particular, differences in the spatial organization of persons may be relevant to health policy and promotion efforts.

Introduction

In the US, residential segregation by race is well known to affect the social and economic well-being of the segregated minority group, particularly that of African Americans. Given relatively higher rates of black poverty, segregation acts to concentrate poverty in space, creating a distinctive milieu in segregated black neighborhoods, one that is characterized by factors such as housing deterioration, lower quality schools, inferior public services, and higher rates of crime, unemployment, families on public assistance, and single parenthood (Massey & Denton, 1993). These and other features associated with segregation have been linked to lower educational and occupational attainments, diminished prospects for socioeconomic mobility, and higher risks for individual social dislocations (e.g., teen pregnancy, welfare dependency, unemployment, criminal behavior, etc.) (Massey, Gross, & Eggers, 1991; Massey & Denton, 1993; Wilson, 1987; Wilson, 1997).

The relationship between racial segregation and health, however, has received less attention, and most empirical work has focused on mortality. Half a century ago, Yankauer (1950) found that black and white infant mortality rates are higher in the more segregated residential areas of New York City. Since then, other ecological studies have demonstrated that black infant mortality is positively associated with segregation at the metropolitan area level (LaVeist (1989), LaVeist (1993); Polednak, 1996b) and at the state level (Bird, 1995). Adult mortality has also been examined, with ecological studies showing that metropolitan area segregation is positively associated with black mortality rates (Collins & Williams, 1999; Hart, Kunitz, Sell, & Mukamel, 1998), as well as with the differential between black and white mortality rates (Polednak (1993), Polednak (1996a)). More recently, a multi-level study found that metropolitan area segregation is associated with increased odds of poor self-rated health among blacks (Subramanian, Acevedo-Garcia, & Osypuk, 2005). Additionally, a few studies have found that residence in a census tract with a higher proportion of blacks is associated with a higher risk of mortality (Leclere, Rogers, & Peters (1997), Leclere, Rogers, & Peters (1998); Jackson, Anderson, Johnson, & Sorlie, 2000).

Though not addressing segregation itself, there are now numerous studies demonstrating that neighborhood-level poverty is inversely associated with health (Pickett & Pearl, 2000). The segmentation of residential areas by race, however, is a key mechanism by which such neighborhood conditions are formed. While Black men in Harlem have a lower life expectancy than their counterparts in Bangladesh (McCord & Freeman, 1990), it is segregation that ultimately gives rise to the concentration of poverty in Harlem.

This study examines the relationship between racial segregation and weight status in US metropolitan areas. In the context of prior work supporting an association between segregation and mortality as well as self-rated health, weight status may function as an intermediary link to such general endpoints. Obesity, which has reached a prevalence of 30% among US adults (Hedley et al., 2004), is well known to be associated with various morbidities (Must et al., 1999), and may also increase the risk of mortality (Manson et al., 1995). There are many pathways by which segregation and the concentration of poverty could affect weight outcomes. For example, a poor consumer income base can negatively influence the quantity and quality of food retail options; a lower tax base can decrease the provision of parks and public recreational facilities; high crime rates can be prohibitive of outdoor activity; and impoverished, disenfranchised neighborhoods may be severely limited in the provision of medical services (Williams & Collins, 2001). Recent studies find that black neighborhoods may have considerably lower access to supermarkets and offer fewer healthy food selections relative to white neighborhoods (Lewis et al., 2005; Morland, Wing, Diez Roux, & Poole, 2001; Zenk et al., 2005).

Aside from the provision of such material services and amenities, the geographic isolation of segregation can translate into social isolation, creating a distinctive sociocultural milieu that may have more direct effects on individual well-being. Not only does racial segregation isolate poor blacks, it spatially concentrates the correlates of individual poverty. Social dislocations such as unemployment, teenage childbearing, and welfare dependency can become the norm rather than the exception, creating a dearth of normatively successful role models and perpetuating attitudes and behaviors that are non-conducive to the fulfillment of conventional expectations. Given that weight status is also correlated with poverty, the status of being overweight may also be fast approaching the norm in segregated enclaves, shifting local weight standards and mitigating the broader social stigma of obesity.

In a related vein, many have argued that alternative status and value schemes may develop to compensate for the disjuncture between dominant socioeconomic goals and structured opportunities. These schemes often reflect a deliberate inversion or disavowal of mainstream counterparts, and can be described as an oppositional subculture (Anderson, 1990; Massey, 1996; Shihadeh & Flynn, 1996; Wilson, 1997). As speech patterns, educational aspirations, and values concerning marriage and childbearing have grown increasingly distant from the mainstream in many poor, predominantly black neighborhoods (Massey & Denton, 1993), we might also expect divergent value schemes in the realm of health promotion efforts and weight preferences. Indeed, some have noted the rise of “health-related subcultures” in such settings (Fitzpatrick & Lagory, 2000). Furthermore, many studies have shown that blacks tend to be more tolerant than whites of heavier weight statuses, and differences in weight preferences or standards of ideal weight may contribute to racial differences in weight outcomes (Chang & Christakis, 2003; Kumanyika, 1998). Although a host of historical and sociocultural factors other than segregation may contribute to racial differences in attitudes with respect to weight and bodily aesthetics, racial isolation would still act to concentrate such standards, shifting the overarching community norm away from mainstream counterparts. Lastly, social isolation may also impede the diffusion of health-related information, and the multifarious nature of stress associated with living with concentrated poverty may precipitate both physiological and coping-type behavioral reactions that contribute to weight gain.1, 2

This study uses nationally representative data to examine the relationship between US metropolitan area segregation and individual weight outcomes. While some prior studies on segregation and health do utilize individual-level data (e.g., Leclere, Rogers, & Peters (1997), Leclere, Rogers, & Peters (1998); LaVeist, 2003; Robert & Reither, 2004; Subramanian et al., 2005), most employ ecological analyses, making the contribution of individual-level covariates difficult to ascertain and potentially overstating the role of contextual-level factors. For example, an association between segregation and health may be confounded by the fact that individual SES is correlated with both place of residence and health, reflecting a compositional effect of individual residents rather than an independent, contextual effect of segregation per se. It should be noted, however, that racial segregation likely affects health, in part, through a negative effect on individual socioeconomic attainments, which can ultimately influence health (Goldman, 2001). In this sense, factors such as individual SES would be on a causal pathway between segregation and health, and their adjustment may lead to an underestimation of the total effect of segregation. Hence, this study will use multi-level data to examine of the influence of segregation on weight status both with and without adjustments for individual-level SES.

The principal hypothesis of this study is that a greater degree of metropolitan area segregation will be associated with higher (heavier) weight status among blacks. For whites, a definitive expectation is more difficult to formulate. On the one hand, segregation (measured at the metropolitan level) may be beneficial to whites, especially poor whites, by isolating or buffering them from the correlates of black poverty (Massey, 1990). On the other hand, empirical results for health have been mixed. For example, while some prior studies do find an inverse association between segregation and mortality among whites (LaVeist (1989), LaVeist (1993)), others show that segregation is either unrelated or positively associated with mortality among whites (Bird, 1995; Collins & Williams, 1999; Jackson et al., 2000; Polednak, 1996a). Lastly, little is known about the relationship between segregation and more specific health outcomes such as weight status.

Massey and Denton (1988) conceptualize segregation along five distinct dimensions: unevenness, isolation, centralization, concentration, and clustering. This study will focus on the isolation dimension, which speaks most directly to proposed theoretical links between black social isolation and weight status. Moreover, an analysis of the isolation dimension is in keeping with several prior studies on segregation and health (Collins & Williams, 1999; Fang, Madhavan, Bosworth, & Alderman, 1998; Guest, Gunnar, & Hussey, 1998; Jackson et al., 2000; Leclere et al., 1997; Subramanian et al., 2005). Massey and Denton (1988) recommend the isolation index as the preferred measure of racial isolation. This concept refers to the degree to which minority group members come into contact primarily with other minority group members, indexing the percentage of neighborhood co-residents who are also minority group members.

Some studies rely on the dissimilarity index, a measure of unevenness in the distribution of blacks and whites. It has been noted, however, that the dissimilarity index is typically chosen arbitrarily or without justification, serving as a default measure with inadequate attention to relevant linkages between specific conceptualizations of segregation and the outcome of interest (Acevedo-Garcia & Lochner, 2003; Collins & Williams, 1999; Shihadeh & Flynn, 1996; Subramanian et al., 2005). Here, the isolation index is used to best capture potential pathways involving the effects of racial social isolation on weight status. Furthermore, in contrast to the dissimilarity index, the isolation index explicitly accounts for the relative sizes of the groups being compared (Massey & Denton, 1988), a property that is critical to this study. As a hypothetical illustration, consider a city where blacks are evenly (or perfectly) distributed, with each neighborhood containing an equal percentage of black residents. In this case, the dissimilarity index would be zero, registering no segregation. The isolation index, however, could still be high if blacks constitute a very high proportion of the total residents in the city, resulting in a relatively high percentage of blacks and little contact with whites in individual neighborhoods, despite even distribution. Lastly, in this regard, the isolation index is also effective in capturing suggested pathways involving the concentration of poverty per se (e.g., insufficient services and amenities). This is because segregation leads to concentrated poverty via an interaction between high minority poverty rates and a high proportion of minority group members in individual neighborhoods, and the isolation index speaks most directly to neighborhood minority proportion.

Section snippets

Data and measures

Individual-level data are from the 2000 Behavioral Risk Factor Surveillance System (BRFSS), a cross-sectional, nationally representative survey administered by the Centers for Disease Control and Prevention. The BRFSS collects data on a wide variety of health-related behaviors and outcomes in the US adult population and utilizes a probability sample of non-institutionalized adults for each state through random-digit-dial telephone surveys. The survey includes data on self-reported height and

Sample characteristics

Table 1 displays descriptive statistics for the BRFSS sample and the metropolitan areas. Whites have a lower mean BMI, percentage overweight, and percentage obese compared to blacks, but both groups have a mean BMI that is in the overweight range, and the majority of both groups is overweight. Both groups show relatively high proportions residing in the South because the sample is restricted to persons residing in metropolitan areas where blacks constitute at least 10% of the population. For

Discussion

To date there have been a limited number of studies on racial residential segregation and health, and most prior work has focused on mortality. This study narrows the scope to weight status, a more specific health outcome that disproportionately affects the most segregated minority group in the US—African Americans—and functions as a potential risk factor for mortality (Hedley et al., 2004; Massey & Denton, 1989). Findings show that for non-Hispanic blacks, racial isolation is positively

Acknowledgements

This research was supported by a Career Development Award (K12-HD-043459) from the National Institutes of Health, National Institute of Child Health and Human Development, the Measy Foundation, and the University of Pennsylvania and Cheney University EXPORT Center of Excellence for Inner City Health. I thank Shiriki Kumanyika for insightful discussions on this topic, Dawei Xie and Michael Elliot for consultation on the analyses, and Jason Schnittker and Theodore Iwashyna for helpful comments on

References (48)

  • V.W. Chang et al.

    Income disparities in body mass index and obesity in the United States, 1971–2002

    Archives of Internal Medicine

    (2005)
  • C. Collins et al.

    Segregation and mortality: The deadly effects of racism? Sociological Forum

    (1999)
  • K. Fitzpatrick et al.

    Unhealthy places: The ecology of risk in the urban landscape

    (2000)
  • N. Goldman

    Social inequalities in health: Disentangling the underlying mechanisms

    Annals of the New York Academy of Sciences

    (2001)
  • A.M. Guest et al.

    The ecology of race and socioeconomic distress: Infant and working-age mortality in Chicago

    Demography

    (1998)
  • K.D. Hart et al.

    Metropolitan governance, residential segregation, and mortality among African Americans

    American Journal of Public Health

    (1998)
  • A.A. Hedley et al.

    Prevalence of overweight and obesity among us children, adolescents, and adults, 1999–2002

    JAMA

    (2004)
  • S.A. Jackson et al.

    The relation of residential segregation to all-cause mortality: A study in black and white

    American Journal of Public Health

    (2000)
  • S.K. Kumanyika

    Obesity in African Americans: Biobehavioral consequences of culture

    Ethnicity and Disease

    (1998)
  • T.A. LaVeist

    Linking residential segregation to the infant-mortality race disparity in U.S. Cities

    Social Science Research

    (1989)
  • T.A. LaVeist

    Segregation, poverty and empowerment: Health consequences for African-Americans

    The Milbank Quarterly

    (1993)
  • T.A. LaVeist

    Racial segregation and longevity among African-Americans: An individual level analysis

    Health Services Research

    (2003)
  • F.B. Leclere et al.

    Ethnicity and mortality in the United States: Individual and community correlates

    Social Forces

    (1997)
  • F.B. Leclere et al.

    Neighborhood social context and racial differences in women's heart disease mortality

    Journal of Health and Social Behavior

    (1998)
  • Cited by (164)

    View all citing articles on Scopus
    View full text