Article

A Model of Racial Residential History and Its Association with Self-Rated Health and Mortality Among Black and White Adults in the United States.

Department of Sociology, Georgia State University, Atlanta, Georgia, USA.
Sociological Spectrum (Impact Factor: 0.31). 07/2009; 29(4):443-466. DOI: 10.1080/02732170902904616
Source: PubMed

ABSTRACT We construct a dynamic racial residential history typology and examine its association with self-rated health and mortality among black and white adults. Data are from a national survey of U.S. adults, combined with census tract data from 1970-1990. Results show that racial disparities in health and mortality are explained by both neighborhood contextual and individual socioeconomic factors. Results suggest that living in an established black neighborhood or in an established interracial neighborhood may actually be protective of health, once neighborhood poverty is controlled. Examining the dynamic nature of neighborhoods contributes to an understanding of health disparities.

0 Followers
 · 
80 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: Racial disparities in obesity among women in the United States are substantial but the causes of these disparities are poorly understood. We examined changes in body mass index (BMI) trajectories for Black and White women as a function of neighborhood disadvantage and racial composition of the neighborhoods within which respondents are clustered. Using four waves of the Americans' Changing Lives (ACL) survey, we estimated multilevel models predicting BMI trajectories over a 16-year period. Even after controlling for individual-level socio-demographics, risk and protective factors, and baseline neighborhood disadvantage and racial composition, substantial racial disparities in BMI persisted at each time point, and widened over time (p < 0.05). Baseline neighborhood disadvantage is associated with BMI and marginally reduces racial disparities in BMI, but it does not predict BMI changes over time. However, without neighborhood-level variables, the BMI trajectory model is misspecified, highlighting the importance of including community factors in future research.
    Health & Place 09/2009; 16(2):191-8. DOI:10.1016/j.healthplace.2009.09.009 · 2.44 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The evidence linking socioeconomic environments and metabolic syndrome (MetS) has primarily been based on cross-sectional studies. This study prospectively examined the relationships between area-level socioeconomic position (SEP) and the incidence of MetS. A prospective cohort study design was employed involving 1,877 men and women aged 18+ living in metropolitan Adelaide, Australia, all free of MetS at baseline. Area-level SEP measures, derived from Census data, included proportion of residents completing a university education, and median household weekly income. MetS, defined according to International Diabetes Federation, was ascertained after an average of 3.6 years follow up. Associations between each area-level SEP measure and incident MetS were examined by Poisson regression Generalised Estimating Equations models. Interaction between area- and individual-level SEP variables was also tested. A total of 156 men (18.7%) and 153 women (13.1%) developed MetS. Each percentage increase in the proportion of residents with a university education corresponded to a 2% lower risk of developing MetS (age and sex-adjusted incidence risk ratio (RR) = 0.98; 95% confidence interval (CI) =0.97-0.99). This association persisted after adjustment for individual-level income, education, and health behaviours. There was no significant association between area-level income and incident MetS overall. For the high income participants, however, a one standard deviation increase in median household weekly income was associated with a 29% higher risk of developing MetS (Adjusted RR = 1.29; 95%CI = 1.04-1.60) CONCLUSIONS: While area-level education was independently and inversely associated with the risk of developing MetS, the association between area-level income and the MetS incidence was modified by individual-level income.
    BMC Public Health 07/2013; 13(1):681. DOI:10.1186/1471-2458-13-681 · 2.32 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Several well-established determinants of health are associated with premature mortality. Using data from the 2010 County Health Rankings, we describe the association of selected determinants of health with premature mortality among counties with broadly differing levels of income. County-level data on 3,139 US counties from the 2010 County Health Rankings were linked to county mortality data from the Centers for Disease Control and Prevention Compressed Mortality database. We divided counties into 3 groups, defined by sample median household income levels: low-income (≤25th percentile, $29,631), mid-income (25th-75th percentile, $29,631-$39,401), and high-income (≥75th percentile, ≥$39,401). We analyzed group differences in geographic, sociodemographic, racial/ethnic, health care, social, and behavioral factors. Stratified multivariable linear regression explored the associations of these health determinants with premature mortality for high- and low-income groups. The association between income and premature mortality was stronger among low-income counties than high-income counties. We found differences in the pattern of risk factors between high- and low-income groups. Significant geographic, sociodemographic, racial/ethnic, health care, social, and behavioral disparities exist among income groups. Geographic location and the percentages of adult smokers and adults with a college education were associated with premature mortality rates in US counties. These relationships varied in magnitude and significance across income groups. Our findings suggest that population health policies aimed at reducing mortality disparities require an understanding of the socioeconomic context within which modifiable variables exist.
    Preventing chronic disease 03/2012; 9:E75. DOI:10.5888/pcd9.110120 · 1.96 Impact Factor