Income Segregation and Health: Evidence from New York City

Background/Purpose: Over the past 30 years, the places where rich and poor Americans live have become increasingly segregated. Americans on the extreme ends of the income distribution are now more likely to be isolated from people who are not as poor or as rich as they are than they have been at any point in the last 3 decades (Taylor & Fry, 2012). 

Prior studies have focused on national, state-level, or city-level trends in residential income segregation; to my knowledge, no empirical studies have been undertaken that examine within-city variation in isolation of the poor. Further, few studies of income segregation have looked at health and healthcare usage as outcomes, while no studies have examined racial heterogeneity in associations between residential income segregation and health within a city. Using recent data from the New York City Longitudinal Study of Wellbeing (n = 1,800), I investigate the relationship between neighborhood-level income segregation and health/healthcare usage in contemporary New York City, both within and across racial/ethnic groups. This holds the potential to illuminate previously unexplored patterns of urban residential isolation as well as to inform discussions of the salience of neighborhood characteristics on health.  

Methods: Ordinary Least Squares regressions were used to generate all coefficients. Regression analyses were conducted using two measures of health/healthcare usage (fair/poor self-rated health and having an unmet health need) as dependent variables and income segregation scores (constructed using data from the 2012 American Community Survey) as key independent variables. Regressions were run for each dependent variable separately, and both univariate and multivariate models (using racial segregation scores, household size, gender, borough of residence, race of head of household, education level of head of household, and income-to-needs ratios of households as covariates) were run for each. In order to examine differences in the association between income isolation and outcomes by racial or ethnic group, both univariate and multivariate models were run for each outcome, stratifying by race/ethnic group.

Results: A 1 point increase in a neighborhood’s low-income segregation score was associated with a 7 percentage point increase in the likelihood of households within that neighborhood experiencing fair or poor health, net of other covariates (p < 0.1). Increases in low-income segregation were only statistically significantly associated with having an unmet health need among white respondents (p < 0.05), though there is evidence that segregation of low-income residents within a neighborhood may be associated with lower chances of having health needs go unmet for black and Hispanic residents.

Conclusions and Implications: Segregation of the poor in the neighborhoods of New York City is associated with worsened self-reported health, regardless of race, ethnicity, racial isolation of black households, or income-to-needs ratios. Somewhat surprisingly, there is also evidence that higher levels of income isolation are associated with lower chances of having unmet medical, dental, or general health needs, though only for black and Hispanic respondents. This may have important implications for the targeting of interventions aimed at mitigating health disparities, especially among racial minorities and poor whites.