Cross-sectional analysis of two social determinants of health in California cities: racial/ethnic and geographic disparities

Abstract

Objective: To study the magnitude and direction of city-level racial and ethnic differences in poverty and education to characterise health equity and social determinants of health in California cities.

Design: We used data from the American Community Survey, United States Census Bureau, 2006-2010, and calculated differences in the prevalence of poverty and low educational attainment in adults by race/ethnicity and by census tracts within California cities. For race/ethnicity comparisons, when the referent group (p₂) to calculate the difference (p₁-p₂) was the non-Hispanic White population (considered a historically advantaged group), a positive difference was considered a health inequity. Differences with a non-White reference group were considered health disparities.

Setting: Cities of the State of California, USA.

Results: Within-city differences in the prevalence of poverty and low educational attainment disfavoured Black and Latinos compared with Whites in over 78% of the cities. Compared with Whites, the median within-city poverty difference was 7.0% for Latinos and 6.2% for Blacks. For education, median within-city difference was 26.6% for Latinos compared with Whites. In a small, but not negligible proportion of cities, historically disadvantaged race/ethnicity groups had better social determinants of health outcomes than Whites. The median difference between the highest and lowest census tracts within cities was 14.3% for poverty and 15.7% for low educational attainment. Overall city poverty rate was weakly, but positively correlated with within-city racial/ethnic differences.

Conclusions: Disparities and inequities are widespread in California. Local health departments can use these findings to partner with cities in their jurisdiction and design strategies to reduce racial, ethnic and geographic differences in economic and educational outcomes. These analytic methods could be used in an ongoing surveillance system to monitor these determinants of health.

Keywords: Epidemiology; Health policy; Public health.

Figure 1

Distribution of within-city differences in (A) poverty rate and (B) low educational attainment for pairwise comparisons of California Whites, Blacks, Latinos and Asians, 2006–2010. *Considering Whites as a socially advantaged reference group, the differences that favour Whites are considered inequities. The legends show the per cent of cities in which the p₂ race/ethnicity group has a better outcome and the per cent of cities in which the outcome is statistically significant (p<0.1). For instance, ‘Black-Asian (77%, 31%)’ indicates that 77% of the cities in which the comparison is possible have a better outcome for the Asian group and 31% of those cities have a significantly better outcome.

Figure 2

(A) Within-city poverty rate differences and overall city poverty rate and (B) low educational attainment and overall city low educational attainment, California cities, 2006–2010. r is the Pearson correlation coefficient between within-city differences and the overall city value. *Considering Whites as a socially advantaged reference group, the differences that favour Whites are considered inequities.

Figure 3

Distribution of within-city differences in the highest and lowest census tract rates for educational attainment and poverty, California, 2006–2010.