Child Self-Rated Health: The effect of longitudinal measures of neighborhood racial composition

The literature on neighborhood context and health suggests that the life-course processes involved in building trajectories of health are not adequately captured in cross-sectional analysis, which has been the empirical focus of much of the research in this area. In this study we use data from the U.S. Early Childhood Longitudinal Study – Kindergarten Cohort (ECLS-K) to examine the impact of longitudinal measures of neighborhood racial composition on child self-rated health between kindergarten and 8th grade. We employ two-level multilevel longitudinal logistic regression models to examine variation in the initial status and trajectories of child self-rated health over time. Since the ECLS-K tracked child mobility over time, we are able to model the impact of changes in neighborhood racial composition. We find that compared to children living in low white neighborhoods, children living in high white neighborhoods had higher odds for initial good health (OR=2.10; p<0.001). Children living in high black, Hispanic, and minority neighborhoods has significantly lower odds of initial good health (OR=0.58; p<0.001, OR=0.31; p<0.001 and OR=0.39; p<0.001, respectively). Hispanic children living in high minority neighborhoods have higher odds of good health status compared to those living in low minority neighborhoods (OR=0.34; p<0.01).