The neighborhood contribution to black-white perinatal disparities: an example from two north Carolina counties, 1999-2001.

Previous studies of black-white disparities in perinatal outcomes have generally not controlled for both observed and unobserved neighborhood inequalities with models that compare only black and white women living in the same neighborhoods. Using 1999-2001 birth certificate data from 2 counties in North Carolina, the authors employed a hybrid fixed-effects approach to assess the total contribution of neighborhood factors to both absolute and relative racial disparities in low birth weight, preterm birth (PTB), and smallness for gestational age at term. Neighborhood factors made a notable contribution to racial disparities for PTB only, accounting for an additional 15% reduction in crude disparities beyond individual sociodemographic characteristics, which accounted for approximately 40% of racial disparities. The neighborhood contribution was greater for moderate PTB (32-36 weeks' gestation) than for very PTB (<32 weeks' gestation). A neighborhood deprivation index accounted for a smaller percentage of PTB disparities than the hybrid fixed-effects estimates, which suggests that measured socioeconomic deprivation does not account for all health-relevant neighborhood inequalities. Contemporaneous individual-level sociodemographic and neighborhood factors together explained one- to two-thirds of perinatal disparities. To fully explain racial disparities in perinatal outcomes, evaluation of other differential exposures (e.g., racism or wealth) and neighborhood factors across the life course may be necessary.