Segregation and preterm birth: the effects of neighborhood racial composition in North Carolina.

Epidemiologic research suggests that racial segregation is associated with poor health among blacks in the United States (US). We used geocoded birth records and US census data to investigate whether neighborhood-level percent black is associated with preterm birth (PTB) for black and white women in two counties in the southern US, whether area-level deprivation modifies this association, and whether the association is influenced by the choice of geographic unit used to approximate a neighborhood. A 20%-point increase in tract-level percent black was found to be associated with increased PTB odds in white (OR=1.09, 95% CI: 1.01, 1.18) and black (OR=1.05, 95% CI: 0.99, 1.11) women. These small associations were similar to those observed in other US regions. Effects were robust to choice of neighborhood proxy and were not modified by area-level deprivation.