County-level racial prejudice and the black-white gap in infant health outcomes

Black mothers are 60 percent more likely than white mothers to have preterm births and twice as likely to have a baby with low birth weight. We examine whether these black-white gaps in birth outcomes are larger in counties with higher levels of racial prejudice.

We use data from the restricted-use natality files in the United States, which provide information on birth weight, gestation, and maternal characteristics for over 31 million births from 2002 to 2012, combined with county-level data measures of both explicit and implicit racial prejudice from Project Implicit from over a million individuals who took the Implicit Association Test during this same period. We compare counties that are one standard deviation above the mean (high prejudice) with those that are one standard deviation below the mean (low prejudice) in terms of their average level of racial prejudice.

The black-white gap in low birth weight is 14 percent larger in counties with high implicit racial prejudice compared to counties with low prejudice. The black-white gap in preterm births is 29 percent larger in the high prejudice counties. The gaps are even larger when we use explicit measures of racial prejudice with high prejudice counties having a black-white gap that is 22 percent larger for low birth weight and 36 percent larger for preterm births. These relationships do not appear to be biased by the way the prejudice sample is constructed, since the racial gap in birth outcomes is unrelated to other county-level biases such as those based on gender or sexual orientation.

The black-white gap in United States' birth outcomes is larger in those counties that have the highest levels of racial prejudice. This is true for both implicit and explicit racial prejudice, though the strength of the relationship is strongest for explicit racial prejudice.