Residential segregation and prenatal depression in a non-Hispanic Black and Hispanic cohort in North Carolina.

ABSTRACT

PURPOSE: Investigate residential segregation and prenatal depression in a non-Hispanic (NH) Black and Hispanic North Carolina pregnancy cohort. METHODS: Demographics, prenatal depression (Center for Epidemiological Studies Depression scale ≥16), and residence from the 2006-2009 Newborn Epigenetic Survey were linked to Census-tract levels of racial and economic segregation (Index of Concentration at the Extremes) from the American Community Survey 2005-2009 5-year estimates. Adjusted prevalence ratios (aPR) for prenatal depression compared living in Index of Concentration at the Extremes tertiles 1 and 2 (higher proportion NH Black or Hispanic and/or low income) to 3 (higher proportion NH white and/or high-income), accounting for neighborhood clustering, age, education, employment, parity, and marital status. RESULTS: Among the 773 survey participants (482 NH Black and 291 Hispanic), 35.7% and 27.2% of NH Black and Hispanic participants had prenatal depression, respectively. For NH Black participants, depression prevalence was 17% lower for tertile 1 versus 3 for the NH Black/white (aPR=0.83; 95% CI=0.62-1.10), low/high income (aPR=0.83; 95% CI=0.62-1.11), and low-income NH Black/high-income NH white (aPR=0.82; 95% CI=0.61-1.09) measures. For Hispanic participants, estimates were weaker in the opposite direction for the Hispanic/NH white (aPR=1.02; 95% CI=0.71-1.47), low/high income (aPr=1.14; (95% CI=0.76-1.69), and low-income Hispanic/high-income NH white (aPR=1.12; 95% CI=0.78-1.60) measures. CONCLUSIONS: Residential segregation's impact on prenatal depression may differ by race/ethnicity and level of segregation, but findings are imprecise due to small sample sizes. Longitudinal research spanning greater geographic areas is needed.