Preterm birth is a leading cause of perinatal morbidity and mortality in the US. The increased risk of preterm delivery among African Americans, compared with White women, has not been explained by traditional risk factors. Taking a sociobiological approach, we examined whether neighborhood characteristics affected preterm delivery and whether these effects were mediated through or modified by cigarette smoking. In this case-control study, women were selected from a cohort of all births at UCSF hospital between 1980 and 1990. The sample included 417 African American and 1244 White women. Smoking was a risk factor for preterm delivery in logistic models, controlling for potential confounders. We explored neighborhood characteristics using multilevel logistic models, including Medi-Cal insurance to control individual SES. Among African American women, median census tract income, neighborhood unemployment and the change in African American residents from 1980-1990, were associated with preterm delivery. Among White women, only the change in unemployment in the census tract between 1980-1990 was associated with preterm delivery. All neighborhood variables had significant squared terms, indicating increased risk of preterm delivery at high and low levels of the variable. Cigarette smoking did not attenuate or modify the effect of the neighborhood variables. The larger impact of neighborhood characteristics on African American women may explain some of the racial/ethnic disparity in preterm delivery rates. Given widening wealth differentials and neighborhood deterioration in the US, interventions on African American health may be more effective if they focus on social conditions rather than on traditional behavioral risk factors.
Learning Objectives: 1. Participants will recognize the importance of neighborhood level socioeconomic characteristics in relation to preterm delivery 2. Participants will consider the importance of evaluating the interrelationship between socioeconomic and behavioral/biologic risk factors 3. Participants will better understand the importance of multilevel modeling
Keywords: Social Inequalities, Birth Outcomes
Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.