UC Berkeley

The purpose of this dissertation is to explore the relationship between neighborhood quality and preterm birth (birth before 37-weeks’ gestation, PTB) among Non-Hispanic Black and African American women (hereafter, Black women). Three interrelated studies assess different dimensions of this relationship within the population of Black women who lived in Oakland, California, at the time of giving birth (2007-2011). The risk of PTB among Black women in the United States remains 1.5 times that of Non-Hispanic White women (hereafter, White women), a gap which has persisted without much improvement for decades. Because this inequity remains largely unexplained by differences in individual-level risk factors, researchers have begun to look farther upstream. The physical and social environments in which Black women live are one source of risk which is being increasingly examined.

Study 1 examines the overall relationship between neighborhood quality and PTB among Black women in Oakland (N=5549). First, the study maps neighborhood quality as defined by the California Healthy Places Index (HPI) and its eight component domains across 107 Oakland census tracts. Next, the study assesses the relationship between neighborhood quality and PTB using log-linear generalized estimating equation (GEE) models with exchangeable correlation structures. Fully adjusted models control for maternal confounding factors: parity, nativity, age, educational attainment for age, and low-income status. The study finds that living in a higher quality neighborhood is significantly associated with reduced risk of PTB among Black women across Oakland neighborhoods.

Study 2 examines the effect modification of relationship between neighborhood quality and preterm birth among Black women in Oakland (N=5549), by maternal characteristic. This study assesses each maternal characteristic as a potential multiplicative effect modifier through a series of log-linear GEE models with exchangeable correlation structures. The study finds significant effect modification by maternal participation the Special Supplemental Nutrition Program for Women, Infants and Children (WIC). The relationship between neighborhood quality and PTB is only significant among women who did not use WIC.

Study 3 examines potential mediation of the relationships identified in Studies 1 and 2 among Black women in Oakland (N=5343) by the number of prenatal care (PNC) visits, adequacy of PNC initiation, adequacy of PNC received services, and adequacy of PNC use (APNCU). The Baron and Kenny method is used to guide a series of models to identify the presence of mediation, and significance of mediation is tested through the calculation of a mediation percentage with non-parametric bootstrap-based 95% confidence intervals. The study finds that while adequacy of PNC initiation, adequacy of PNC received services, and APNCU are significant risk factors for PTB in this population, independent of neighborhood quality, there is no evidence of significant mediation of the relationship between neighborhood quality and PTB.

The findings of Studies 1, 2, and 3 can inform efforts in Oakland to improve neighborhood quality in order to have a long-term impact on risk of PTB among Black women. Building on these findings, future quantitative and qualitative research is warranted to understand in greater detail the specific pathways through which neighborhood quality influences Black women’s risk of PTB in Oakland. Additional research in other urban environments would enhance a broader understanding of the diverse relationships between place and PTB among Black women in the United States. Our studies contribute to the growing body of literature investigating how place may matter and exploring intervention opportunities to reduce Black women’s risk of PTB.