Original Research
Obstetrics
Associations of neighborhood-level racial residential segregation with adverse pregnancy outcomes

Presented in part in an oral format at the Society for Maternal-Fetal Medicine Pregnancy Meeting, Dallas, TX, Feb. 3, 2018.
https://doi.org/10.1016/j.ajog.2018.01.022Get rights and content

Background

Previous analyses utilizing birth certificate data have shown environmental factors such as racial residential segregation may contribute to disparities in adverse pregnancy outcomes. However, birth certificate data are ill equipped to reliably differentiate among small for gestational age, spontaneous preterm birth, and medically indicated preterm birth.

Objective

We sought to utilize data from electronic medical records to determine whether residential segregation among Black women is associated with an increased risk of adverse pregnancy outcomes.

Study Design

The study population was composed of 4770 non-Hispanic Black women who delivered during the years 2009 through 2013 at a single urban medical center. Addresses were geocoded at the level of census tract, and this tract was used to determine the degree of residential segregation for an individual’s neighborhood. Residential segregation was measured using the Gi* statistic, a z-score that measures the extent to which the neighborhood racial composition deviates from the composition of the larger surrounding area. The Gi* statistic z-scores were categorized as follows: low (z < 0), medium (z = 0-1.96), and high (z > 1.96). Adverse pregnancy outcomes included overall preterm birth, spontaneous preterm birth, medically indicated preterm birth, and small for gestational age. Hierarchical logistic regression models accounting for clustering by census tract and repeated births among mothers were used to estimate odds ratios of adverse pregnancy outcomes associated with segregation.

Results

In high segregation areas, the prevalence of overall preterm birth was significantly higher than that in low segregation areas (15.5% vs 10.7%, respectively; P < .001). Likewise, the prevalence of spontaneous preterm birth and medically indicated preterm birth were higher in high (9.5% and 6.0%) vs low (6.2% and 4.6%) segregation neighborhoods (P < .001 and P = .046, respectively). The associations of high segregation with overall preterm birth (odds ratio, 1.31; 95% confidence interval, 1.02–1.69) and spontaneous preterm birth (odds ratio, 1.37; 95% confidence interval, 1.02–1.85) remained significant with adjustment for neighborhood poverty, insurance status, parity, and maternal medical conditions.

Conclusion

Among non-Hispanic Black women in an urban area, high levels of segregation were independently associated with the higher odds of spontaneous preterm birth. These findings highlight one aspect of social determinants (ie, segregation) through which adverse pregnancy outcomes may be influenced and points to a potential target for intervention.

Introduction

Compared to non-Hispanic White women, non-Hispanic Black women are more likely to experience adverse pregnancy outcomes such as preterm birth (PTB), defined as birth <37 weeks’ gestation.1, 2 This disparity in PTB is directly related to disproportionately high levels of infant mortality among Black infants.1, 2 Because differences in individual-level maternal characteristics have been unable to account fully for differences in birth outcomes, investigators have considered the environmental factors that may contribute to obstetric disparities.2, 3, 4, 5

Aspects of the social environment, particularly racial residential segregation, appear to be associated with a higher risk of experiencing adverse pregnancy outcomes.6, 7, 8, 9, 10, 11 The studies that have demonstrated these associations largely have used administrative data and have been limited to outcomes such as low birthweight (applied to newborns weighing <2500 g) or overall rates of PTB. These studies, however, have not been able to adequately assess the underlying reasons for PTB, the main subtypes being spontaneous or medically indicated etiologies. Moreover, low birthweight, which includes neonates born preterm as well as those born due to fetal growth restriction, is a grouping that ignores the differences in underlying etiologies.12 Accordingly, the relationship between segregation and small-for-gestational-age birth, which is defined as birthweight <10th percentile for their gestational age and that better reflects abnormal intrauterine growth, has not been well evaluated.12, 13, 14 Furthermore, birth certificate data often lack information regarding maternal health conditions, limiting the ability of these studies to analyze or control for individual-level variations and confounders, and this has been identified as an area needing further investigation by previous authors.11

In this study, we utilized electronic medical records, and the more granular data they provide, to examine associations of segregation among non-Hispanic Black women with overall PTB, spontaneous PTB, medically indicated PTB, and small-for-gestational-age birth.

Section snippets

Study population

Patient data from Northwestern University Memorial Hospital in Chicago, IL, were obtained from the Northwestern Medicine Enterprise Data Warehouse system. We extracted birth records for all non-Hispanic Black women who delivered singleton gestations from Jan. 1, 2009, through Dec. 31, 2013. We limited our sample to women who had a home address within the Chicago-Joliet-Naperville, IL-IN-WI metropolitan statistical area (MSA). Non-Hispanic White women were excluded from this study due to

Results

During the study period from 2009 through 2013, there were 5638 singleton births to 5114 non-Hispanic Black women at Northwestern Medicine Prentice Women’s Hospital. Of these, there were 5235 births to 4770 women with valid address information within the Chicago MSA and were thus considered for analysis. We then excluded 61 births to 51 women (1%) with missing or implausible clinical or demographic information. The final analytical sample included 5174 births to 4719 unique women who lived in

Comment

The current analysis has used electronic medical records to demonstrate the association between high levels of neighborhood-level racial segregation and the higher odds of overall PTB. Further analysis of type of PTB demonstrated a stronger relationship between spontaneous PTB than medically indicated PTB. These findings support those of prior studies,6, 7, 9 and add to those findings by differentiating between spontaneous and medically indicated PTB.

We did not find a significant association

References (39)

Cited by (0)

This study was funded internally and funding sources had no involvement in design, analysis, writing, or submission of this article.

The authors report no conflict of interest.

Cite this article as: Salow AD, Pool LR, Grobman WA, et al. Associations of neighborhood-level racial residential segregation with adverse pregnancy outcomes. Am J Obstet Gynecol 2018;218:351.e1-7.

View full text