Association of Neighborhood-Level Social Vulnerability with Preterm Birth in King County, Washington 2018-2022

Background: Preterm birth (PTB) affects one in ten US infants and remains to be a leading cause of infant morbidity, mortality, and life course adverse health outcomes. In King County, WA, the rate of PTB is higher among Black (12.1%) and American Indian/Alaska Native (15.6%) birthing people, as well as those without a college degree (12.0%), when compared to King County’s average (9.3%). Recent studies suggest that social determinants of health play a significant role in PTB risk. However, no prior study examined associations between social disadvantage and PTB risk in King County, as well as potential roles of race in this relationship. This study examined the association between social vulnerability index (SVI), a tool developed by CDC to characterize social vulnerability, and PTB in King County, WA. It also examined the role of race as a potential effect modifier of the association. Methods: This cross-sectional study used singleton birth data in King County, WA, from 2018 to 2022. Exposure was social vulnerability, defined using SVI quartiles (Q1, Q2, Q3, and Q4), while outcome was PTB, defined as birth before 37 weeks of gestation. Participants were excluded if they had missing exposure, outcome, or census tract information. The study analyzed data from 89,320 live singleton births in King County, WA. Three logistic regression models were fit to estimate odds ratios (ORs) and corresponding 95% confidence intervals (CIs). Model 1 (minimally adjusted) included exposure and outcome variables, along with birthing person’s race/ethnicity and King County region. Model 2 (partially adjusted) included Model 1 variables and variables for infant sex, birthing person’s age, educational attainment, marital status, and Women, Infant and Children (WIC) program status. Model 3 (fully adjusted) included Model 2 variables and variables for Kotelchuck index, parity, diabetes status, hypertension status, PTB history, and smoking status. Additionally, we fit logistic regression models stratified by race/ethnicity (Asian, Black, Hispanic, Multiple, Other/Unknown, and White) to compare associations across racial/ethnic groups. A likelihood ratio test was used to determine the statistical significance of differences in associations across race/ethnic groups.
Results: The majority (59%) of study participants were in the 25-34 years age group, non-White (54%) and South King County residents (43%). Over half (58%) held Bachelor’s degree or higher, and the majority (78%) were married. About 54% were parous with no history of preterm birth (97%). Participants in the highest SVI quartile were younger (18% aged 18-24 years), unmarried (36%), enrolled in WIC (34%), racial minorities (69%), and South King County residents (75%), compared with participants in the lowest three SVI quartiles. PTB prevalence increased across increasing SVI quartiles: 6.2%, 6.8%, 7.5%, and 9.2% in the first, second, third, and fourth quartiles, respectively. In general, PTB prevalence was highest in Q4, and lowest in Q1, for all racial/ethnic groups except Black birthing people. In Model 1, the odds ratios of PTB risk for the upper three SVI quartiles, compared with Q1, were as follows: Q2 (OR:1.04, 95% CI: 0.96-1.13), Q3 (OR:1.08, 95% CI:1.00-1.17) and Q4 (OR:1.17, 95% CI:1.08-1.27) (trend p-value <0.001). In Model 2, the OR estimates for Q2 (OR:1.02, 95% CI: 0.94-1.11), Q3 (OR:1.04, 95% CI:0.96-1.13) and Q4 (OR:1.07, 95% CI:0.99-1.17) were closer to the null and became statistically insignificant (trend p-value=0.093). Findings from Model 3 were largely similar to findings from Model 2 (trend p-value=0.107). In stratified analysis using Model 2, the ORs across SVI quartiles suggested a positive SVI-PTB risk relationship among Asian residents (Q2 OR: 1.05, 95% CI: 0.90-1.23; Q3 OR: 1.09, 95% CI: 0.93-1.28; Q4 OR: 1.15, 95% CI: 0.96-1.37) (trend p-value=0.112), whereas among Black residents it suggested an inverse relationship (Q2 OR: 0.84, 95% CI: 0.56-1.27; Q3 OR: 0.79, 95% CI: 0.54-1.16; Q4 OR: 0.74, 95% CI: 0.51-1.07) (trend p-value=0.094). Model 3 findings of the stratified analyses were in general similar to Model 2 findings of the stratified analyses. The p-values for the likelihood ratio tests assessing the interaction between SVI quartiles and race/ethnicity on PTB risk were 0.011 for Model 1, 0.072 for Model 2, and 0.108 for Model 3.
Discussion: We found potential positive associations between SVI and PTB risk that varied across racial/ethnic groups (suggestive positive associations among Asians and suggestive inverse associations among Blacks). The estimates relating SVI and PTB were not statistically significant after adjusting for other variables in our models. This study suggests a potential relationship between SVI and PTB with significant public health implications. Future studies in this area are warranted to inform targeted program or policy interventions that address maternal and child health equity in the population.