Aims: This study assessed the relationship between spatial social polarization measured by the index of the concentration of the extremes (ICE) and preterm birth (PTB) and infant mortality (IM) in New York City. A secondary aim was to examine the ICE measure in comparison to neighborhood poverty. Methods: The sample included singleton births to adult women in New York City, 2010–2014 ( n=532,806). Three ICE measures were employed at the census tract level: ICE − Income (persons in households in the bottom vs top 20th percentile of US annual household income), ICE −Race/Ethnicity (black non-Hispanic vs white non-Hispanic populations), and ICE – Income + Race/Ethnicity combined. Preterm birth was defined as birth before 37 weeks’ gestation. Infant mortality was defined as a death before one year of age. A two-level generalized linear model with random intercept was utilized adjusting for individual-level covariates. Results: Preterm birth prevalence was 7.1% and infant mortality rate was 3.4 per 1000 live births. Women who lived in areas with the least privilege were more likely to have a preterm birth or infant mortality as compared to women living in areas with the most privilege. After adjusting for covariates, this association remained for preterm birth (ICE – Income: Adjusted Odds Ratio (AOR) 1.16 (1.10–1.21); ICE – Race/Ethnicity: AOR 1.41 (1.34–1.49); ICE – Income + Race/Ethnicity: AOR 1.36 (1.29–1.43)) and IM (ICE – Race/Ethnicity (AOR 1.80 (1.43–2.28) and ICE – Income + Race/Ethnicity (AOR 1.54 (1.23–1.94)). High neighborhood poverty was associated with PTB only (AOR 1.09 (1.04–1.14). Conclusions: These results provide preliminary evidence for the use of the ICE measure in examining structural barriers to healthy birth outcomes.