Neighbourhood Socioeconomic Disadvantage and Severe Maternal Morbidity: Secondary Analysis of a Prospective Cohort.

Abstract

To examine whether neighbourhood socioeconomic disadvantage, as measured by the Area Deprivation Index (ADI) in early pregnancy, was associated with severe maternal morbidity (SMM) at delivery hospitalisation. A prospective multi-site observational cohort. A secondary analysis of the Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-To-Be study (nuMoM2b) across eight United States (US) sites from 2010 to 2013. Participant residential address in the first trimester was geocoded at the US census-tract level to calculate the ADI, a standardised metric of neighbourhood socioeconomic disadvantage. We used modified Poisson regression with robust error variance and adjusted for individual-level covariates: age, pre-pregnancy body mass index, chronic hypertension, and pregestational diabetes to examine the association between the ADI [modelled in quartiles from the least (quartile 1, Q1, reference) to the most (Q4) disadvantage] and SMM. Differences in the association between ADI and SMM by self-reported race and ethnicity as a social construct were evaluated with effect modification via an interaction term in the adjusted model. SMM, based on the US Centers for Disease Control and Prevention definition, and secondarily, SMM without transfusion. Among 9588 nulliparas, 2.3% (n = 221) experienced any SMM and 0.5% (n = 48) experienced non-transfusion SMM. Individuals living in the most disadvantaged neighbourhoods (Q4) were more likely to experience SMM compared with those in the least disadvantaged neighbourhoods (Q1) (3.4% vs. 2.1%; aRR 1.73; 95% CI: 1.17, 2.58). This association was also significant for non-transfusion SMM (1.0% vs. 0.3%; aRR: 2.82; 95% CI 1.15, 6.93). Individuals who self-identified as non-Hispanic Black were more likely to experience SMM than non-Hispanic White individuals (3.9% vs. 2.1%; p < 0.001). There was no evidence of effect modification by self-reported race and ethnicity (interaction p > 0.05). Nulliparous pregnant individuals who lived in the most disadvantaged US neighbourhoods were at increased risk of experiencing SMM. Known racial and ethnic disparities in SMM may be related to adverse neighbourhood-level social determinants.