Geographic and Racial Disparities in Maternal Health in New Jersey

Background and Purpose
Maternal health impacts the well-being of women, children, and their families and communities. Severe maternal morbidity (SMM) includes potentially life-threatening or near-miss events that would have resulted in a maternal death during pregnancy, childbirth, or within 42 days of termination of pregnancy if not for significant medical intervention. Remarkably, SMM is a pressing issue in New Jersey despite having one of the lowest poverty rates and highest median incomes in the U.S. The state has the 4th highest maternal mortality rate in the nation, one of the highest maternal morbidity rates, and the rates continue to rise. Further, there are large SMM disparities within NJ by race yet the reasons of these disparities remain understudied. In addition, the overall state racial differences may mask geographic variation in the state. It is therefore unclear whether disparities are concentrated in particular communities or whether the racial disparities are driven by Black women residing disproportionately in communities with poorer maternal health outcomes for both White and Black women. We addressed this knowledge gap by investigating geographic variation across and within counties in racial disparities in SMM in NJ. We also examined the extent to which geography explains the racial disparities in the state overall.

Methods
We used data from New Jersey State Health Assessment Data system for hospital discharge records of all 463,366 women admitted for birth- and pregnancy-related reasons in 2008-2015. We identified women with SMM using International Classification of Diseases diagnosis and procedure codes in the discharge records, based on the Centers for Disease Control and Prevention SMM criteria. We used generalized linear models to estimate Black-White disparities (both risk differences and risk ratios) overall, by county, and in models that controlled for race-specific county poverty rate. We also used Blinder-Oaxaca decomposition to determine the extent to racial disparities in SMM are explained by county of residence.

Results
Black women admitted for childbirth were more likely to live in counties with higher overall SMM rates, and counties with high SMM rates had a larger proportion of births to Black women compared to counties with low SMM rates. There were statistically significant racial disparities in SMM in most counties in NJ, and racial differences in the county of residence explained less than 15% of the racial disparity in the state overall.

Conclusions and Implications
Some of the racial disparity in SMM is driven by Blacks living disproportionately in counties with high SMM rates for both Blacks and Whites. Reducing racial disparities in these counties will not eradicate disparities at the state level. Most of the racial disparities in SMM are local, and policies and interventions directed at improving overall maternal health of Blacks, particularly in ‘hot spot’ counties, hold the greatest promise for reducing disparities at the state level.