The researchers wanted to study the purpose of looking at the relationship between individual-level and neighborhood-level risk variables and severe maternal morbidity. This was a cohort analysis in retrospect for all pregnancies delivered at the University of Pennsylvania Health System between 2010 and 2017. According to Centers for Disease Control and Prevention recommendations, severe maternal morbidity was characterized using International Classification of Diseases codes. Individual-level risk variables for severe maternal morbidity, such as maternal age and preeclampsia diagnosis, were examined using logistic regression modeling.
The study also utilized spatial autoregressive modeling to examine Census-tract, neighborhood-level risk variables for severe maternal morbidity, such as violent crime and poverty. There were 63,334 pregnancies in total, with a 2.73% severe maternal morbidity rate, or 272 births with severe maternal morbidity per 10,000 delivery hospitalizations. In the multivariable model that examined individual-level risk variables for severe maternal morbidity, patients having a cesarean delivery (adjusted odds ratio [aOR] 3.50, 95% CI 3.15–3.89), stillbirth (aOR 4.60, 95% CI 3.31–6.24), or preeclampsia (aOR 2.71, 95% CI 2.41–3.03) had the largest magnitude of risk.
When accounting for the number of violent crimes and the percentage of people identifying as White, the rate of severe maternal morbidity increased by 2.4% (95% CI 0.37–4.4%) for every 10% increase in the percentage of individuals in a Census tract who identified as Black or African American in our final multivariable model assessing neighborhood-level risk factors for severe maternal morbidity. Risk variables at both the individual and neighborhood levels were linked to severe maternal morbidity.
A better understanding of the risk factors for severe maternal morbidity is required for designing of clinical and public health treatments aimed at lowering the rates of severe maternal morbidity and maternal mortality.