Risk of Severe Maternal Morbidity Associated with Maternal Comorbidity Burden and Social Vulnerability

Abstract

Objective We evaluated the associations of the obstetric comorbidity index (OB-CMI) and social vulnerability index (SVI) with severe maternal morbidity (SMM). Study Design Multicenter retrospective cohort study of all patients who delivered (gestational age > 20 weeks) within a university health system from January 1, 2019, to December 31, 2021. OB-CMI scores were assigned to patients using clinical documentation and diagnosis codes. SVI scores, released by the Centers for Disease Control and Prevention (CDC), were assigned to patients based on census tracts. The primary outcome was SMM, based on the 21 CDC indicators. Mixed-effects logistic regression was used to model the odds of SMM as a function of OB-CMI and SVI while adjusting for maternal race and ethnicity, insurance type, preferred language, and parity. Results In total, 73,518 deliveries were analyzed. The prevalence of SMM was 4% (n = 2,923). An association between OB-CMI and SMM was observed (p < 0.001), where OB-CMI score categories of 1, 2, 3, and ≥4 were associated with higher odds of SMM compared with an OB-CMI score category of 0. In the adjusted model, there was evidence of an interaction between OB-CMI and maternal race and ethnicity (p = 0.01). After adjusting for potential confounders, including SVI, non-Hispanic Black patients had the highest odds of SMM among patients with an OB-CMI score category of 1 and ≥4 compared with non-Hispanic White patients with an OB-CMI score of 0 (adjusted odds ratio [aOR] 2.76, 95% confidence interval [CI] 2.08–3.66 and aOR 10.07, 95% CI 8.42–12.03, respectively). The association between SVI and SMM was not significant on adjusted analysis. Conclusion OB-CMI was significantly associated with SMM, with higher score categories associated with higher odds of SMM. A significant interaction between OB-CMI and maternal race and ethnicity was identified, revealing racial disparities in the odds of SMM within each higher OB-CMI score category. SVI was not associated with SMM after adjusting for confounders. Key Points