Severe maternal morbidity and mortality in the U.S. disproportionately affect African-American (AA) women. Inequities occur at many levels, including community, provider/practice, and health system levels. This proposal will test the effectiveness and cost-effectiveness of a multilevel intervention to address AA maternal morbidity and mortality in two Michigan counties: Genesee County (which includes Flint) and Kent County (which includes Grand Rapids). Interventions were developed or co-developed by our partners in these counties, who include AA women residents, enhanced prenatal and postnatal care (EPC) staff (including race-matched community health workers), and physician/health system staff and providers. Community level intervention. We will expand access to EPC services (i.e., home visiting programs, Healthy Start programs) using telehealth and flexible scheduling. Despite being designed for minority women, about 60% of eligible AA women in Michigan do not enroll in EPC services. Pilot work indicates that 50% of minority women who declined EPC services would participate if a telehealth option was available. We will provide this option. Provider/practice level intervention. We will address provider and health system implicit and explicit bias and corresponding structures and practices and make this learning actionable using daylong experiential trainings. Training will include didactics, reflection, discussion, windshield tours, and brainstorming ways to tailor participants’ practices and settings to better meet the needs of perinatal AA women. Training will include everyone from physicians to front desk staff. System level intervention. We will implement community care patient safety bundles targeting maternal health disparities throughout the intervention counties. We will test the effects of the multilevel intervention using a quasi-experimental difference-in-difference with propensity scores approach to compare pre (2016-2019) to post (2021-2024) changes in outcomes among Medicaid women in the two intervention counties with similar women in other Michigan counties. The sample will include all Medicaid insured women observed during pregnancy, at birth, and/or up to 1 year postpartum, who delivered in Michigan from 2016 – 2024 (approximately 540,000 births, including ~162,000 births to AA women). Measures will be taken from a pre-existing linked dataset that includes Medicaid claims, death records, birth records, and EPC program data. The specific aims are to: (1) Assess the effectiveness of the multilevel intervention on AA severe maternal morbidity and mortality; (2) Test improved service utilization and non-severe maternal morbidity as mechanisms of the effect of the multilevel intervention on severe maternal morbidity, and (3) Evaluate the cost-effectiveness of the multilevel intervention. This project will be among the first to evaluate a multilevel intervention to reduce AA maternal morbidity and mortality at the population level. The trial tests whether the intervention engages the mechanisms presumed to underlie intervention effects and provides cost-effectiveness data that systems need to make informed decisions about adoption, speeding implementation.

African American women are three to four times more likely to die of pregnancy related complications than non- Hispanic white women. This study will scale a community, provider, and system-level intervention to reduce African American maternal morbidity and mortality disparities and will test the intervention using data from all Medicaid insured women who deliver in Michigan from 2016-2019 and 2021–2024 (approximately 540,000 births, including 162,000 births to African American women).