In this project we will test a multilevel intervention in a large-scale pragmatic trial to reduce Non-Hispanic Black (NHB) and Hispanic pregnancy-related and -associated maternal morbidity and mortality (PRAMM) disparities in three Michigan counties with over ¼ (>2.5M) of the Michigan population. Interventions at each level were co- developed with our partners in these counties, including NHB and Hispanic pregnant persons, community lead- ers, providers, and health system representatives. Community level intervention. We will enhance a commu- nity health worker (CHW)-inclusive home visiting model by increasing pre- and postnatal focus on maternal health and by designing and implementing specific CHW interventions to directly address PRAMM equity in Black and Hispanic Medicaid birthing persons. This approach takes advantage of existing statewide infrastruc- tures in MI and many other states to expand the reach of maternal health disparities interventions. The pro- vider/practice level intervention will address patient-provider interactions through actionable, experiential pro- vider trainings focused on communication (e.g. listening to birthing persons), provider and practice racism, bias, cultural relevance, and awareness of and referrals to community care services including CHWs and home visit- ing, for both clinical and community settings. The study is innovative in that (i) it is the first large-scale test of a scalable and integrated CHW – home visiting program specifically designed for reducing PRAMM and disparities among NHB and Hispanic birthing persons; (ii) will include interventions for specific comorbid conditions relevant to PRAMM and disparities, including multimorbidity; (iii) will be one of the first to address patient-provider inter- actions both from the provider and patient perspective and will incorporate innovative upstream provider training methods such as community health detailing; (iv) and will enhance the coordination of clinical and community care at multiple levels. The study will use a quasi-experimental, stepped wedge design. Participants will be Medicaid insured persons in Wayne, Kent, and Genesee counties observed during pregnancy, at birth, and up to 1 year postpartum, who deliver a birth between 2021-2028 (~101,000 births, including ~39,000 NHB and Hispanic). Analyses will use a statewide linked data system including all Medicaid birth and death records, Med- icaid claims, and other program data. In the context of Michigan extending pregnancy-related Medicaid coverage to 12 months postpartum starting in 2022, the study will be able to evaluate the impact of the proposed multilevel intervention vs usual care from early pregnancy through 12 months postpartum. Specific aims are to: (1) Assess the effectiveness of the intervention (vs usual care) in reducing NHB and Hispanic PRAMM (up to 1-year post- partum; overall & relative to NHW persons) and NHB and Hispanic severe maternal morbidity and pregnancy- associated (which includes pregnancy-related) mortality (overall & relative to NHW persons); (2) Test access to care, quality, and social conditions as mechanisms of the effect of the multilevel intervention on NHB and His- panic PRAMM disparities; (3) Evaluate the cost-effectiveness of the multilevel intervention.