What is the impact of free community-based doula care on maternal and infant health and health care use: an observational study of the New York City citywide doula initiative

Doula care, in which trained non-clinical personnel provide support during pregnancy, birth, and postpartum, is a promising intervention for improving maternal health outcomes and care experiences. State Medicaid programs, which finance nearly half of births in the U.S., have rapidly expanded doula coverage. Community-based doula care models, which are rooted in disinvested communities and provide no-cost, comprehensive, and culturally concordant doula care across the perinatal period, may be particularly well-suited to address the health and social needs of pregnant Medicaid beneficiaries. While there is growing evidence of the effectiveness of community-based doula care at improving maternal and infant outcomes, evaluations have primarily focused on small, highly localized programs that do not use a consistent model. In 2022, the New York City (NYC) Health Department expanded its evidence-based community doula model by launching the Citywide Doula Initiative (CDI), scaling free, community-based doula care citywide. This study aims to build on prior evidence by evaluating whether the benefits of the community-based doula model persist at-scale and extend to a broader population and set of outcomes. Using data linked at the individual level from CDI program records and New York State Medicaid claims, we will compare outcomes for CDI clients who had a Medicaid-paid in-hospital live birth from March 2022 through December 2025 to a comparison group of similar Medicaid beneficiaries matched on demographics, geography, pregnancy/birth characteristics, Medicaid enrollment pathway, social needs, and delivery hospital. The primary outcomes will be spontaneous vaginal birth, preterm birth, and low birthweight. Secondary outcomes will include additional delivery outcomes, maternal morbidity, maternal mental health, and prenatal and postpartum health care use. We will use adjusted regression models to estimate differences between the matched groups overall and for several subgroups based on demographics and clinical characteristics. Findings will be interpreted alongside qualitative data from a related mixed-methods evaluation. With most U.S. states now implementing or expanding Medicaid doula benefits amid constrained federal funding, this evaluation will offer timely, policy-relevant evidence on how community-based doula care, implemented at-scale, affects health outcomes and health care use for Medicaid beneficiaries.