Panel Paper: The Impact of Strong Start on Birth Outcomes and Expenditures Among Women Covered By Medicaid.

Saturday, November 9, 2019
Plaza Building: Concourse Level, Plaza Court 8 (Sheraton Denver Downtown)

*Names in bold indicate Presenter

Lisa Dubay1, Bowen Garrett1, Fredric Blavin1, Emily M. Johnston1, Embry Howell1, Justin Morgan1, Brigette Courtot1, Sarah Benatar1, Ian Hill1 and Caitlin Cross-Barnet2, (1)Urban Institute, (2)Centers for Medicare & Medicaid Services


The Strong Start for Mothers and Newborns Initiative aimed to reduce preterm births and improve birth outcomes among Medicaid beneficiaries through three enhanced prenatal care approaches: birth centers (BC), group prenatal care (GPC), and maternity care homes (MCH). This study estimates the effects of each Strong Start model on birth outcomes and expenditures among Strong Start participants relative to Medicaid beneficiaries obtaining non-Strong Start prenatal care in “typical” practices.

In the largest study of its kind, birth certificate and Medicaid eligibility/claims data were obtained from state agencies and linked in 13 states. Propensity score re-weighting was used to compare women participating in Strong Start to Medicaid beneficiaries in the same counties who had similar social and medical risk profiles and received typical prenatal care. Data from case-studies were used to interpret results. Analyses included Strong Start participants (N=15,120) and other Medicaid-covered pregnant women (N=875,374) who delivered in 2014-2016 in counties with Strong Start.

We found that infants born to BC participants were 2.2 percentage points (26 percent) less likely to be pre-term and l.5 percentage points (20 percent) less likely to be low-birthweight than infants born to women receiving typical care through Medicaid. Rates of C-sections were 11.5 percentage points (40 percent) lower, rates of vaginal birth after cesarean 11.6 percentage points (93 percent) higher, rates of weekend deliveries (a proxy for unscheduled deliveries) 4.0 percentage points (20 percent) higher, and expenditures for mother and infant during the prenatal care and delivery period $1759 (21 percent) lower for women who received care in a BC compared to those receiving typical care. There were few positive differences in birth outcomes for participants in GPC practices or MCHs compared to women receiving typical care, but women participating in these two models had slightly higher rates of weekend deliveries. Results were robust to specifications that used diagnoses from claims to control for underlying health status. Results mentioned above are significant at p<=0.05 level.

Receiving prenatal care in BCs, which employ the holistic, time-intensive midwifery model of care and serve women at lower medical risk, was associated with large improvements in outcomes for women covered by Medicaid. Moreover, better outcomes were achieved at lower costs. Improvements were not observed in either GPC or MCHs.

BCs are an effective and efficient option for serving Medicaid-covered pregnant women with lower levels of medical risk. Increased use of the midwifery model of care in birth centers under Medicaid could lead to significantly improved birth outcomes for these women while reducing costs to Medicaid. GPC may not be well suited to Medicaid beneficiaries as many women faced challenges attending the set appointment times due to fluctuating work schedules and transportation barriers. Care-coordination layered on typical care, the general approach of MCHs, does not appear sufficient to improve birth outcomes, perhaps due to challenges care coordinators faced addressing the full scope of client needs. Moving forward, comprehensively addressing the broader needs faced by low-income women, including many social determinants of health, may be necessary to achieve improved outcomes.