Explaining variation in birth outcomes of Medicaid-eligible women with variation in the adequacy of prenatal support services.

OBJECTIVES: This study examines the contribution of the adequacy of nutrition, psychosocial, and health education support service delivery in explaining variation in birth outcomes among Medicaid-eligible women, their provider sites, and practice settings. METHODS: Logistic regression models for low birthweight and preterm birth outcomes are first fitted with medical record data on maternal risks and use of prenatal visits for more than 3,485 women receiving care at 27 ambulatory sites, correcting for clustering of women within sites. RESULTS: The change in variation explained by these models with the addition of the adequacy of support services indicates that providing at least one nutrition, psychosocial, and health education service session each trimester of care contributes significantly to explaining better birth outcomes when compared with providing fewer sessions. When the expected outcome rates calculated with the estimated effects in the models are compared with their observed rates across provider sites and setting types, however, adequacy of service delivery does not help to explain differences in outcomes at individual sites or types of settings. CONCLUSIONS: Although repeated support service sessions during prenatal care improve the chances of avoiding poor birth outcomes in low income women, further adjustments for other differences between women or service delivery are needed to explain variation in outcomes at different sites and practice settings.