Evaluation Design for a Transformed Healthy Start Program

The Maternal and Child Health Bureau (MCHB) within the Health Resources and Services Administration provides national leadership for the Healthy Start Program, which aims to reduce racial and ethnic disparities in infant mortality. The program is undergoing a transformation as new science and opportunities have become available in the past decade. This transformation requires a strong evaluation that will contribute to the program’s evolution by helping to shape key programmatic decisions, identify successful implementation strategies, and strengthen the evidence base for the transformed program model.

MCHB selected Mathematica to develop an evaluation design and plan for the transformed Healthy Start Program. The evaluation design includes four components:

• An outcomes study that involves a quasi-experimental design with a comparison group
• A multi-level study that employs hierarchical linear modeling techniques to assess the associations between community-, grantee-, and participant-level characteristics and outcomes
• A network study to assess organizational collaboration and coordination within communities toward collective goals
• An implementation study to investigate fidelity in and facilitators and barriers to implementation, to provide information for the evaluation

Together, these complementary design components will provide information about outcomes for the programs and factors contributing to these outcomes at all levels of the socioecological model. To support empirical analysis for the design components, Mathematica developed five data collection tools, including an individual data collection form that will be administered to all Healthy Start participants and comparison women, a program survey, a community action network survey, a site visit protocol, and a focus group protocol.