Association for Public Policy Analysis & Management

In the past several years, evidence-based home visiting programs have received unprecedented support from the federal government in an effort to improve maternal and child outcomes for our highest risk families. Through funding provided by the Affordable Care Act, nearly every state has implemented a Maternal, Infant, and Early Childhood Home Visiting (MIECHV) program, which requires states to implement home visiting programs that have demonstrated impacts on maternal and child outcomes through rigorous evaluation methods. The early intervention programs provide expectant and newer parents with resources, education, and support through regular visits in the family’s home by a trained professional. 

The home visiting evidence base suggests that the impacts of several programs are larger or only observed for families considered high-risk, whereas the impacts for less disadvantaged families are modest or negligible on most outcomes. However, findings from implementation studies of home visiting programs suggest that attrition is a serious concern, and that it is greatest among the highest risk families. This set of findings poses a challenge for states as they aim to recruit and serve high-risk families to make the greatest impact on family outcomes, while simultaneously struggling to retain high-risk families in their programs. If the high-risk families drop out, then will the states see only minimal impacts on maternal and child outcomes?

This paper examines the extent to which families with the highest level of risks are more likely to leave the home visiting programs prior to completion, and whether their higher rates of attrition help explain differences in maternal and child outcomes. We focus on the home visiting program in Texas, which is one of the largest and most diverse in the country. The MIECHV program in Texas incorporates four home visiting program models across fifteen counties, and serves over 3,000 mothers per year. The communities that implement the programs were selected because their residents have high levels of risk; however, there is considerable variation in risk level across families, within communities. The data for this study come from a multi-year, ongoing evaluation of the statewide implementation of the Texas MIECHV program, and include program administrative data and longitudinal survey data from home visitors and mothers participating in home visiting.

Analyses will examine the effect of several risk factors on the program dosage families receive, operationalized as length of time families remain in a home visiting program and the number of home visits the family completes. Risk factors examined will include poverty and economic hardship, teen parent status, substance abuse, family violence, family instability, household chaos, poor maternal health, father incarceration, child developmental disability status, and parental immigration status. Outcomes examined will include parental support for child’s learning and length of breastfeeding. The model will control for several selection variables including parent and child race/ethnicity, primary language of parent, and parental employment and education status. We anticipate the results of this study will uncover the potential complications of recruiting and retaining high-risk families in home visiting programs.