MIHOPE Newsletter January 2016
Enrollment in MIHOPE is complete! We are pleased to announce that the final sample size is 4,229, with 2,111 in the home visiting group and 2,118 in the comparison group. This achievement is due to the dedication of our partners in this work, including the participating families, local sites, states, model developers, our evaluation contractors, and federal staff. With this sample size, MIHOPE becomes one of the largest studies of home visiting ever conducted, and it will be able to reliably examine impacts of home visiting on key outcomes and for key subgroups of interest. Thanks to everyone for helping us achieve this goal!
We are also pleased that, in addition to a large overall sample, the study has a large number of families representing each of the four models in the study. This is important because without enough families in each group, we may not be able to answer questions about how outcomes potentially vary across different home visiting models.
To examine how outcomes vary across models (or any other subgroup), it is preferable to have equal numbers of families in each group. Despite everyone’s best efforts, the study did not reach this goal. The final sample includes more participants from models that have the largest number of sites in the study and that enroll larger numbers of women eligible for the study (that is, women who are pregnant or have a child under six months of age). The study enrolled 1,458 participants through 26 Healthy Family America (HFA) programs, 1,234 through 22 Nurse-Family Partnership (NFP) programs, 965 through 21 Parents as Teachers (PAT) programs, and 572 through 19 Early Head Start-Home Visiting (EHS) programs.
The MIHOPE Design Report published at the beginning of the project predicted that there would be differences by model in outcome impacts since each model has slightly different goals and works with somewhat different target populations. For instance, some models may be more focused on health outcomes, others on child maltreatment, and still others on child development. This variation, along with the sample size for each model, will be taken into account in the final analysis.
The final analysis plan is due to be published in Spring 2016. The final implementation report is due to be completed in mid-2018, and the final impact report will be available in late 2018. We will keep everyone informed as the project moves towards these goals!
Measuring Family Engagement — Home visitors and the families participating in home visiting programs have to work together for home visiting services to be helpful and successful. The home visitor plays a key role in initially engaging families in the program and ensuring that they can develop good working relationships. The Mother and Infant Home Visiting Program Evaluation (MIHOPE) research team is examining the strategies that home visitors use to connect with families and the effectiveness of those strategies. To learn as much as possible, the study is using three types of information about engagement: (1) documentation (a.k.a. weekly home visit logs) kept by home visitors and supervisors, (2) interviews with staff, and (3) videos capturing what happens during home visits.
Weekly home visit logs. One way to measure family engagement is by examining how often a family has home visits, how long the visits last, and how long the family is enrolled in the program. Local programs usually capture this information in their local data systems, but not necessarily in a similar way across programs. To make sure we have the same information for all sites, MIHOPE is collecting it uniformly across all of the MIHOPE programs through home visitors’ weekly logs. In these logs, home visitors also capture more nuanced information about engagement, such as their impression of how actively the parents participate during each visit and how much parents follow through on suggested activities between visits.
Interviews with home visitors and other staff. The research team conducted 90-minute interviews with 112 home visitors, 24 supervisors, and 21 program managers in part to gain insight about family engagement. These interviews provide information about the activities conducted in the home with families and how home visitors build trusting relationships with different types of families they are working with. For example, home visitors were asked to share the strategies they used to build a relationship with two different families, discuss how each family has been responding to services, and describe how their relationship with each family has changed over time.
Video recordings of home visits. Visitor-parent communication is the “active ingredient” of home visiting. Thus, understanding how visitors and families actually interact during visits is key to understanding how much families are likely to benefit from the visits. To that end, MIHOPE has video-recorded visits for a subsample of participating families. The research team reviews and codes the videos to document the quality of visitor-family interactions. Communication between providers and recipients has many functions, such as building a helping relationship, addressing concerns and uncertainty, managing information, solving problems, and empowering recipients. Since families enroll in home visiting for a broad range of reasons, our measures are designed to capture information about how the interactions proceed in conversations about different subjects, such as parenting or prenatal health. Our measures will provide indicators of quality, such as how family-centered the conversations are, the degree of shared decision-making between home visitor and parent, and how visitors elicit and respond to family concerns.
Examining the information contained in logs, interviews, and videos will help the MIHOPE team summarize how home visitors and families engage with one another.
Using Data to Support Program Operations — The MIHOPE and MIHOPE-Strong Start teams have engaged program managers in a process that uses data to better understand the stream of referrals and program enrollment. Some program managers were already using data extensively to manage intake, but for others this was a newer endeavor. By examining the rate of referrals, the conversion of referrals to enrollments, and retention in the program once enrolled, the study team worked with 108 local program managers to estimate the rate of referrals and enrollment in home visiting compared to each program’s capacity.
Early on in the MIHOPE and MIHOPE-Strong Start studies, each program gathered key information about program capacity, referrals, eligibility, and new enrollments in a typical month. A step-by-step process and a “funnel” graphic were used to examine the data and understand how many families moved through each step in the recruitment process.
Example of a Program Intake “Funnel”
Used in MIHOPE and MIHOPE-Strong Start Planning With Local Partners
Some local programs use this type of data-driven process to plan for the flow of parents into their program, even outside of the study. This type of process can help staff systematically understand how many families need to be recruited in order to enroll a targeted number of families. It can also help to identify any specific bottlenecks, or points where families tend to drop out of the enrollment process, so that staff can develop strategies for improving the process.
MIHOPE RESEARCH ACTIVITY UPDATES
Although study enrollment is complete, local staff are still making important contributions to the study. For example:
- Until follow-up surveys with families are complete, staff will be screening referrals to ensure that families who were already assigned to the non-home visiting group are not enrolled in home visiting and are instead referred to appropriate services in the community.
- Home visitors and supervisors continue to complete weekly electronic logs at a high rate (90%) for the MIHOPE study. This is a fantastic response rate! The data collected through the logs, which will continue through June 2016, will be used in the implementation study and the cost analysis.
All in-person research visits to the local programs have been completed. The research team visited 24 of the 88 MIHOPE program sites, across seven states. During these visits, the MIHOPE team conducted interviews with home visitors and other local staff. The information gathered through these visits is a valuable aspect of the implementation study, which will shed light on how and why services varied across local sites and models.
MIHOPE-STRONG START RESEARCH ACTIVITY UPDATES
MIHOPE-Strong Start sample recruitment is complete – THANK YOU! Thank you to all the program sites, model developers, federal and state partners, and other stakeholders for supporting MIHOPE-Strong Start. It will be the largest study to examine the effectiveness of home visiting services on improving birth outcomes and infant and maternal health care use. Twenty local programs in California, Illinois, Indiana, Massachusetts, Nevada, New Jersey, New York, North Carolina, Pennsylvania, Tennessee, and Washington have enrolled families in MIHOPE-Strong Start. Many MIHOPE sites that enrolled families that met the MIHOPE-Strong Start eligibility criteria, which are narrower than the MIHOPE criteria, are also included in this study. A full description of the sites and sample will be included in the 3rd Annual Report due out this Spring (and mentioned below). We appreciate all the local staff who explained MIHOPE-Strong Start to eligible families — the study would not be possible without you!
The 3rd Annual MIHOPE-Strong Start Report: An Early Look at Families and Local Programs will be published in 2016! The report will present descriptive information on about 40 percent of the families who enrolled in the study and discuss selected characteristics of staff and local programs. This report lays the groundwork for the final report, to be published in 2018, which will examine local program implementation processes, impacts on family outcomes, and cost analyses.
CONFERENCES and EVENTS
The Secretary’s Advisory Committee (SAC) on the Maternal, Infant and Early Childhood Home Visiting Program Evaluation met for its fifth session on Monday, September 21, 2015. The purpose of the meeting was to get the Committee’s feedback on the analysis plan for the MIHOPE project, including impacts, implementation, impact variation, and cost analyses. The analysis is intended to deepen understanding of the features of local programs that are associated with greater benefits for families and thus to strengthen the future implementation of home visiting programs.
The MIHOPE research team presented at the 2015 Association for Public Policy Analysis and Management (APPAM) Fall Research Conference on November 12, 2015, in Miami, FL. The presentation, “Early Findings from the Federal Home Visiting Program Evaluation: Results from the MIHOPE Report to Congress,” included an overview of the characteristics of families enrolling in MIHOPE programs and summarized the characteristics of local programs in MIHOPE.