Multi-level Determinants of Retention in a Home-visiting Child Abuse Prevention Program

Objective: This exploratory study investigated how attributes from multiple domains influence retention in a voluntary home-visiting child abuse prevention program.

Method: A sample of 1093 "at-risk" families participated in a home-visiting child abuse prevention program. A total of 71 Family Support Workers (FSWs) provided home visitation services in 12 different communities. Hierarchical general linear modeling (HGLM) was used to examine the community, home visitor, and maternal attributes that predicted retention in the program beyond 1 year.

Results: Multi-level analyses (HGLM) revealed significant community, home visitor, and maternal level effects. Families were less likely to remain in a home-visiting child abuse prevention program for at least 1 year if they lived in an area with high community violence. Families were more likely to remain when the home visitor received more hours of direct supervision. Older mothers were more likely to remain for at least 1 year than were younger mothers. Hispanic mothers were more likely to remain than were White non-Hispanic mothers.

Conclusions: These findings demonstrate the utility of looking across multiple levels of influence when examining retention in home-visiting child abuse prevention programs. To increase retention rates home visitors will need to be adaptable to fit the needs of families in violent communities. Supervisors can influence retention rates by providing more hours of direct supervision. Other strategies that may contribute to program retention include meeting the scheduling needs of younger mothers and involving the families of younger mothers more effectively.