A Randomized Trial of Healthy Families New York (HFNY): Does it Prevent Child Maltreatment?

The proposed study requests funds to extend a randomized trial of Healthy Families New York (HFNY) to test the program's effectiveness in preventing child abuse and neglect seven years following random assignment. HFNY, recently designated a 'proven program' by RAND's Promising Practice Network, uses the Healthy Families America (HFA) community-based prevention model to improve the well-being of children at risk for maltreatment through the provision of comprehensive and intensive home visitation services. Through age two, the HFNY randomized trial found reductions in both the prevalence and frequency of violence toward children (Mitchell-Herzfeld et al., 2005; DuMont et al., 2006). Since previous research suggests that the strongest benefits of home visitation programs may not become evident for several years (Olds et al., 1997; 2004), it is conceivable that these early effects may be strengthened and new effects may emerge as the target children develop. The proposed evaluation builds on baseline data already collected for 1173 women who met assessment criteria for HFNY and were randomly assigned to either an intervention group that was offered HFNY services or a control group that was given information and referrals to other appropriate services. The trial also includes follow-up data obtained from administrative maltreatment records, case record reviews, the HFNY management information system, in-depth interviews with mothers at the time of the target child's first, second, and third birthdays, and videotaped observations of parent-child interactions at Year 3. Study retention rates have been high, with 91% reinterviewed at Year 1 and 84% reinterviewed at Year 2. This study will extract information from administrative maltreatment records and to reinterview about 912 mothers and, for the first time, to interview their target children. We will use structural equation modeling to combine information about parenting from these three sources and derive two latent constructs of child maltreatment ' one dichotomous and the other continuous. The indicators will be used in logistic and Poisson regression models to examine the extent to which the intervention 1) effectively prevents and/or reduces child maltreatment at seven years; 2) has effects that are mediated by early parenting attitudes and behaviors and access to health care; and 3) impacts target children by reducing precursors of juvenile delinquency. Findings will inform national child welfare policy about the long-term outcomes and costs and benefits associated with the HFA home visitation model and how the program can be targeted and enhanced to maximize its success. CA/NCF