Economic evaluation research in the context of Child Welfare policy: a structured literature review and recommendations

Objectives

With over 1 million children served by the US Child Welfare system at a cost of $20 billion annually, this study examines the economic evaluation literature on interventions to improve outcomes for children at risk for and currently involved with the system, identifies areas where additional research is needed, and discusses the use of decision-analytic modeling to advance Child Welfare policy and practice.

Methods

The review included 19 repositories of peer-reviewed and non-peer-reviewed “gray” literatures, including items in English published before November, 2009. Original research articles were included if they evaluated interventions based on costs and outcomes. Review articles were included to assess the relevance of these techniques over time and to highlight the increasing discussion of methods needed to undertake such research. Items were categorized by their focus on: interventions for the US Child Welfare system; primary prevention of entry into the system; and use of models to make long-term projections of costs and outcomes.

Results

Searches identified 2,640 articles, with 49 ultimately included (19 reviews and 30 original research articles). Between 1988 and 2009, reviews consistently advocated economic evaluation and increasingly provided methodological guidance. 21 of the original research articles focused on Child Welfare, while 9 focused on child mental health. Of the 21 Child Welfare articles, 81% (17) focused on the US system. 47% (8/17) focused exclusively on primary prevention, though 83% of the US system, peer-reviewed articles focused exclusively on prevention (5/6). 9 of the 17 articles included empirical follow-up (mean sample size: 264 individuals; mean follow-up: 3.8 years). 10 of the 17 articles used modeling to project longer-term outcomes, but 80% of the articles using modeling were not peer-reviewed. Although 60% of modeling studies included interventions for children in the system, all peer-reviewed modeling articles focused on prevention.

Conclusions

Methodological guidance for economic evaluations in Child Welfare is increasingly available. Such analyses are feasible given the availability of nationally representative data on children involved with Child Welfare and evidence-based interventions.

Practice implications

Policy analyses considering the long-term costs and effects of interventions to improve Child Welfare outcomes are scarce, feasible, and urgently needed.     

Parent-child participation in planning children's activities outside of school in European American and Latino families

This longitudinal research used a sociocultural perspective to examine planning competence in the everyday experiences of European American and Latino children from 7 to 9 years of age. Data on children's participation in planning their activities outside of school, parental expectations about children's planning competence, and children's planning in the classroom were collected yearly from Grades 2 to 4 from 140 children and their mothers, and the children's teachers. Results indicate that decision-making practices and parental expectations change with development and vary by ethnicity. Decision making at home was related to children's classroom planning; however, the nature of these relations changed over middle childhood. Results are discussed in terms of cultural and parental contributions to the development of planning skills.     

Disclosing unwnated sexual experiences: results from a national sample of adolescent women

OBJECTIVE: The aims of this study are to identify factors that influence the disclosures made by female survivors of unwanted sexual experiences (USE) in childhood and adolescence. The predictors of both the timing of disclosure (short delay, long delay, non-disclosure) and the recipient of the disclosure (disclosing ever to an adult, disclosing to peers only, non-disclosure) were investigated. Participant characteristics, USE characteristics, and family contextual variables were explored. METHOD: A sub-sample of 263 adolescent females who reported unwanted sexual experiences in the National Survey of Adolescents (NSA) provided data on the characteristics of their experience as well as the timing and recipients of disclosure. Two multinomial logistic regression analyses were conducted to determine significant predictors of each aspect of disclosure. RESULTS: Age of onset, a known perpetrator, a familial relationship with the perpetrator, and a history of drug abuse in the household are related to the timing of disclosure. Age of onset, penetration, fear for one's life during the USE, injury during the USE, family structure, and the age differential between the survivor and the perpetrator are linked to the recipient of the disclosure. CONCLUSION: Age is a critical variable in the disclosure process. Whereas aspects of the abuse experience were more important in predicting whom a victim would tell, the relationship to the perpetrator was more important in deciding to delay disclosure. As survivors grow into adolescence, the importance of peers provides a source of support that becomes increasingly important in decisions to disclose.     

Randomized trial of a statewide home visiting program to prevent child abuse: impact in reducing parental risk factors

OBJECTIVES: To assess the impact of a home visiting program in reducing malleable parental risk factors for child abuse in families of newborns identified, through population-based screening, as at-risk of child abuse. METHODS: This randomized trial focused on Healthy Start Program (HSP) sites operated by three community-based organizations on Oahu, HI, USA. From 11/94 to 12/95, 643 families were enrolled and randomly assigned to intervention and control groups. Mothers in both groups were interviewed annually for 3 years (88% follow-up each year) to measure mental health, substance use, and partner violence. HSP records were reviewed to measure home visiting services provided. Home visitors were surveyed annually to measure their perceived competence. RESULTS: Malleable parental risks for child abuse were common at baseline. There was no significant overall program effect on any risk or on at-risk mothers' desire for and use of community services to address risks. There was a significant reduction in one measure of poor mental health at one agency and a significant reduction in maternal problem alcohol use and repeated incidents of physical partner violence for families receiving > or =75% of visits called for in the model. Home visitors often failed to recognize parental risks and seldom linked families with community resources. HSP training programs were under-developed in preparing staff to address risks and to link families with community resources. CONCLUSIONS: Overall, the home visiting program did not reduce major risk factors for child abuse that made families eligible for service. Research is needed to develop and test strategies to improve home visiting effectiveness in reducing parental risks for child abuse.