Maltreatment prevention through early childhood intervention: A confirmatory evaluation of the Chicago Child–Parent Center preschool program
Highlights
► In this study we examine the impacts of a preschool program on child maltreatment. ► We test mediating paths of effect on maltreatment overall and neglect specifically. ► Family and school support processes mediate program effects on all outcomes. ► Changes in child behavior mediate program impacts on maltreatment but not neglect. ► Well-designed preschool programs have the potential to prevent child maltreatment.
Introduction
Rates of child victimization in the United States, including officially reported child maltreatment, have been on the decline for many years. According to the National Child Abuse and Neglect Data System (U.S. Department of Health and Human Services, Administration on Children, Youth, and Families, 2010), for example, approximately 763,000 children were abused or neglected in 2009, down more than 20% from peak levels in the early 1990s. The newly released Fourth National Incidence Study of Child Abuse and Neglect (NIS-4) revealed a similar pattern, as the number of children reported as having experienced maltreatment decreased substantially from the levels reported by the NIS-3 in 1993 (Sedlak et al., 2010).
This downward trend is welcomed news, though it is counterbalanced by several sobering realities. First, data from reported and recorded events likely underestimate the true incidence of child maltreatment each year, and annual incidence rates undoubtedly fall short of lifetime prevalence rates. Therefore, despite the recent decrease in maltreatment, millions of American children each generation are at risk of being victimized before reaching majority age. Second, the rate reductions described above have varied by type of maltreatment; whereas verified physical and sexual abuse have declined, no significant change in child neglect has been observed (Finkelhor and Jones, 2006, Sedlak et al., 2010). Third, it is unclear why the incidence of maltreatment has decreased. The convergence of trends across data sources enhances confidence in the estimates, yet it is still uncertain whether the drop in maltreatment is authentic or artifactual. Unobserved sources of variance such as altered standards for screening and verifying abuse and neglect reports may affect official rates. Moreover, we do not know if programs and policies designed to prevent maltreatment have contributed to the apparent decrease.
While the accuracy of epidemiological data remains open to dispute, evidence compiled over the past two decades has left little doubt that exposure to abuse and neglect increases a child's risk of developmental and functional impairments. Findings from studies using various design, measurement and analytic strategies have shown that child maltreatment is associated with pervasive and persistent consequences. Maltreated children are at an elevated risk of poor physical health (e.g., Hussey et al., 2006, Lanier et al., 2010), mental health (e.g., Cohen et al., 2002, Springer et al., 2007), and behavioral health (e.g., Moran et al., 2004, Topitzes et al., 2010) outcomes relative to their non-maltreated peers. Child maltreatment is also associated with decreased educational and economic attainments (McGloin and Widom, 2001, Mersky and Topitzes, 2010, Stone, 2007, Zielinski, 2009) as well as delinquency and crime (English et al., 2002, Ireland et al., 2002, Mersky and Reynolds, 2007). Compounding its impacts on victims, child maltreatment results in substantial costs to society due to lost productivity and expenditures for public services (Wang & Holton, 2007).
Increased awareness of the consequences outlined above has led to greater interest in programs and policies that have the potential to reduce maltreatment or associated risks. As summarized in several recent reviews (e.g., Geeraert et al., 2004, Klevens and Whitaker, 2007, MacLeod and Nelson, 2000, MacMillan et al., 2009, Mikton and Butchart, 2009, Reynolds et al., 2009, Sweet and Appelbaum, 2004), maltreatment prevention initiatives have often targeted expectant primiparous mothers and families with young children. Collectively, early childhood programs represent a preferred maltreatment prevention strategy because, among all victims, roughly one-third are younger than 3 and nearly one-half are below age 6 (U.S. DHHS, 2010).
Of the early childhood interventions that have been cited for their potential to prevent maltreatment, home visitation models have received the most attention. This is due in large part to the renown of Olds' Nurse–Family Partnership (NFP) program. An oft-cited randomized trial of the NFP discovered that children of young mothers who participated in the program were significantly less likely to have been victims of substantiated maltreatment than control children (Olds et al., 1997). Subsequent investigations of the NFP and other home visitation interventions have yielded mixed results with respect to maltreatment and related outcomes, however (Cicchetti & Toth, 2004). Divergence from the original findings of Olds, Sadler, and Kitzman, (2007) may be attributable to differences in salient features (e.g., intensity; fidelity; and staff qualifications) of the programs that have been tested.
Despite receiving less public attention and scientific scrutiny, several other intervention models have been linked to reduced maltreatment or its associated risks, including parent education programs (Barth, 2009, Britner and Reppucci, 1997), health services programs (Brayden et al., 1993), and multi-component programs (DePanfilis and Dubowitz, 2005, Prinz et al., 2009, Stevens-Simon et al., 2001). Evaluations of center-based early childhood interventions have also produced encouraging results. For example, one randomized investigation found that parents whose children attended Head Start, the well-known school readiness program that serves low-income families, were less likely than control parents to report spanking their children (Puma et al., 2005). Other studies have produced similar findings, suggesting that participation in programs like Head Start and Early Head Start may reduce the incidence and frequency of parents' self-reported spanking (Love et al., 2005, Magnuson and Waldfogel, 2005). These results reinforce prior findings from Reynolds and Robertson (2003) who evaluated the impacts of participation in the Chicago Child–Parent Center preschool program, an early childhood intervention that shares many features with Head Start. The authors discovered that children who attended the program were less likely to have an indicated1 maltreatment report by age 18 than a comparable group of children who did not attend preschool in a Child–Parent Center (Reynolds & Robertson, 2003).
Although promising approaches to preventing the maltreatment of young children have begun to emerge, lingering gaps in the literature remain. First, as aforementioned, researchers have principally focused on home visitation programs and, to a lesser degree, parent education programs and school-based sexual abuse prevention programs (Mikton & Butchart, 2009). Without minimizing the potential benefits of these prevention models, alternative approaches also warrant attention. Second, most evaluations of maltreatment prevention initiatives have suffered from significant methodological limitations. Few studies, for instance, have examined programmatic effects on actual abuse and neglect outcomes (Reynolds et al., 2009). Caution should be exercised when extrapolating from a program's effects on risks associated with maltreatment to actual impacts on maltreatment. Third, there is limited knowledge of how prevention programs impact different forms of maltreatment, and there is a dearth of interventions with demonstrated impacts on child neglect (DePanfilis and Dubowitz, 2005, Mersky et al., 2009).
Last, even among well-designed studies of high-quality interventions, little is known about the mechanisms that link programs to their prevention effects. To our knowledge, only one exploratory study has examined mediating paths leading from participation in an early childhood intervention to reduced maltreatment (Reynolds & Robertson, 2003). Fortunately, future studies in this area can draw from a more developed body of etiological research that has cataloged risks associated with child maltreatment. Numerous family-level indicators are known correlates of maltreatment, including poverty, early childbearing, low parent educational attainment, family conflict, poor parent–child relations, parent mental health and substance use problems, as well as low levels of parent involvement in school (e.g., Berger and Brooks-Gunn, 2005, Mersky et al., 2009, Sidebotham and Heron, 2006, Stith et al., 2009). Broader structural and ecological factors, including neighborhood poverty and unemployment, have also been shown to contribute to maltreatment prediction models net of family-level sources of variance (See Coulton, Crampton, Irwin, Spilsbury, & Korbin, 2007 for review). Less is known about individual markers that elevate a child's risk of maltreatment, but a few studies have documented associations between child temperament characteristics or externalizing behaviors and maltreatment victimization (Harrington et al., 1998, McBride et al., 2002, Stith et al., 2009). Further research is needed to determine whether changes in the individual, family and extrafamilial processes listed above, along with other mechanisms of effect, explain why certain programs impact rates of maltreatment.
Using data from the Chicago Longitudinal Study (CLS), our investigation makes two significant contributions to the maltreatment prevention literature. First, we extend the work of Reynolds and Robertson (2003) who conducted hierarchical regression analyses to explore select mediators that link the Chicago Child–Parent Center (CPC) preschool program to a reduction in child maltreatment. We use structural equation modeling (SEM) to test a more comprehensive set of child, family, and school measures hypothesized to mediate the CPC-maltreatment connection. Second, despite being the most prevalent form of maltreatment and a serious threat to child well-being (Chaffin and Bard, 2006, Tyler et al., 2006), there has been remarkably limited scholarly activity devoted to neglect and its prevention. Therefore, we build on evidence suggesting the CPCs generated a significant reduction in child neglect (Mersky et al., 2009) by examining mediators of the CPC-neglect connection.
Section snippets
Sample and design
The CLS is a quasi-experimental panel study of 1539 underprivileged minority individuals born in 1979 or 1980. The original sample included a cohort of 989 children who participated in the CPC preschool program (described below) in 1983 or 1984 and completed CPC kindergarten in 1986. Eligible families that enrolled their children in the program resided in high-poverty neighborhoods served by one of 20 different CPC preschool sites. A comparison group of 550 children from comparable family and
Results
Table 1 displays descriptive statistics for study measures and results from bivariate correlation analyses. Each mediator displayed was significantly associated (p ≤ .05) with the preschool program measure and with both indicators of maltreatment. We tested other measures that fulfilled the double correlation criterion but failed to meet mediator criteria in the multivariate context. For instance, we found a significant bivariate correlation between troublemaking behavior and neglect, but the
Discussion
This investigation reinforces previous CLS findings indicating that sample children who attended CPC preschools were significantly less likely to be maltreated than a comparable group of children who attended other public school programs. Consistent with a confirmatory program evaluation approach (Reynolds, 1998), we systematically investigated hypothesized mediators of the program's main effects. We discovered that family and school indicators contributed significantly to explaining the CPC
References (93)
- et al.
Childhood predictors of adult substance abuse
Children and Youth Services Review
(2010) Parent involvement in elementary school and educational attainment
Children and Youth Services Review
(2004)- et al.
A prospective study of secondary prevention of child maltreatment
The Journal of Pediatrics
(1993) Is it time to rethink healthy start/healthy families?
Child Abuse & Neglect
(2004)- et al.
How neighborhoods influence child maltreatment: A review of the literature and alternative pathways
Child Abuse & Neglect
(2007) Enrollment strategies in early home visitation to prevent physical child abuse and neglect and the ‘‘universal versus targeted’’ debate: A meta-analysis of population-based and screening-based programs
Child Abuse & Neglect
(1999)- et al.
Programs for the promotion of family wellness and the prevention of child maltreatment: A meta-analytic review
Child Abuse & Neglect
(2000) - et al.
Interventions to prevent child maltreatment and associated impairment
Lancet
(2009) - et al.
School mobility and achievement: A meta-analysis
Children and Youth Services Review
(2004) - et al.
Comparing early adult outcomes of maltreated and non-maltreated children: A prospective longitudinal investigation
Children and Youth Services Review
(2010)