Pathways to prevention: Early Head Start outcomes in the first three years lead to long-term reductions in child maltreatment
Introduction
There is growing evidence that early childhood prevention programs can have positive benefits for families and children. Reviews of home visiting and other early childhood programs have shown generally positive impacts on outcomes assessed in the short term, including maternal health and well-being (Avellar & Supplee, 2013; Casillas, Fauchier, Derkash & Garrido, 2016); parenting (Casillas, et al., 2016; Chen & Chan, 2016); and child health and development (Anderson et al., 2003; Peacock, Konrad, Watson, Nickel, & Muhajarine, 2013) outcomes. At the same time, outcomes of specific studies and program models have been variable and have yielded generally modest effects (Howard & Brooks-Gunn, 2009; Sama-Miller, Akers, Mraz-Esposito, Coughlin, & Zukiewicz, 2017; Shonkoff & Fisher, 2013). This has led to an increasing call for research that can better address more nuanced questions such as how such programs lead to outcomes and for whom (Avellar & Supplee, 2013; Howard & Brooks-Gunn, 2009). However, there have been few evaluations of programs for infants, toddlers and their families that have sufficient rigor, breadth of measurement, and length of follow-up period that are needed to answer these more complex questions (Supplee & Duggan, 2019). To help address this gap in research, this study used data from a large-scale national, longitudinal randomized trial the Early Head Start program, a federal program providing services to low-income pregnant women and families with children up to age three, linked to child welfare agency records from birth through age fifteen years. This study provides a unique opportunity to assess not only long-term prevention impacts, but also to identify the key pathways through which the early impacts of a widely implemented early childhood program may lead to later benefits for children.
Early Head Start (EHS) was created with bipartisan support through the reauthorization of the Head Start Act in 1994, and was designed based on growing empirical evidence of the importance of the first three years of life for children’s neurological and brain development (ACF, 1994; Shonkoff & Phillips, 2000). EHS is a primary prevention program offering services to an at-risk (selected) population of low-income families including pregnant women and families with children through age three years. It was designed to promote healthy child development and prevent negative child and family outcomes, starting prenatally. All EHS programs must follow the high standards for comprehensive services for families (including education, nutrition, health and mental health) set by the Head Start Performance Standards (U. S. Department of Health & Human Services [USDHHS], Head Start Bureau, 1992) but can be designed to fit the needs of local communities. Programs can offer child care, home visiting, or a mixture of the two services. Services provided to individual families are tailored based on each family’s unique needs and goals. In 2018–2019, the federal EHS program served well over 160,000 children per year in over 1000 programs, making it one of the largest programs serving low-income infants and toddlers in the U.S., although it still only serves approximately 11 percent of eligible children (National Head Start Association, 2018).
When Congress first authorized the EHS program they mandated a rigorous evaluation of the new program. Using an experimental design, the evaluation was conducted in 17 communities across the country. This evaluation resulted in extensive information on the positive impacts of EHS that is available in government reports and peer reviewed journals (Administration for Children and Families, 2002; Love et al., 2005). A 2013 Society for Research in Child Development Monograph provides an extensive series of articles detailing the theoretical framework of the EHS program, the methodology used in the randomized study, and documenting positive impacts at ages 2, 3, & 5 (Love, Chazan-Cohen, Raikes, & Brooks-Gunn, 2013). In brief, at the end of the program, when children were three years old, EHS was found to benefit families across a wide range of child, parent, and family outcomes, including all domains of child development assessed (health, social emotional, cognitive, language) although impacts were modest in size. There were also reductions in children’s emergency room visits due to unintentional injuries (perhaps a proxy for neglect). EHS parents were more supportive and less negative in play interactions with their children and provided home environments that were more supportive of learning and development. EHS also reduced punitive parenting (spanking) by both mothers and fathers. Furthermore, mothers who had been in EHS were more likely to be employed or in education or job training (Vogel et al., 2013). Two years after the program ended, positive impacts remained in the areas of children’s social-emotional outcomes, parenting, and parent well-being (Chazan-Cohen & Kisker, 2013; Love, Chazan-Cohen, Raikes, & Brooks-Gunn, 2013). The program thus has been shown through a rigorous trial to successfully address many documented risk and protective factors associated with maltreatment (Stith et al., 2009).
While EHS programs focus on reducing many of the risk factors associated with child maltreatment, the effectiveness of EHS in preventing child abuse and neglect had not been directly studied until a recent exploratory study by the current investigators and colleagues (Green et al., 2014). Initial findings based on a subset of 7 out of the original 17 EHS program sites suggested that by age nine, compared to children in the control group, EHS children had fewer total child welfare reports, and that these preventive impacts occurred subsequent to children’s fifth year of life. The current study expands the findings from this prior research in several ways, using data collected from a larger sample of 2794 children across 16 of the original 17 EHS program sites, and including data through children’s fifteenth birthday. While the current study did not find significant overall impacts on maltreatment outcomes (Green et al., 2017), following O’Rourke & Mackinnon (2018) we sought to assess the extent to which there are specific pathways, or mechanisms, through which early impacts of EHS reduce the likelihood of maltreatment later in life.
While scientists, policymakers, and funders have frequently called for research that not only examines questions about program impacts (e.g., “Does the program work?”) but provides information about the mechanisms underlying program effectiveness (“How does the program work?”), few studies have been conducted that have sufficient sample size and breadth to pinpoint how changes during the early years of children’s lives may lead to improved longer term outcomes (Supplee & Duggan, 2019). Such studies hold great potential for informing policy and practice, and strengthening the effectiveness of early childhood interventions, yet only a handful of such studies could be identified in the early childhood intervention literature. Heckman, Pinto, & Savelyev (2013), for example, found that the Perry Preschool program’s impact on criminal arrests in adulthood were driven largely by early effects on children’s externalizing behavior. In the only direct analysis of program-mediated outcomes of an infant-toddler prevention program that could be identified, Eckenrode et al. (2017) found that long term benefits of the Nurse-Family Partnership (NFP) home visiting program on reductions in child maltreatment were related to short-term impacts associated with reductions in the number of subsequent children born to participating women, which in turn improved economic self-sufficiency.
While prevention research in areas such as education (e.g., Loughlin-Presnal & Bierman, 2017) and substance abuse prevention (Longabaugh, 2007; Longabaugh & Magill, 2011) have a longer history of intervention studies focused on mechanisms, the notion of using mediational analyses to identify and examine a program’s “active ingredients” has more recently been introduced to the field of early childhood prevention (Supplee & Duggan, 2019). Further, researchers have typically not conducted mediational analyses in the absence of significant overall treatment effects, a practice likely based on earlier analytic approaches to testing mediation (e.g., Baron & Kenny, 1986) that have since been replaced with more sophisticated methods (Mackinnon, 2008). Recent work by O’Rourke and MacKinnon, 2015, O’Rourke and MacKinnon, 2018 points to the importance of conducting mediation analyses in all intervention studies, even those in which overall impacts do not reach statistical significance. Specifically, they argue that under some conditions, statistical power may be higher for analyses including mediational effects than for the impact model alone (O’Rourke & Mackinnon, 2015). They further state that using mediational approaches to assessing whether and how early prevention programs influence later outcomes is a “research imperative” in order for the field to begin to more accurately describe and understand the effectiveness of interventions (O’Rourke & Mackinnon, 2018).
While EHS was not designed explicitly to reduce child abuse and neglect, by addressing key child, parent, and family outcomes, the model de facto addressed a number of the key theoretical predictors of child maltreatment. For the current study we drew on this literature, as well as on the documented early program outcomes, to examine the pathways through which the EHS program may have led to reduced rates of child abuse and neglect. Specifically, early program impacts on the following domains have been theoretically linked to increased risk for child maltreatment, and thus were the central focus of the current study. First, considerable research has shown that parenting behavior, especially harsh and negative parenting and unrealistic expectations for young children’s development, is associated with increased risk for child abuse and neglect (Barth, 2009; Chaffin et al., 2004; Slack et al., 2011). Second, factors related to adult and family challenges such as mental health issues, parenting stress, and family conflict are associated with child maltreatment (Stith et al., 2009), including both neglect (Mulder, Kuiper, Van der Put, Stams, & Assink, 2018) and physical abuse (Thomas, Abell, Webb, Avdagic, & Zimmer-Gembeck, 2017). Finally, a substantial research literature suggests that children’s behavior and other characteristics also increase the risk of maltreatment. In particular, children with delayed cognitive and/or language development are at increased risk for abuse, as are children with higher levels of aggression and externalizing behavior (MacKenzie et al., 2011; Mulder et al., 2018).
The current study addresses the overarching research question: Do early positive impacts of EHS mediate subsequent impacts on child maltreatment? We hypothesized specific pathways to reduced maltreatment based on three domains in which positive early impacts of the EHS programs had been found in prior research (Love et al, 2013), namely: (1) parenting, and in particular harsh and negative parenting; (2) parental stress and family conflict; and (3) child emotional and cognitive development. More specifically, we hypothesized that the long-term impact of the EHS program on child maltreatment will be mediated by earlier positive program impacts on:
- 1.
Parenting behavior, such that children in EHS whose parents showed positive impacts on improved parenting skills early in life will experience lower rates of child maltreatment through age 15 years.
- 2.
Family conflict and parental stress, such that children in EHS whose families benefited from EHS in that parents reported less stress and lower levels of family conflict when children were young, will experience lower rates of child maltreatment through age 15 years.
- 3.
Child cognitive and social-emotional development, such that children in EHS who showed positive impacts on language, cognitive, and behavioral/social-emotional development early in life will experience lower rates of child maltreatment through age 15 years.
Section snippets
Site selection and recruitment
In 1996, the Administration on Children and Families (ACF) selected 17 of the first EHS programs to participate in the national impact study. The 17 programs were chosen to obtain diversity in geographic distribution, population served, service delivery approaches, and strength of local research partners. As part of the national study, these EHS programs recruited twice as many families as they could serve and randomly assigned 3001 primary caregivers with children under the age of one year
Child welfare involvement
A total of 504 children in the sample, or 18.0% of the 2794 children submitted for matching had at least one involvement (substantiated report and/or out-of-home placement episode) with the child welfare system. The EHS (18.7%, n = 265) and control (17.3%, n = 239) groups were similar (Table 2). The two groups also had similar percentages of children with substantiated maltreatment reports (overall 14.9%, n = 416) and out-of-home placement episodes (overall 7.6%, n = 211). Eighty-two percent
Discussion
Results of this study found that the Early Head Start program can effectively reduce child abuse and neglect after the end of program services - but primarily to the degree that services promote specific shorter-term outcomes among participating families. Children were at lower risk for later maltreatment if the programs were successful in: (1) reducing levels of family conflict and parenting-related stress; (2) supporting emotionally warm, engaged, and responsive parenting practices; and (3)
Funding
This research was funded by the Division of Violence Prevention, National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, Grant #5R01CE002317-02.
CRediT authorship contribution statement
Beth L. Green: Conceptualization, Methodology, Investigation, Writing - original draft, Supervision, Project administration, Funding acquisition. Catherine Ayoub: Conceptualization, Methodology, Investigation, Writing - original draft, Supervision, Funding acquisition. Jessica Dym Bartlett: Conceptualization, Methodology, Investigation, Funding acquisition. Carrie Furrer: Writing - original draft, Formal analysis, Data curation, Project administration, Supervision. Rachel Chazan Cohen:
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The Early Head Start Child Welfare Study (EHSCWS) projects were funded by the Division of Violence Prevention, National Center for Injury Prevention and Control Centers for Disease Control and Prevention (Contract #s 200-2010-35155 & 1RO1CE002317-01). The findings reported here are also based on research conducted as part of the national Early Head Start Research and Evaluation Project funded by the Administration for Children and Families (ACF), U.S. Department of Health and Human Services
Disclaimer
The findings reported herein were performed with the permission of the California Department of Social Services. The opinions and conclusions expressed are solely those of the authors and should not be considered as representing the policy of the collaborating agencies or of any agency of the California government.
References (58)
- et al.
Task force on community preventive services. The effectiveness of early childhood development programs: A systematic review
American Journal of Preventive Medicine
(2003) - et al.
Implementation of evidence-based home visiting programs aimed at reducing child maltreatment: A meta-analytic review
Child Abuse & Neglect
(2016) - et al.
It's not as simple as it sounds: Problems and solutions in accessing and using administrative child welfare data for evaluating the impact of early childhood interventions
Children and Youth Services Review
(2015) - et al.
The effect of Early Head Start on child welfare system involvement: A first look at longitudinal child maltreatment outcomes
Children and Youth Services Review
(2014) - et al.
Toward a cumulative ecological risk model for the etiology of child maltreatment
Children and Youth Services Review
(2011) - et al.
Risk factors for child neglect: A meta-analytic review
Child Abuse & Neglect
(2018) - et al.
Risk and protective factors for child neglect during early childhood: A cross-study comparison
Children & Youth Services Review
(2011) - et al.
Risk factors in child maltreatment: A meta-analytic review of the literature
Aggression & Violent Behavior
(2009) Parenting stress index: Professional manual
(1995)- et al.
Manual for the ASEBA preschool forms & profiles
(2000)
Statement of the advisory committee on services for families with infants and toddlers
Missing data
Effectiveness of home visiting in improving child health and reducing child maltreatment
Pediatrics
The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations
Journal of Personality and Social Psychology
Preventing child abuse and neglect with parent training: Evidence and opportunities
The Future of Children
Bayley scales of infant development
Using the HOME inventory to assess the family environment
Pediatric Nursing
Parent-child interaction therapy with physically abusive parents: Efficacy for reducing future abuse reports
Journal of Consulting and Clinical Psychology
Links between early care and education experiences birth to age 5 and prekindergarten outcomes
Monographs of the Society for Research in Child Development
Effects of parenting programs on child maltreatment prevention: A meta-analysis
Trauma, Violence, & Abuse
The multiphase optimization strategy (MOST) and the sequential multiple assignment randomized trial (SMART): New methods for more potent eHealth interventions
American Journal of Preventive Medicine
Preventing child abuse: An evaluation of services to high-risk families
The prevention of child maltreatment through the nurse family partnership program: Mediating effects in a long-term follow-up study
Child Maltreatment
II. Design and methods in the Early Head Start Study
Monographs of the Society for Research in Child Development
Introduction to mediation, moderation, and conditional process analysis: A regression-based approach
Cited by (8)
Risks of adverse childhood experiences on healthcare utilization and outcomes in early childhood
2023, Child Abuse and NeglectEarly childhood education and care policies in the U.S. And their impact on family violence
2022, Children and Youth Services ReviewCitation Excerpt :Child maltreatment was measured in three different ways in the identified articles. Most of the identified studies utilized parent-report questions to assess use of physical punishment (Ansari et al., 2016; Ansari & Gershoff, 2016; Berlin et al., 2011; Green et al., 2020; R. Lee et al., 2014; Love et al., 2005; Magnuson & Waldfogel, 2005; Pratt et al., 2015). One study examined both abuse and neglect using a validated measure of child maltreatment, the PC-CTS (Zhai et al., 2013).
Families with young children in homeless shelters: Developmental contexts of multisystem risks and resources
2023, Development and PsychopathologyParenting and family self-sufficiency services contribute to impacts of Early Head Start for children and families
2023, Frontiers in PsychologySupporting Fathers to Prevent Child Maltreatment: How Paid Family Leave and Child Care Subsidies Are Part of a Public Health Approach
2022, International Journal on Child Maltreatment: Research, Policy and Practice