Child Maltreatment Research, Policy, and Practice for the Next Decade: Workshop Summary (2012)

Appendix D


Background Paper: Major Research
Advances Since the Publication of the 1993
NRC Report Understanding Child Abuse and
Neglect: Highlights from the Literature ¹

By Rosemary Chalk²

Introduction

Definitions and Frameworks

Identification, Reporting, and Data Sources

Causes

Consequences

Prevention Efforts

Treatment Interventions

Services and Systems-Level Issues

Social Policy

Conclusion

References

______________________

¹This paper was commissioned by the Institute of Medicine (IOM) and National Research Council (NRC) to provide background for the January 30-31 Workshop on Child Maltreatment Research, Policy, and Practice for the Next Generation, hosted by the IOMNRC Board on Children, Youth, and Families. The responsibility for the content of this article rests with the author and does not necessarily represent the views of the IOM, the NRC, or their committees and convening bodies.

²The author wishes to acknowledge the contributions of several individuals who reviewed early drafts of this paper, including Lucy Berliner, Mark Chaffin, Lisa Jones, Melissa Jonson-Reid, John Leventhal, Joy Osofsky, Anne Petersen, Andrea Sedlak, and Melissa Welch-Ross. Yeonwoo Lebovitz, IOM research associate, provided extensive bibliographic and research assistance in the preparation of the paper.

INTRODUCTION

The problem of child maltreatment has persisted as one of the most serious threats to child health and safety in the United States. The most recent National Incidence Study (NIS-4) reported that more than 1.25 million children, involving 1 in 58 children from the general population, were abused or neglected by a parent during the 2005-2006 survey period according to the evidence of harm standard (Sedlak et al., 2010b). When the broader standard of endangerment is applied (which includes maltreatment by adult caretakers other than parents, or by teenaged caretakers in the case of sexual abuse), the number of children in substantiated cases increases to nearly 3 million children, involving 1 in 25 children, according to NIS-4 data.

More recent data, provided in the FY 2010 report Child Maltreatment (HHS, 2011, p. ix), indicates that “the unique victim rate was 9.2 victims per 1,000 children in the population” when considering substantiated reports of child abuse and neglect. The overall rate of child maltreatment deaths, the most tragic consequences of abuse and neglect, was 2.07 deaths per 100,000 children, based on estimates provided by state child welfare agencies (HHS, 2011).

The statistical figures mask a complex picture of child maltreatment, one that frequently challenges the general public’s perception of the nature of the problem of child abuse and neglect. For example:

•  Maltreatment is frequently viewed as physical or sexual abuse,yet child neglect reports consistently account for the large majority of the reported cases in national surveys and official records.

•  The NISs report that the general incidence of child maltreatmentdeclined by 19 percent (harm standard) in the 12 years between the data reported in NIS-3 (which collected data in 1993) and NIS-4 (Sedlak et al., 2010b). This decline occurred during a period of growth in the child population in the United States. When adjusted to account for such growth, the rate of decline per 1,000 children equals 26 percent, approaching the 1986 incidence level reported in the NIS-2 estimate.

•  Most of the rates of decline can be explained by significant decreases in reports of physical or sexual abuse of children; the level of child neglect reported in NIS-4 has remained about the same as that reported in NIS-3. Finkelhor and Jones (2006) offer

additional explanations for the causes of the reported decline in child maltreatment rates.

•  One recent study suggests that the perceived decline in NIS reports of abuse and neglect is not consistent with other data regarding trends in hospital admissions for child maltreatment injuries, which have remained stable from 1997 (Gilbert et al., 2012). However, their analysis relied on data that are not nationally representative, excluded indicators of sexual abuse, and also classified as maltreatment cases where injuries in question were classified as of “undetermined cause” (Personal communication, L. Jones, University of New Hampshire, February 10, 2012).

•  Although a relatively small proportion of reported cases of child abuse and neglect meet legal criteria for substantiation, several studies have suggested that unsubstantiated cases face equal risks (Hussey et al., 2005; Kohl et al., 2009).

These changes, as well as persistent trends, need to be considered in light of the changing demographics of the U.S. child population. Statistics in the report America’s Children (FIFCFS, 2011), based on data compiled by the Census Bureau through the Current Population Survey, notes that more than one in five children now live in poverty (see www.childstats.gov/americaschildren/eco1.asp). The report states:

•  The percentage of children in families with incomes below the poverty threshold (defined as $21,756), which is a significant risk factor for abuse and neglect, rose from a low of 16 percent in 2000 and 2001 to 21 percent in 2009.

•  The wealth disparities that now characterize American society also affect children: The percentage of children living in families in extreme poverty (defined as 50 percent of $21,756) rose from 6 percent in 2000 to 9 percent in 2009, which is the highest estimate for related children since 1997.

•  The percentage of children who lived in families with very high incomes (600 percent or more of the poverty threshold) remained unchanged between 2000 and 2009 (13 percent). The rising number of children living in poverty is particularly noteworthy among younger children. In 2009, 24 percent of related children ages 0 to 5 lived in poverty, compared with 18 percent of older related children.

The reported decline in child maltreatment rates is part of a larger decline in other forms of violent crime, which began in the early 1990s and continues today (Blumstein and Wallman, 2006). In recent decades, however, dramatic shifts have occurred in the American economic arena as well as the racial and ethnic composition of America’s children. According to data from the 2010 Census that is included in the America’s Children report, 54 percent of U.S. children were white, non-Hispanic; 23 percent were Hispanic; 14 percent were Black; 4 percent were Asian; and 5 percent were in the category “all other races” http://www.childstats.gov/americaschildren/demo.asp). Rapid increases have been recorded particularly among the percentage of children who are Hispanic, who made up only 9 percent of the child population in 1980 (http://www.childstats.gov/americaschildren/demo.asp).

In light of these economic trends and shifting demographics, it is important to consider at this time how the health and safety of children have changed over the past decade. A particular area of interest is research on child maltreatment, which focuses on the characteristics and needs of children and families who experience physical abuse, sexual abuse, emotional maltreatment, and neglect.

1993 Academy Study on Child Abuse and Neglect

Nearly two decades ago, the National Research Council (NRC) published the report Understanding Child Abuse and Neglect (NRC, 1993). The report was prepared by a panel of national experts, following a comprehensive study and critique of the existing research literature as well as discussions with hundreds of practitioners from the social services, healthcare, and legal systems that serve vulnerable children and their families. The NRC report embraced a developmental and ecological perspective in examining the various dimensions of the problem of child maltreatment, and the study panel offered a general conceptual childoriented framework to guide new approaches to child and family services as well as to set priorities that could integrate a diverse, fragmented, and interdisciplinary research literature.

The 1993 report included 10 chapters that offered a synthesis of the key research studies under the designated topics:

1.  Introduction

2.  Identification and Definitions

3.  Scope of the Problem

4.   Etiology of Child Maltreatment

5.  Prevention

6.  Consequences of Child Abuse and Neglect

7.  Interventions and Treatment

8.  Human Resources, Instrumentation, and Research Infrastructure

9.  Ethical and Legal Issues in Child Maltreatment Research

10.  Priorities for Child Maltreatment Research

Based on this review, the study panel highlighted 17 research priorities, organized within a research agenda that addressed four objectives:

1.  Clarify the nature and scope of child maltreatment;

2.  Provide an understanding of the origins and consequences of child maltreatment in order to improve the quality of future policy and program efforts;

3.  Provide empirical information about the strengths and limitations of existing interventions as well as guiding the development of more effective ones; and

4.  Develop a science policy for child maltreatment research that recognizes the importance of national leadership, human and financial resources, instrumentation, and appropriate institutional arrangements.

Since the publication of the 1993 report, the field of child maltreatment studies has continued to expand. While the Office on Child Abuse and Neglect (formerly the National Center for Child Abuse and Neglect) within the U.S. Department of Health and Human Services (HHS) continued to support a modest research portfolio, other federal sponsors invested in child maltreatment studies, including a national consortium on child neglect research organized by several institutes within the National Institutes of Health. In the intervening years, a national child abuse prevention initiative within the Doris Duke Charitable Foundation has also emerged. In addition, a wave of animal and human research studies focused on stress, trauma, and the regulation of adverse environmental influences (including threats and violence) has embraced the significance of child maltreatment as a major influence on health and well-being (see, e.g., papers produced by the Center on the Developing Child at Harvard University, http://developingchild.harvard.edu).

The expansion of research in the neurosciences, including the development of new tools that are capable of imaging brain structures and functions, has advanced our understanding of the intricate and complex

processes associated with the regulation of adverse stimuli. Additional biological studies, focused on other systems such as the immune function and interactions between genetic structures and the social environment, are shaping the ways in which researchers view complex interactions among threats and protective factors in forming the pathways to and consequences of child maltreatment. Research that is focused on selected childhood injuries, such as head trauma, has also converged with studies of child maltreatment, especially in highlighting selected stages of development (e.g., infancy) or child behaviors (e.g., prolonged crying) that may be especially vulnerable to particular forms of abuse or trauma among young children.

Purpose and Scope of This Paper

This paper highlights some of the major research advances since the publication of the 1993 NRC report, with a particular emphasis on studies published in the past decade. The objective is to provide an initial guide to recent research that offers a significant guide for our understanding of a multifaceted and disturbing subject, in preparation for a January 2012 workshop on child abuse and neglect research convened by the Institute of Medicine (IOM) and the NRC. The paper is designed for a general audience that may not be acquainted with the full range of relevant studies in the social, behavioral, health, and biological sciences.

This paper cannot offer a comprehensive review of the literature concerned with child maltreatment or cover all the topics addressed in the initial 1993 study. Rather, it provides a brief overview of selected research within most of the nine categories that mirror the chapters of the original report (the topic of ethical and legal issues is not addressed in this paper, although there is a brief section on social policy that incorporates some of this discussion).

In keeping with the original NRC report, the paper has a child orientation rather than a broader review of perpetrator, family, neighborhood, or cultural characteristics associated with abuse and neglect, which deserve further attention in a more comprehensive analysis. Research studies focused on specific aspects of child welfare, such as the experience with alternative forms of foster care placements or disproportionality in the foster care population, are not addressed. Nor is attention directed toward topics such as the reliability of child testimony, or interventions in judicial settings for victims of child abuse and neglect. While these other areas are certainly suitable for a more comprehensive research re-

view, they fell beyond the scope of what was feasible to address in the 3-month time period for this effort.

Research Methods

The research studies selected for inclusion in this paper were identified through a search of the bibliographic databases operated by the Web of Science and a comprehensive library search function of the National Academy of Sciences, which includes 13 separate databases (e.g., Academic Search Premier, SCOPUS, and Science Direct). The initial search focused on the identification of research review papers that received a significant number of citations in other articles and narrowed the list to 30 from the top-ranking 50 articles. The initial database review was then supplemented by searches of additional research sources, such as the National Criminal Justice Research Service, the Child Abuse and Neglect Digital Library maintained by Cornell University, and websites maintained by selected HHS agencies, including the Administration for Children and Families (ACF), Health Resources and Services Administration (HRSA), Centers for Disease Control and Prevention (CDC), National Institutes of Health, and Substance Abuse and Mental Health Services Administration (including the Catalog of Federal Domestic Assistance and other background materials that describe grants supported by these agencies). In some cases, reports not archived in the scientific bibliographic databases were identified through searches of websites of selected academic and professional organizations (e.g., the American Academy of Pediatrics, American Psychological Association, or Mt. Hope Family Center). Early drafts of the paper were reviewed by members and staff of the IOM-NRC planning committee for the January 2012 workshop as well as by workshop speakers and participants; their suggestions were particularly helpful in highlighting specific areas of emphasis and gaps in the literature review.

This paper strives to highlight areas of research that are characterized by multiple, theory-informed empirical studies with study populations that include children and families who have experienced abuse and neglect. Where possible, attention is directed toward those interventions that are the focus of comprehensive research reviews, striving to create a reliable evidence base to guide policy and practice. The author recognizes that multiple other studies exist that focus on common risk factors for abuse and neglect (e.g., depressed parents, domestic violence, or substance abuse). For the most part, these research areas are not included

unless the study specifically addressed child maltreatment issues as a primary outcome of interest.

DEFINITIONS AND FRAMEWORKS

Child maltreatment studies are consistently challenged by variations in definitions that characterize the events, behaviors, and experiences under review. Coulton et al. (2007) highlight the importance of distinguishing among the definition, recognition, reporting, and agency administrative classifications of child maltreatment cases that are recorded by social services agencies, and the types of child maltreatment behaviors or experiences that are self-identified by victims or offenders. These differences are not just semantic; the datasets that support official reports of child maltreatment may differ in important ways from other types of administrative records or self-report data that are obtained through household or victimization surveys.

The 1993 NRC study described an array of research studies on definitions of child maltreatment and various principles that could guide efforts to achieve greater consistency in future research studies. Since then, additional efforts have been made to improve the quality of definitions of child maltreatment used in both clinical and general population studies. Most notably, public health agencies and clinicians have made efforts to identify uniform definitions and data elements, including International Classification of Disease (ICD) codes, that can be used to classify child maltreatment injury and related health data and to incorporate these data into national health information databases, surveillance efforts, and diagnostic procedures. For example:

•  The Child Abuse Prevention and Treatment Act (CAPTA) (42 U.S.C.A. 5106g), as amended by the CAPTA Reauthorization Act of 2010, defines child abuse and neglect as, at minimum: “Any recent act or failure to act on the part of a parent or caretaker which results in death, serious physical or emotional harm, sexual abuse or exploitation” or “An act or failure to act which presents an imminent risk of serious harm.” This definition of child abuse and neglect refers specifically to parents and other caregivers. A “child” under this definition generally means a person who is younger than age 18 or who is not an emancipated minor. This legislative definition guides federal policy and programs, and sets minimum standards for states that accept

CAPTA funding. However, each state provides its own definitions of maltreatment within civil and criminal statutes, resulting in significant variation in terms of the scope of actions (or inactions) that may constitute abuse or neglect (http://www.childwelfare.gov/can/defining/federal.cfm).

•  The CDC website states that: “A consistent definition is needed to monitor the incidence of child maltreatment and examine trends over time. In addition, it helps determine the magnitude of child maltreatment and compare the problem across jurisdictions.” (http://www.cdc.gov/violenceprevention/childmaltreatment/definitions.xhtml).

•  In January 2008, the CDC published the report Child Maltreatment Surveillance: Uniform Definitions for Public Health and Recommended Data Elements, Version 1.0 (Leeb et al., 2008). The report includes recommendations “designed to promote consistent terminology and data collection related to child maltreatment” that were developed through an extensive expert consultation process. The 2008 CDC report notes that prior efforts by the research and legal communities to develop consistent and uniform definitions of child maltreatment were not adequate for use in public health surveillance because many of the data sources used by the research and legal communities are not available to state and local public health officials:

Because no public health-based definitions for child maltreatment exist, public health officials continue to use terms related to child maltreatment in different ways and use different terms to describe the same acts. Not surprising, these inconsistencies have contributed to varied conclusions about the incidence and prevalence of child abuse and neglect. (Leeb et al., 2008, p. 12)

The CDC therefore developed uniform definitions and a set of recommended data elements to guide surveillance efforts by public health agencies. However, the report does not provide specific instruments for surveillance nor does it offer clinical information for identifying child maltreatment.

The uniform definition included in the CDC report is: “Child Maltreatment is any act or series of acts of commission or omission by a parent or other caregiver that results in harm, potential

for harm, or threat of harm to a child. Additional definitions are offered for each term in bold with specific examples.”

The 2008 CDC report also describes the use of the ICD 9/10 codes for child maltreatment, which vary by hospital and region. The ICD codes provide a standard system for hospitals to code all discharge diagnoses and are used for reimbursement rates in the United States. These codes provide an alternative data source for monitoring the scope of child maltreatment, but it is important to note that many physicians may or may not use them. In addition to recording information about the nature, severity, and physiological location of injury, the ICD data include “E-codes” that provide information about the source or cause of the injury. In the case of child maltreatment, the E-code may identify several characteristics of the perpetrator, including the person’s relationship to the child (Leeb et al., 2008). (See Box 3 for further details about the use of ICD codes in classifying child maltreatment injuries.)

In its first Report to Congress on High Priority Evidence Gaps for Clinical Preventive Services, the U.S. Preventive Services Task Force identified “Interventions in Primary Care to Prevent Child Abuse and Neglect” as one of the high-priority areas in “Behavioral Intervention Research Topics That Deserve Further Research.”

Approximately 1 million abused children are identified in the United States each year. Despite the dedication and hard work of people in many sectors, no one has discovered an effective role for the primary care system and primary care professionals in preventing child abuse and neglect. The Task Force recognizes that the solution to this issue will include many other efforts and hopes that needed research to find effective interventions initiated in primary care will be conducted. Early research suggests that clinician referrals to home visitation by nurses during pregnancy and early childhood may reduce child abuse and neglect in selected populations, but additional research is needed. Future research must examine both the potential benefits and the potential unintended harms of interventions aimed at preventing child abuse and neglect. (Moyer et al., 2011, p. 12; see http://www.uspreventiveservicestaskforce.org/annlrpt/tfannrpt2011.pdf)

Recent controlled studies have followed this lead and the field has begun to identify pediatric primary care-based prevention models that reduce maltreatment reports (Dubowitz et al., 2009).

•  The IOM’s 2002 report entitled Confronting Chronic Neglect: The Education and Training of Health Professionals on Family Violence called for health professional organizations to develop and provide guidance to their members, constituents, institutions, and stakeholders regarding violence and abuse education. Specifically, these recommendations emphasized the need for organizations to provide guidance in (1) competencies to be addressed in health professional curriculums, (2) effective teaching strategies, and (3) approaches to achieving sustained behavior changes among health professionals.

•  The IOM further recommended that health professional organizations identify and disseminate information on approaches for overcoming barriers to training on family violence. Although some progress has been made, training and education about the health problems related to violence and abuse remain highly variable and often marginalized in the curriculums of most health profession schools as well as within the individual practices of physical and behavioral health professionals and the U.S. health delivery system. Even within directly related academic disciplines, such as social work and psychology, specific training or coursework focused on child maltreatment or child welfare may be sparse.

•  Although the governing bodies in some health disciplines have recognized the need for core competencies appropriate to practitioners in their fields, the call for an overarching set of principles remains unmet. The Academy on Violence and Abuse was founded in 2005 to address these concerns and to support actions to achieve the IOM recommendations (http://www.avahealth.org). Increasingly, clinicians are exploring ways to distinguish pediatric trauma that is related to child maltreatment from other forms of injury experienced by children. One study of pediatric injuries drew on data from a 10-year period as recorded in the National Pediatric Trauma Registry (DiScala et al., 2000). The authors sought to highlight distinctive patterns in the nature, severity, survival, and functional outcomes of patients hospitalized for an acute injury in hospitals during the study period. They concluded that those injured by abuse “sustain more severe injuries, use more medical services, and have worse survival and functional outcome than children with unintentional injuries.”

In many cases, child maltreatment involves more than a single act of abuse or neglect. Multiple forms of abuse often co-occur. In some cases, maltreatment may be a routine part of a child’s life rather than a few aberrant events. In other cases, maltreatment may be only one part of a broader social environment that involves other forms of violence (most notably, domestic violence), high interpersonal conflict, substance abuse, parental mental illness, inadequate housing, poverty, chaotic and unpredictable schedules, and bereavement. Significant variations also occur in the extent to which a perpetrator needs to be a parent or caregiver. This relationship is necessary in cases of physical abuse or neglect (because other forms of assault, e.g., peer or sibling violence or school bullying, do not fall within state-based definitions of child abuse). Yet, any form of child sexual abuse is considered within the scope of social service agencies, regardless of the relationships of the offender to the child.

Recognizing the inadequacies of using legal or public health definitions that are focused on the commission or omission of individual events or actions, some researchers have identified maltreatment through child-oriented studies by drawing on multiple fields of research. This approach is built on the assumption that child maltreatment is a set of behaviors and experiences rather than a specific disorder. As noted by Damashek and Chaffin (in press): “The phenomenon of child maltreatment is [composed] of two elements, maltreating behaviors and maltreatment experiences, that together constitute a socially defined problem with mental health relevance.”

This interest in examining experiences, as well as behaviors, has stimulated research in the areas of developmental psychopathology (Cicchetti and Toth, 1995), developmental traumatology (De Bellis, 2001), and, more recently, building a public health surveillance system that can monitor the effects of adverse childhood experiences (Anda et al., 2010). These approaches, and others, are extending the research agenda to focus not only on behaviors and experiences, but also to consider the impact of the disruption of parent-child relationships on underlying biological systems that influence the regulation of stress and trauma. In addition, greater attention is being directed to the effects of trauma on children and their families and interventions that can work effectively within a family context.

•  “Developmental psychopathology is an emerging discipline that seeks to unify, within a developmental, life-span framework, the many contributions to the study of the mood disorders emanating

from multiple fields of inquiry, including psychology, psychoanalytic theory, psychiatry, the neurosciences, sociology, cultural anthropology, and epidemiology.” (Cicchetti and Toth, 1995, p. 373)

•  “Developmental psychopathology represents a movement toward comprehending the causes and determinants, course, sequelae, and treatment of the disorders through its synthesis of knowledge from multiple disciplines within a developmental framework.” (Cicchetti and Toth, 1995, p. 373)

•  “Developmental traumatology is the systemic investigation of the psychiatric and psychobiological impact of overwhelming and chronic interpersonal violence (child maltreatment) on the developing child. This is a relatively new area of study in child psychiatry that synthesizes knowledge from developmental psychopathology, developmental neuroscience, and stress and trauma research.… Active areas of research investigate the consequences of child maltreatment and related family and psychosocial stressors and their effects on the development and regulation of major biological stress response systems and their influence on childhood brain development and function” (De Bellis, 2001, pp. 539-540). The focus on trauma allows researchers to consider not only the act of maltreatment itself, but also the relationship of the victim to the offender (De Bellis, 2001, p. 540).

•  A few surveys, such as the Adverse Childhood Experience Study, have incorporated the developmental approach by scoring respondents on the number and severity of traumatic events (including child maltreatment) occurring during childhood that may affect their adult health status (Anda et al., 2010). Evidence suggests that the aggregate burden of childhood adverse experiences rather than the specific type of adversity may best predict developmental sequelae, emphasizing the importance of viewing maltreatment within a context of associated adversities (Finkelhor, 2008).

•  The National Child Traumatic Stress Network (NCTSN), established by Congress in 2000, is a government-funded network made up of health care providers, community service centers, academic researchers, and families to raise the standard of care for child traumatic experiences and to increase access for families. NCTSN is jointly coordinated by the University of Califor-

nia, Los Angeles, and Duke University, governed by an advisory board (http://www.nctsn.org/), and divided according to three areas’National Center for Child Traumatic Stress, Treatment and Services Adaptation Centers, and Community Treatment and Services Centers. Sample activities include public awareness campaigns, community program development, educator toolkits,and workshops for foster parents (Lott, 2011).

In many cases, however, child neglect may be a chronic experience rather than an accumulation of discrete traumas. Nor is neglect frequently categorized as a type of interpersonal violence. In such cases, it may be the prolonged absence of parenting behaviors (affection, interaction) that lead to significant harm (Dubowitz et al., 2005). Early neglect may also result in intellectual delays due to a lack of appropriate stimulation (Strathearn et al., 2001).

IDENTIFICATION, REPORTING, AND DATA SOURCES

One of the key challenges associated with research on child abuse and neglect is determining the scope and severity of experiences with child maltreatment within the general population. Discrepancies between official reports of child abuse and neglect and other data sources (e.g., health records and household surveys) as well as other methodological challenges have raised basic questions about the scope of the problem, the types of families that are affected, trends over time, and outcomes associated with selected prevention and treatment interventions (Fallon et al., 2010). While government statistics are often based only on victims of substantiated or indicated cases, other administrative records are often available in many states that can provide additional insights into the types and characteristics of the much broader set of cases that are brought to the attention of child protection agencies.

Child Maltreatment 2010 is the most recent governmental report of the number and types of cases of child maltreatment reported to child protection (HHS, 2011). The report relies on official data provided by state agencies through the National Child Abuse and Neglect Data System (NCANDS), which is described below. Key findings from the Child Maltreatment 2010 report include

•  More than 3.6 million (duplicate) children were subjects of at least one report and received one or more dispositions in FY

2010. The duplicate count includes a child each time he or she was included in a report of abuse or neglect during the year. The unique count of child victims counts a child only once regardless of the number of times he or she was found to be a victim during the reporting year. One-fifth of these children were found to be victims, with dispositions of substantiated (19.5 percent), indicated (1 percent), and alternative response victim (0.5 percent).

•  Of the 1,793,724 reports that received an investigation, 436,321 were substantiated (the allegation of maltreatment or risk of maltreatment was supported or founded by state law or policy); 24,976 were found to be indicated (maltreatment could not be substantiated under state law or policy, but there was reason to suspect that at least one child may have been maltreated or was at risk of maltreatment); and 1,262,118 were found to be unsubstantiated (there was not sufficient evidence under state law to conclude or suspect that the child was maltreated or at risk of being maltreated).

•  The most common form of reported victimization is child neglect (78.3 percent). Other forms include physical abuse (17.6 percent) and sexual abuse (9.2 percent).

•  Per 1,000 children, 9.2 were reported to be “unique victims” of child abuse and neglect in the total population of U.S. children. The number of nationally estimated duplicate victims was 754,000 and the number of nationally estimated unique victims was 695,000.

•  Victims in the age group of birth to 1 year had the highest rate of reported victimization at 20.6 per 1,000.

•  About 1,560 children died from abuse and neglect, based on data from state reports.

•  Eighty-eight percent of victims were composed of three races or ethnicities—African American (21.9 percent), Hispanic (21.4 percent), and White (44.8 percent).

The statistics derived from these data offer important insight into the ways in which different states handle reports of child abuse and neglect. Significant questions persist about the extent to which the frequency, severity, and other characteristics of the reported cases resemble those of cases of child maltreatment within the general child population that are not reported to child protection agencies. To reconcile questions about the quality of the data based on reported cases, some researchers have

compared data based on administrative records with other data sources, such as those drawn from child health and injury records. For example, Gilbert et al. (2012) published a study in Lancet that compared indicators of child maltreatment in six developed countries and provinces (England [United Kingdom], Manitoba [Canada], New Zealand, Sweden, the United States, and Western Australia). They observed:

Existing research on how child maltreatment is changing in developed countries is conflicting. Studies that rely on officially recorded or substantiated maltreatment measure only a small part of the bigger picture—for example, in some settings as few as one in 30 of the children who experience physical abuse every year have their abuse officially recognized. One reason is that most child maltreatment is hidden and not recognized by professionals dealing with children. Another reason is that health, education, and other community professionals in contact with children consistently report to child protection agencies only a proportion of children whom they recognize as being maltreated. Therefore, studies based on self-reported or parentreported incidents of maltreatment come closest to measurement of the occurrence of maltreatment, although these studies might still underestimate the scale of the problem. However, many of the events identified in self-report studies might not be sufficiently severe to require intervention. (Gilbert et al., 2012, p. 2)³

The authors of the Lancet article highlight several concerns in conducting research on the identification, disclosure, and reporting of child abuse and neglect. First, government policies exert strong influence on the types of contacts and reports made to child protection agencies, which in turn affect the official records used to construct national databases of reports of abuse and neglect. Second, health records maintained by hospitals and other medical centers provide important sources of data about maltreatment-related injury or violent death for young children. Third, professionals who serve children and families who are at risk of

______________________

³It is useful to note that in the United States, mandatory child abuse reporting laws are required by federal law and limit the extent to which children and their parents can be directly surveyed about maltreatment behavior or experiences. Certificates of Confidentiality, issued by federal agencies to protect research participants from required disclosure of other types of sensitive research data, do not extend to disclosure of research data about identifiable maltreatment victims, and therefore self-report studies can involve exposing parents to jeopardy.

maltreatment often are aware of many behaviors that may constitute abuse or neglect, but only a proportion of these behaviors (usually the most severe cases) are reported to child protective services (CPS) officials. Fourth, many instances of abuse or neglect may be known only by parents or children themselves. One of their key findings is: “Large differences between countries in the rate of contacts with child protection agencies contrasted with little variation in rates of maltreatment-related injury or violent death.… Overall, one or more child protection agency indicators (notification, investigation, officially recognized physical abuse or neglect, or out-of-home care) increased in five of six countries and states, particularly in infants, possibly as a result of early intervention policies” (Gilbert et al., 2012, p. 1). This finding suggests that child protection agencies are increasingly responsive to cases that involve lower levels of severity than in prior years.

The field continues to grapple with the difficulties of reconciling rates of child victimization from different data sources, including those that rely on voluntary state-based administrative data systems (e.g., reports generated by the NCANDS), those that involve nationally representative surveys (e.g., the NIS studies), and those that rely on health records that require some judgment in classifying selected injuries as child maltreatment cases (e.g., the Gilbert et al. [2012] study). It is useful, therefore, to consider the characteristics of different data sources used to estimate the scope and severity of child maltreatment. Each of the following sections highlights the key data sources used to develop indicators of child abuse and neglect. These data sources include official reports, health records, and research studies that include periodic household surveys as well as those that follow one population cohort over time.

Official Reports

Official sources of data on child maltreatment draw on reports from CPS and child welfare agencies (at both the county and state levels); interviews with personnel in human services agencies; “sentinel reports” from persons such as teachers, health care providers, and others who are in frequent contact with children and who may observe injuries or other signs of maltreatment; and vital statistics (which provide death records).

Several reports and information systems are based on official reports from state or county agencies. These include

National Incidence Studies (NISs)^*4 The four NISs (NIS-1,-2,-3, and-4) are legislatively mandated and have been conducted since 1979 under contract to the ACF. The results of the most recent study (NIS-4) were reported to Congress in January 2010 and were authorized by Public Law 108-36 in 2003 (Sedlak et al., 2010b). Data for NIS-4 were collected over two 3-month periods in 2006 from two sources in a nationally representative sample of 122 counties in the United States: CPS and “sentinels” who include professionals working in the same counties in human services and health agencies as well as a variety of other settings, including education, public housing, law enforcement, judiciary, child care agencies, and shelters for runaway and homeless youth or victims of domestic violence.

National Child Abuse and Neglect Data System^* NCANDS is a voluntary data reporting system, authorized by legislation (Public Law 93-247, as amended). NCANDS draws on data provided only by CPS and include two components: the Summary Data Component (SDC) and the Detailed Case Data Component (DCDC). SDC is a compilation of key aggregate child abuse and neglect statistics from all states, including data on reports, investigations, victims, and perpetrators. DCDC is a compilation of case-level information from child protective services agencies drawn from electronic child abuse and neglect records (all states, with the exception of Oregon, and the District of Columbia and Puerto Rico provided detailed case data in FY 2010; Oregon provided summary data). The DCDC consists of two data files that (1) collect information about the characteristics of all children included in a report of alleged maltreatment as well as characteristics such as the source and disposition of the report (known as the Child File), and (2) information about the type of maltreatment, the support services provided to the family, and any special problems that were identified for the child, caretaker, or family (known as the Agency File). Only children identified as substantiated or indicated victims are included in the Child and Agency Files. NCANDS does not collect data on the alleged perpetrators, and reports of child fatalities are excluded from both files. Data drawn from the state reports within NCANDS provided the basis for the annual Child Maltreatment reports prepared by the ACF (HHS, 2011).

______________________

⁴Data from studies marked with the asterisk (*) are archived at the National Data Archive on Child Abuse and Neglect at Cornell University (http://www.ndacan.cornell.edu).

Adoption and Foster Care Analysis and Reporting System (AFCARS)^* AFCARS is a third legislatively mandated source of data collection. AFCARS is an automated information system that relies on semiannual reports by the states, which provide information on (1) all children in foster care for whom state child welfare agencies have responsibility for placement, care, or supervision; (2) children who are adopted under the auspices of the state’s public child welfare agency; and (3) information on foster and adoptive parents. The Children’s Bureau has developed an assessment review process for the purpose of more fully assessing and evaluating states’ AFCARS. The AFCARS Assessment Review process is a “validation and verification” of the automated information system to ensure that it can provide valid and reliable data (see http://www.acf.hhs.gov/programs/cb/systems/afcars/about.htm for further information).

Statewide Automated Child Welfare Information Systems (SACWIS) Unlike the three data sources described above, SACWIS is a case management system that most states use as a tool and technical resource. SACWIS allows agencies to integrate multiple data sources and facilitate the delivery of child welfare support services, including family support and family preservation. Fifteen states (including California) do not yet have operational SACWIS information systems; a number of these states rely on non-SACWIS models to manage their child welfare information sources.

SACWIS allows the states to receive enhanced federal funding to develop and implement their case management files, and by law, a SACWIS is required to support the reporting of data to AFCARS and NCANDS. States are also encouraged to link their child welfare services case information with other federally supported programs, such as Title IV-A (Temporary Assistance for Needy Families), Title XIX (Medicaid), and Title IV-D (Child Support) systems, among others (see http://www.acf.hhs.gov/programs/cb/systems/sacwis/about.htm for further information). In a few states (most notably California, Missouri, and Washington), researchers have been able to draw on these types of administrative data sources to examine the service histories of children and families that come into contact with child protection agencies.

National Crime Victim Survey (NCVS) In addition to the data drawn from child protection and child welfare agencies, the U.S. Census Bureau conducts an annual crime victim survey on behalf of the U.S. Department of Justice. This household survey uses rigorous definitions and re-

petitive interviews to ensure the validity of the information provided by the respondents and to avoid duplication of reports. NCVS is the nation’s primary source of crime victim statistics. Interviews are conducted with a nationally representative sample of 76,000 households regarding the frequency, characteristics, and consequences of criminal victimization in the United States. The survey includes interviews with 10,000 youth ages 12 to 17, who report victimization experiences such as violence in school, neighborhood, or family settings, including bullying (see http://www.icpsr.umich.edu/icpsrweb/NACJD/NCVS/#About_NCVS for further information).

Sources Based on Health Records

Data about abuse and neglect may be drawn from surveys of health care providers and patients as well as health services records, such as hospital discharge information, clinical records, and insurance databases.

ICD codes for child injury and fatality data As noted in Box 3, the CDC collects data from hospital records that are coded according to ICD-9 and ICD-10. Several research studies have demonstrated difficulties with the data sources that are used to identify child fatalities and injuries that result from maltreatment.

For example, Crume et al. (2002) conducted a review of all cases involving the death of all children in Colorado (ages birth to 16 years, who died between January 1, 1990, and December 1, 1998) and identified those deaths that were a result of maltreatment, as defined by a statewide child fatality review team (CFRT). The authors concluded that the child deaths in Colorado that were coded for maltreatment (149 deaths) were nearly half of those fatalities that the CFRT classified as resulting from abuse and neglect (295 deaths) (Crume et al., 2002). They noted several difficulties:

Problems with using death certificate data to estimate child maltreatment deaths stem in part from limitations in the ICD-9 and ICD-10 coding system. In our study, of the 295 maltreatment deaths, only 16 (5 percent) were coded with N995.5 and 42 (14 percent) were coded with E967. The concern with using these codes is that they are not specific for child maltreatment and include homicides that would not be considered maltreatment, e.g., gang violence between teens.(p. 5)

Schnitzer et al. (2011) have noted other instances of state-level differences in the public health surveillance of fatal child maltreatment.

Healthcare Cost and Utilization Project Kids’ Inpatient Database (KID) Gilbert et al. (2012) also sought to use health records to identify child deaths or serious injuries that resulted from violence by parents or other caregivers (physical abuse) as well as those that were perpetrated by other adults or children and, but not always, reflect inadequate supervision (neglect). Their study involved an analysis of coded data from a large dataset provided by 2,521-3,739 hospitals in 22-38 states derived from four 1-year periods (1997, 2000, 2003, and 2006, from KID) (AHRQ, 2011). The state-reported data are then compiled by the National Center for Health Statistics as part of a national dataset.

The analysis focused on younger children (ages 0 to 11 years) “because injuries related to physical assault or neglect in older children are more likely to be due to peer, sibling, or stranger violence, or to adverse environments, than to be related to parental or caregiver violence or poor supervision” (Gilbert et al., 2012, p. 761) and included coded data from four categories:

•  “Maltreatment syndrome (i.e., [ICD] codes directly referring to abuse or neglect or a perpetrator of abuse);

•  Assault;

•  Injuries of undetermined cause; and

•  Codes reflecting concern about adverse social circumstances that are indicators of neglect or broader welfare concerns (e.g., problems related to the social environment, family support, upbringing, or lifestyle).” (Gilbert et al., 2012, p. 762)

The analysis by Gilbert et al. (2012) suggests that the rates of violent deaths and maltreatment-related injury have remained stable in the United States and the other countries that were included in their analysis since the mid-1990s, although decreases in violent deaths coincided with decreases in admission related to maltreatment injury in both Sweden and Manitoba.

The use of KID offers an additional tool for building epidemiological data on child maltreatment. For example, hospitalizations due to serious injury can be further distilled to identify specific characteristics of physical abuse, assault, or child battery that can offer added insights for tracking child hospitalization trends and effects of prevention programs. Certain limitations exist, however, as the information derived from KID

refers to hospitalizations per year, rather than per child, and physicians’ use and accuracy of ICD-9 diagnostic codes for child abuse cannot be determined (Leventhal et al., 2012).

Longitudinal Studies and Household Surveys

Research studies frequently rely on individual interviews or responses to survey questionnaires, which in some cases may be part of an ongoing longitudinal study. Data regarding personal experiences with abuse and neglect are usually collected from parents or older children, or through retrospective reports from adults who may have experienced maltreatment when they were children. Retrospective reports are, of course, not equivalent to contemporaneous determinations, and may be particularly problematic for certain types of maltreatment, such as neglect during preschool years (which is perhaps the single most common form of maltreatment). Biological materials are collected in some studies, especially genetic and hormonal samples (e.g., cortisol) that are thought to be related to reactions to stress and trauma. Observational studies are virtually nonexistent as a data source in the child maltreatment literature.

Longitudinal Studies

Longitudinal Survey of Child Abuse and Neglect (LONGSCAN) ^* LONGSCAN is a consortium of research studies that were initiated in 1990 with multiple federal grants. The consortium consists of five separate cohort studies conducted by satellite sites (Chicago, Baltimore, North Carolina, San Diego, and Seattle) that are coordinated through a center at the University of North Carolina. Each site conducts a separate and unique research project on the etiology and impact of child maltreatment using different child and family populations. For example, the Chicago study compares the life course of infants among three different types of families: (1) those who received comprehensive services after a report of child maltreatment; (2) those who received follow-up by the state welfare agency; and (3) a control group of matched infants. The San Diego study is a cohort study that initially followed maltreated children who were placed in foster care in the first 18 months of life and followed until age 4. This sample was recruited into LONGSCAN at age 4 to study outcomes of kinship versus nonfamily foster care, the consequences of reunification, and the use and impact of healthcare and mental health services. Papers currently in press that draw on LONGSCAN data

are focusing on topics such as identifying children at high risk of child maltreatment; the influence of caregiver network support and caregiver psychopathology on child mental health need and service use; and suicidal ideation in adolescence (http://www.iprc.unc.edu/longscan/pages/publist/index.htm).

North Carolina Cohort within the LONGSCAN study The North Carolina cohort specifically focused on high-risk infants based on poor prenatal care, serious medical complications, low birthweight, or external influences, such as maternal substance abuse, maternal health and mental health, and household structure (i.e., single parenthood). Researchers compared interviews with mothers and reports to North Carolina’s Registry of Child Abuse and Neglect to examine the reliability of predictive factors of first-year maltreatment reports and potential interventions to reduce the risk of second-and third-year maltreatment reports (Hunter and Knight, 1998; Kotch et al., 1997).

National Children’s Study (NCS) The NCS was authorized by the Children’s Health Act of 2000 and is sponsored by a collaboration among four federal agencies: the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), the National Institute of Environmental Health Sciences, the CDC, and the Environmental Protection Agency. The NICHD is the lead agency for the NCS. The study design involves a total of 36 study centers that include 105 study locations covering 79 metropolitan areas (urban, suburban, and small cities) and 26 rural communities (see http://www.nationalchildrensstudy.gov/studylocations/pages/overview.aspx).

The study objective is to recruit and follow a nationally representative sample of 100,000 children from before their birth until age 21 years, examining the effects of the physical, chemical, and social environments on the growth, development, and health of children across the United States. The study includes attention to family dynamics, community and cultural influences, and genetics. The goal of the study is to improve the health and well-being of children and contribute to understanding the role that various factors have on health and disease. The NCS has not yet published work on data related to child abuse or neglect, although planning efforts indicated that (1) investigation of the causes and consequences of child maltreatment was an appropriate scientific hypothesis to include in the study, and (2) the study of these issues justified the need for a large (100,000+) cohort and longitudinal design:

To document the “natural history” of child maltreatment and to understand how environmental, child, and parent characteristics influence occurrences of child maltreatment and subsequent child development, large-scale prospective longitudinal research, such as the NCS, is required.… The ability to identify early markers of problematic parent-child interactions and factors that contribute to the likelihood of child maltreatment across different stages in children’s and families’ lives will provide invaluable information for the timing and delivery of cost-effective services to prevent child maltreatment.… The NCS also can provide information about the timing, dosage, and content of interventions necessary to address the consequences of child maltreatment and facilitate healthy child development through the study of interventions occurring within the sample and through using the NCS cohort as a control group in prevention and intervention research involving independent samples (Lewin report, http://www.nationalchildrensstudy.gov/about/organization/advisorycommittee/2003Sep/Pages/Injury-document-1.pdf).

Dunedin study The Dunedin Multidisciplinary Health and Development Study (“Dunedin study”) has followed an original cohort of 1,037 babies born in Dunedin, New Zealand, between April 1, 1972, and March 31, 1973, for nearly 40 years (Jaffee et al., 2007; Ouellet-Morin et al., 2011). The study is housed in the Department of Preventive and Social Medicine, Dunedin School of Medicine at the University of Otago and is funded primarily by the Health Research Council of New Zealand. The Dunedin study is notable for its capacity to capture both biological (including genetic) materials as well as social and environmental measures. As noted by AlEissa et al. (2009), the Dunedin study “provides a rich source of information on child abuse, the causes of antisocial behavior and resulting life course outcomes. Projects of particular interest from a child protection context will be research into the relationship between genetic and environmental factors, and how they interact to predispose people to conditions like hyperactivity, violence, and alcoholism” (p. 3). (See http://dunedinstudy.otago.ac.nz/index.xhtml for further information.)

Widom study An older but continuing longitudinal study has been conducted by Cathy Spatz Widom and associates (Widom, 1999; Widom et al., 2007; Yanos et al., 2010). They identified a sample of 900 children who had been abused or neglected before age 11 and compared them with a sample matched on age, gender, race, and place of residence. A 20-year follow-up survey of the original cohort provided key findings

regarding the impact of earlier maltreatment experience with juvenile violence as well as adverse health outcomes (Yanos et al., 2010). Farrington (2011) referred to the Widom survey as “the most famous study of child abuse and neglect” (p. 138).

Jonson-Reid study A more recent study with a service system perspective was designed by Jonson-Reid and associates with a sampling frame intentionally patterned after the Widom study (Jonson-Reid et al., 2012; Lanier et al., 2010). They identified a sample of children reported for abuse and neglect in 1993-1994 prior to the age of 12 in families receiving Aid to Families with Dependent Children and matched this to a sample of children in families by birth year and residential location with children in families with no histories of such reports. The study of more than 12,000 children uses administrative records from child welfare, income maintenance, special education, juvenile court and highway patrol, vital statistics, health, mental health, Census data, corrections, hospital records, and runaway shelters. Children were followed using dated records through 2009, when subjects were ages 15 to 27.

Minnesota Longitudinal Study of Parents and Children In 1975, the University of Minnesota’s Institute of Child Development organized a study that observed 267 first-time mothers from their third trimester of pregnancy and followed the families through the children’s early adulthood (age 28). The researchers conducted a wide variety of assessments, including behavioral, environmental, and social factors, to identify themes about the course of individual development (Stroufe et al., 2005). An in-depth analysis of high-risk children and their families traced the pathways between maltreatment and antisocial behavior, and concluded that alienation in early childhood—whether through neglect or abuse—was linked to impaired self-regulation of emotion (i.e., antisocial behavior) in later life (Egeland et al., 2002).

Romanian orphan studies Another body of research is often used to examine the short-and long-term effects of global deprivation and grossly inadequate institutional care on the health and development of young children (Bos et al., 2009; Nelson et al., 2007). The study population includes a group of infants and young children who resided in public orphanages in Bucharest, Romania, some of whom were subsequently placed in foster care homes in the same region or adopted by parents in the United Kingdom or the United States. Several studies have been conducted with selected samples of the Romanian orphan population, including the Bucharest Early Intervention Project in the United States (Zeanah

et al., 2003, 2005) and the English and Romanian Adoptees Study in the United Kingdom (Colvert et al., 2008). These studies strive to identify areas of cognitive, emotional, and biological functioning and regulation that are especially susceptible to inadequate care during the child’s early years of development, as well as highlighting those areas that are particularly resilient and respond positively to health and caregiving interventions.

Household Surveys

National Survey of Child and Adolescent Well-Being (NSCAW)^* NSCAW is a national study of children who have come into contact with the child welfare system. Congress authorized the study in the Personal Responsibility and Work Opportunities Reconciliation Act of 1996, and NSCAW is supported by the ACF. The study sample is drawn from administrative records, which are used to identify families who are then invited to participate in household surveys. Firsthand reports are obtained from children, parents, and other caregivers; reports from caseworkers and teachers; and data from administrative records. NSCAW includes a longitudinal component that follows cases for several years and collects data on the types of abuse or neglect involved, agency contacts and services, and out-of-home placements. The study also includes child and family well-being outcomes in detail and explores the relationship between those outcomes and experience with the child welfare system, family characteristics, community environment, and other factors. Data have been collected in two waves, NSCAW I and II. NSCAW I involved data from 5,501 children (ages 0 to 14) from 97 child welfare agencies nationwide, who entered the child welfare system within a 15-month period (October 1999-December 2000); NSCAW II involved 5,873 children, from 83 counties nationwide, who ranged from birth to 17.5 years old at the time of sampling and who entered the child welfare system between February 2008 and April 2009. NSCAW I also included a supplemental sample of 727 children who have been in foster care for a longer period to support additional analyses. A series of 16 research briefs have been prepared that present findings based on NSCAW. The study is conducted through a collaborative effort among the Research Triangle Institute,Walter R.MacDonald Associates,and Caliber Associates (http://www.acf.hhs.gov/programs/opre/abuse_neglect/nscaw/reports/summary_nscaw/nscaw_research_brief_main_findings.pdf).

Developmental Victimization Survey (DVS)^* Finkelhor and colleagues at the University of New Hampshire (2005) conducted a household survey to estimate the range of victimization experiences (including physical and sexual abuse) by children and youth from ages 2 to 17. The survey is unique in that older youth are able to provide firsthand, confidential accounts of their own experiences. The sample population includes 2,020 children. Results indicate that “over half of all children experienced a physical assault in the course of the previous year, much of it by siblings and peers” (Finkelhor, 2007, p. 18). These findings present a stark contrast to the narrower and more rigorous results of the NCVS, which the authors attribute to the likelihood that the DVS may include incidents that “observers might dismiss as ’not real crimes,’ such as sibling and peer assaults and disciplinary acts” (p. 18), including being beaten by a parent.

National Survey of Adolescents This survey was a household study conducted in 1995 by the National Crime Victims Research and Treatment Center at the Medical University of South Carolina, and funded through the National Institute of Justice and the CDC. American youth, ages 12-17 and living with a parent or guardian, were surveyed using Computer-Assisted Telephone Interviewing to test hypotheses regarding relationships among victimization experiences, mental health effects, substance abuse, and delinquent behavior in adolescents (Kilpatrick and Saunders, 1999, 2009). Findings from the survey point to nationwide correlations between child maltreatment and subsequent effects on mental health, rate of future victimization, and delinquent behaviors (Hanson et al., 2006; Kilpatrick et al., 2003; Knopf et al., 2008; Waldrop et al., 2007).

Measures and Methods

One of the research priorities recommended in the 1993 NRC report (NRC, 1993) highlighted the need for “reliable and valid clinicaldiagnostic and research instruments for the measurement of child maltreatment” (p. 345). Nearly 20 years later, broad fragmentation persists in the scope and quality of measures that are routinely used in studies of child maltreatment.

For example, a recent (December 2011) search of the measures index organized by the National Data Archive on Child Abuse and Neglect (NDACAN) at Cornell University (http://www.ndacan.cornell.edu/abis/abisMeasuresIndex.cfm) identified 367 individual measures that

have been used in the child maltreatment studies available within the Data Archive. Only 69 of these measures are commonly used (defined as used in five or more studies). Nearly half of the remainder are limited to one-time use (163 of 367 measures), and a third are dedicated solely to LONGSCAN and NSCAW measures (110 of 367 measures). For those measures that are used in five or more studies, half (38 of 69) are used only in the studies directed by the same research team. In particular, most of the datasets using LONGSCAN and NSCAW measures were intended for studies directly related to these measures, respectively.

Measures that are used in studies that are not archived in NDACAN have not been indexed (e.g., the Dunedin study or the studies of the Romanian orphans). The extent to which these longitudinal efforts rely on measures or sampling procedures that are common in other child maltreatment studies is not known. The author is not aware of how this analysis of measures used in child maltreatment studies compares with other social or behavioral studies that focus on child development or family processes and relationships. However, the broad and persistent fragmentation of measures across multiple studies inhibits comparative studies and presents a continuing challenge to investigators who wish to pool data across multiple samples in order to increase the power of statistical analyses.

CAUSES

Although initial studies focused on single risk factors for child maltreatment (e.g., parent’s mental health, poverty, or parent’s personal history of maltreatment), more recent work has drawn on developmental research that places the causes of abuse and neglect within a multilevel framework involving both environmental and individual factors. These studies draw on early ecological-developmental theory by Bronfenbrenner (1979), which Belsky (1980), Cicchetti and Rizley (1981), and others used to portray an etiological-transactional model for development that described interactive factors operating across multiple environmental, contextual, familial, and individual ecologies.

The etiological-transactional model has been further adapted to suggest that maltreatment emerges from a disordering of the balance between positive (“potentiating”) and negative risk factors that may occur within and across each of the four levels of the model (Cicchetti and Valentino, 2006). The transient or enduring presence of these disruptions

are thought to affect not only the selected ecological level, but may also influence processes in the surrounding levels as well. As stated by Cicchetti and Valentino (2006): “These dynamic transactions, which operate both horizontally and vertically throughout the levels of ecology, determine the amount of risk for maltreatment that an individual faces at any given time” (p. 134).

The complexity of these behaviors and experiences suggests that a broad set of causal and contributing factors is involved, including not only the presence of certain risk factors, but also the absence of protective or positive assets that can prevent the occurrence of abuse and neglect. Additional complexity results from the need to consider not only the individual risk and protective factors for the child as well as the parent or offending caregiver, but also to consider how these factors influence the relationship and interactions between the adult and child in multiple settings.

Research studies have also traditionally built on a framework that involves three general categories: the parent’s contribution, the child’s contribution, and social context (Belsky and Vondra, 1989). The parent’s contribution might include the psychological state of the individual parent or caregiver (e.g., anger or depression) or behavioral disorders (e.g., substance abuse or domestic violence). The child’s contribution includes factors such as age, temperament, or birthweight (e.g., infants, and lowbirthweight infants in particular, are at greater risk of maltreatment than older children). Social context might include factors that stress or support parents in their caregiving experiences and interactions, such as poverty, unemployment, or inadequate housing.

Multiple waves of the NISs (Sedlak et al., 2010b) and a range of other studies (Coulton et al., 2007; Drake and Pandey, 1996) have demonstrated that poverty is among the strongest predictors of child maltreatment. However, the actual mechanisms by which poverty leads to maltreatment are still under study (Slack et al., 2004). In addition, the role of race in child maltreatment is not fully settled. Administrative data and the most recent wave of the NIS correspond with official report data (Sedlak, 2010a) showing that African Americans are reported at a higher rate than Whites, while Hispanics are not. The findings of higher risk among African Americans are consistent with the higher economic risks faced by that population (Sedlak et al., 2010a), while the lower rates experienced by Hispanics are consistent with the concept that certain cultural factors are protective for Hispanic children, known as the “Hispanic paradox” (Drake et al., 2011). Several regional studies focused on levels

of social organization in poor neighborhoods have concluded that Black children are at greater risk than White children, even when controlling for the overrepresentation of Black children among poor families (Coulton and Korbin, 1995; Korbin et al., 1998; Wulczyn, 2009).

More recent attention has been directed toward situational factors that may affect family life (e.g., family disruptions or turmoil, social isolation, or movement of unrelated adults into a child’s household) (Finkelhor, 2008). However, limited attention has been directed toward characterizing the presence or absence of positive and protective factors, such as highlighting the importance of a safe and stable relationship with an adult caregiver who can nurture and protect the child when a parent is unable to do so. The “bundling” of child maltreatment with other life adversities (Damashek and Chaffin, in press) consistently challenges studies of the underlying factors that contribute to abuse and neglect. Such adversities may include poverty, intimate partner violence, substance use, inadequate housing, unemployment, parental illness, mental health disorders, and family dysfunction. Certain psychological, psychiatric, or behavioral factors may precipitate some forms of maltreatment (e.g., the relationship between pedophilia and child sexual abuse or the link between parental depression or substance use and physical abuse). In other cases, such as child neglect, the overall cumulative burden of adversity, rather than one precipitating factor, may be the more significant contributor to the offending events.

Early studies suggested an intergenerational cycle was involved in abusive and neglectful behaviors, with children who were initially victims becoming offenders as they became responsible for the care of their own children (Egeland et al., 1988). Yet this finding is inconsistent, with later studies failing to support the intergenerational continuity of child physical abuse (Widom, 1989). A critique of these and other studies of international continuity has highlighted the methodological challenges associated with establishing such causal relationships, especially when intervening factors (e.g., sociodemographic characteristics during different times of abuse) are not addressed (Ertem et al., 2000).

More recent studies have given particular attention to the biological and psychological underpinnings that are associated with individual transactions between certain caregiving contexts and parenting behavior, suggesting that a caregiver’s own early experience with abuse, deprivation, or trauma as a child may shape particular regulatory systems in the brain or other biological systems that either disrupt their capacity to nurture or soothe a child when under stress, or trigger abusive behaviors that

they experienced during their own childhood. These “maladaptations,” which may persist across generations, may consequently encourage behaviors and representations of self and caregiver practices, including unrealistic expectations of their children, that can contribute to future acts of maltreatment (Cicchetti and Toth, 2005).

Recognizing that children are victimized in numerous ways besides maltreatment (e.g., bullying or pornography), some researchers have proposed a new framework to support “a holistic approach to child victimization” (Finkelhor, 2008, p. 21). On this same page, Finkelhor has suggested the term “developmental victimology” to guide this holistic approach. His work gives particular attention to neighborhood and residential factors that contribute to victimization, including frequent moves; family loss, conflict, and turmoil (that might be associated with health problems or inadequate housing); greater exposure of children to unfamiliar, unrelated, and potentially predatory or aggressive people; and compromises in child supervision that result from these circumstances (Finkelhor, 2008). Some children may also exhibit emotional deficits and difficulties that make them more vulnerable to victimization, especially when such problems lead to “dependent, sexualized, or indiscriminately affiliative behavior” or “a sign of vulnerability that serves to attract offenders” (Finkelhor, 2008, p. 53).

Others have proposed a model, built on the disruption of basic regulatory processes associated with stress and adversity, that draws on research from the biological, behavioral, and social sciences. Several committees of the American Academy of Pediatrics (AAP), for example, have recently published a report that focuses on the “new morbidities” that affect the health and well-being of today’s children—noninfectious and prevalent societal changes (e.g., the growing numbers of single parents and families with two working parents). The AAP report emphasizes that

it is not adversity alone that predicts poor outcomes. It is the absence or insufficiency of protective relationships that reinforce healthy adaptations to stress, which, in the presence of significant adversity, leads to disruptive physiologic responses (i.e., toxic stress) that produce “biological memories” that increase the risk of healththreatening behaviors and frank disease later in life. (Garner et al., 2012, p. e225)

Research on causes and contributing factors for abuse and neglect frequently draws on theoretical models to identify pathways and relationships among diverse processes as well as opportunities for preventive interventions that can discourage risk exposures and enhance opportunities for protective or buffering interactions. Early models drew heavily on the psychological and parent-child studies of Urie Bronfenbrenner (1979; Bronfenbrenner and Morris, 2006) and others that organize these relationships according to their proximity to the child. Other psychopathology models, such as Cicchetti’s work (Cicchetti and Toth, 1995, 2005), focused on the individual’s understanding of and exposure to a shifting balance between risk and “potentiating” factors that contribute to or diminish the probability of maltreatment. More recently, work by Finkelhor (2008) draws on and adapts other criminal research models, suggesting a dynamic multistage model that involves instigation processes, selection processes, and protection processes, organized on levels that separate individual victim characteristics from those of the surrounding social and physical environment.

Although each of these approaches contributes to our understanding of the causes of child maltreatment, no particular model fully explains the broad array of behaviors associated with different forms of abuse and neglect. Yet, studies of the newer versions of the etiological-transactional model suggest that there may be opportunities to identify the specific processes by which disruptions occur in the “average expectable environment for promoting normal development” and that a range of environmental conditions may occur by which it is possible to support “normative developmental processes” (Cicchetti and Valentino, 2006, p. 120).

CONSEQUENCES

Early studies of the consequences of child maltreatment focused on the acute, often severe, physical effects of abuse and neglect, such as fractures, head injuries, and death. The X-ray images of repetitive fractures resulted in an effort by Henry Kempe and his colleagues to describe the “battered child syndrome” as a specific area of concern in pediatric research and healthcare services (Kempe et al., 1962). This work most notably contributed to the initial passage of CAPTA and the creation of a federal program to support state-based child protection and child welfare services.

While initial studies focused on the immediate and severe physical injuries associated with child abuse and neglect, more recent work has focused on long-term consequences of repetitive adverse experiences, particularly those associated with the stress and trauma of chronic maltreatment. Gunnar et al. (2006, p. 652, citing the work of Manly et al., 2001) has noted that hundreds of research studies have demonstrated “unequivocal evidence that childhood maltreatment increases the risk of psychological and behavioral disorders.” Shenk et al. (2010) state that the developmental correlates of childhood maltreatment include increased aggression, emotion dysregulation, anxiety, depression, and posttraumatic stress disorder (PTSD) (Cicchetti and Rogosch, 2001; Kaufman et al., 1997; Paolucci et al. 2001; Shields and Cicchetti, 1998; Shipman and Zeman, 2001; Trickett et al., 1998).

While significant attention has been directed toward the mental health consequences of abuse and neglect, multiple studies have indicated that maltreated children—especially those placed in foster care—also experience a high level of physical health problems that are frequently untreated and that also involve developmental and educational challenges (for a review, see Horwitz et al., 2010). Specific examples of these physical and psychological health problems include growth failure, obesity, asthma, vision and dental problems, and a range of chronic medical diseases (Christian and Schwarz, 2011). Prospective studies have also found that maltreated children have higher rates of participation in special education than matched comparisons (Sullivan and Knutson, 2000). Research on child welfare populations indicates that many of these delays are evident in early childhood and go unserved (Stahmer et al., 2005).

Estimates of the range of mental health and behavioral problems for maltreated children and those in foster care range from 50 to 80 percent (Christian and Schwarz, 2011). For older adolescents, these rates are particularly high. One study suggests that older adolescents in the foster care system (many of whom have been placed there because of maltreatment histories) have “rates of major depression and posttraumatic stress disorder [that] are 2 to 3 times greater than in the general population” (Christian and Schwarz, 2011, citing McMillen et al., 2005). Yanos et al. (2010) and others have indicated that prior history of abuse or neglect is associated with greater use of mental health and social services in adulthood, and that such service use is frequently mediated by a psychiatric disorder, such as PTSD or depression.

In exploring the underlying causes of such disorders, researchers are examining the developmental and biological effects of maltreatment that

may contribute to persistent and chronic disorders. These studies draw on new frameworks and imaging technology that illustrate the complexity of early human development, focusing on the array of positive and negative effects that occur through interactions among individual biology (including genetic makeup), social environments, caregiving experiences, and physiological and psychological processes. Brain imaging studies have demonstrated that maltreatment experiences may affect specific areas of functioning, particularly in domains such as cognition, social learning, memory, and threat perception. Researchers studying the origins of many adult diseases have focused on the importance of early experience in shaping the neurological and hormonal pathways through which individuals handle stress and physical and emotional threats. These studies conclude that “advances in a wide range of biological, behavioral, and social sciences are expanding our understanding of how early environmental influences (the ecology) and genetic predispositions (the biologic program) affect learning capacities, adaptive behaviors, lifelong physical and mental health, and adult productivity” (Garner et al., 2012, p. e224). In addition, researchers are studying processes associated with arousal, sleep patterns, and infection and immune functions that may be affected by maltreatment experiences.

As noted by Gunnar et al. (2006), two lenses initially dominated studies of the psychological and behavioral effects of maltreatment: the developmental-organizational perspective and the social learning perspective. The developmental-organizational perspective focused attention on stage-salient effects, such as disruption of attachment relationships, or the development of a disorganized/disoriented (Type D) attachment. Both could contribute to poor outcomes in childhood and beyond (Lyons-Ruth, 2003, cited in Gunnar et al., 2006).

The social learning perspective emphasized the ways in which specific internalizing or externalizing behaviors by the child that represent adaptive responses to abuse or neglect may interfere with future peer relationships or adult-child interactions (e.g., with teachers). Such adaptive “learned” behaviors can be disruptive and contribute to social isolation, and also present risks for future engagement with antisocial peer groups, externalizing behaviors, conduct problems, and substance abuse (Egeland et al., 2002, cited in Gunnar et al., 2006). This perspective is consistent with the observation that many children who experience early abuse or neglect are especially vulnerable to other problematic behaviors, such as conduct disorder or aggression. The concept of a “cycle of violence” has also emerged to describe the intergenerational effects of abuse

and neglect, in which young people who experience maltreatment are at risk of using these same behaviors with their own children (Widom and Brzustowicz, 2006).

A third perspective, which has more recently emerged in the research literature, has called attention to the interactions among early experiences, the neurobiology of stress, and brain development. These studies rely largely on animal models (primarily rodents) that allow experimental manipulation of experiences with adversity in adult-juvenile interactions (including biological and “adoptive” mothers) during selected developmental stages, including prenatal and early infancy. These early experience studies have shown that “early parental care profoundly influences brain development, regulates gene expression, and shapes the neural systems that in humans are involved in vulnerability to affective disorders in response to later stressful life events” (Gunnar et al., 2006, p. 653).

Teicher et al. (2003) have described “a cascade of neurobiological events” produced by early severe stress and maltreatment, including the potential to cause enduring changes in brain development. “These changes occur on multiple levels, from neurohumoral (especially the hypothalamicpituitary-adrenal [HPA] axis) to structural and functional. The major structural consequences of early stress include reduced size of the midportions of the corpus callosum and attenuated development of the left neocortex, hippocampus, and amygdala. Major functional consequences include increased electrical irritability in limbic structures and reduced functional activity of the cerebellar vermis. There are also gender differences in vulnerability and functional consequences” (p. 33). In summary, the neurobiological sequelae of early stress and maltreatment may contribute in significant ways to the emergence of psychiatric disorders during development.

The complex biological mechanisms associated with individual response to stress and trauma have prompted some researchers to suggest that an evolutionary process is involved, one that allows the brain to adapt to the early experience of severe stress and deprivation. Teicher et al. (2003), for example, state that “early stress signals the nascent brain to develop along an alternative pathway adapting itself to survive and reproduce in a malevolent stress-filled world” (p. 34). This adaptive, alternative process may involve specific areas of brain development, especially individual myelinated regions such as the corpus callosum, which are “potentially susceptible to the impacts of early exposure to high levels of stress hormones” (Teicher et al., 2003, p. 35). There are suggestions that exposure to corticosteroids is a crucial factor in early stages of

infancy and childhood, organizing the brain to develop in this manner. Other studies (Seckl, 1998, cited in Teicher et al., 2003) have indicated that glucocorticoids may exert an organizing effect on development, contributing to low birthweights in infants as well as substantially increasing risks for the development of cardiovascular disease and type 2 diabetes during adulthood.

In addition to these studies of interactions among maltreatment, the neurobiology of stress, and caregiving experiences, some researchers have focused on selected genetic components that may influence the effects of maltreatment experiences. Caspi et al. (2002) conducted a longitudinal study that provides significant evidence of “inherited vulnerability,” involving interactions between a specific candidate gene (Monoamine Oxidase A or MAOA) and the occurrence of child maltreatment that contribute to enduring patterns of antisocial behavior. Jonson-Reid et al. (2010) observed that more recent work has confirmed these interactions (Taylor and Kim-Cohen, 2007), suggesting that “the effect of environmental adversity may be conditional on an individual’s genotype” (McCrory et al., 2010, p. 1079). Yet, additional studies have suggested that the presence or absence of numerous modifying factors such as gender, ethnicity, and the severity of adversity of life events can influence the magnitude of effect of inherited vulnerability (whether incurred by MAOA or other parameters of genetic risk) (Hicks et al., 2009; Weder et al., 2009; Widom and Brzustowicz, 2006). Other studies have begun to identify some of the complex ecological interactions among genotype, child behavior problems, parenting, and family stress, suggesting, for example, that genotype may be more likely to predict development of an escalating cycle of harsh parenting and child behavior problems only under certain conditions, such as when families are under stress (Riggins-Casper et al., 2003).

Studies of the effects of maltreatment on stress responses and other physiological reactions are not limited to young children. A prospective study by Shenk et al. (2010), for example, explored how a prior history of childhood sexual abuse might predict certain physiological responses to stress in late adolescence that also then predicted both higher levels of depression and antisocial behaviors in young adulthood. Their study was an empirical test of an interactive model developed by Bauer and others (2002) that is described as an “asymmetric response to stress”: a response to stress is observed in one regulatory system (e.g., the HPA axis) and an understimulated or blunted response is observed in another (e.g., the central nervous system). The study by Shenk et al. (2010) suggests

that “childhood sexual abuse may sensitize females to respond to moderate daily stressors in a manner that places them at higher risk for experiencing depressive symptoms and antisocial behaviors over time” (p. 752).

Yet, researchers who are focused on the biological pathways between child maltreatment and health outcomes consistently lament the difficulties of working with populations whose maltreatment histories are highly varied. As noted by McCrory et al. (2010, p. 1088):

There is an increasing recognition of the need to improve the construct validity of measures that assess maltreatment type (Herrenkohl and Herrenkohl, 2009) as well as improve our accuracy in gauging maltreatment severity (Litrownik et al., 2005). Future studies need to meet the challenge of becoming more systematic in delineating maltreatment type, chronicity, frequency, and even perpetrator identity, if findings across studies are to be meaningfully compared. There are some notable exceptions where researchers are already working to address these challenges (e.g., Andersen et al., 2008; Cicchetti and Rogosch, 2001).

This challenge is complicated by the observation that some of the more salient dimensions of maltreatment may be subjective (including whether events are experienced as traumatic). The objective dimensions of maltreatment, as reflected in measures of maltreatment behavior or events, may not be adequate to represent such experiences.

Although maltreatment affects children of all ages, infants and young children (under age 5) experience higher rates of abuse and neglect, according to an analysis of NCANDS data (DeVooght et al., 2011). Young children are at particular risk of adverse consequences, including death—87 percent of all child maltreatment fatalities in FY 2009 involved children in this age group, and infants less than a year old comprise 46 percent of all child maltreatment fatalities for this same period (DeVooght et al., 2011). The extreme dependency on caregivers during critical periods of development creates particular vulnerabilities for this age group. A CDC analysis of NCANDS data also demonstrated a concentration of nonfatal maltreatment and neglect at age <1 week, including high risks of homicide (CDC, 2008).

Medical personnel were more frequently cited as the sources reporting child maltreatment among infants and children under age 5 in other NCANDS analyses, whereas older children were more likely to be re-

ported by social services or educational sources (Palusci, 2011). Other characteristics for infants and younger children were different as well: They were more likely to have drug exposure and other medical problems, and their families were more likely to have alcohol and drug problems and as well as other medical problems and violence between caregivers (Palusci, 2011).

Other studies have not supported this finding, however. NCANDS data are not nationally representative, and datasets for many states are incomplete. The NIS-4 study, for example, reported, “In most cases, the 0-to 2-year-olds had significantly lower maltreatment rates than older children,” especially in the category of physical abuse (Sedlak et al., 2010b, p. 8). The authors noted, however, that “It is possible that the lower rates at these younger ages reflect some undercoverage of these age groups.”

PREVENTION EFFORTS

The 1993 NRC report Understanding Child Abuse and Neglect observed:

In the field of child maltreatment, the goals of preventive interventions are to reduce risk factors associated with child abuse and neglect, to improve the outcomes of individuals or families exposed to such risk factors, and to enhance compensatory or protective factors that could mitigate or buffer the child from the effects of victimization. (p. 161)

Since the publication of the NRC report, new dimensions have evolved in the field of prevention studies. These efforts place greater emphasis on a strengths-based approach and the promotion of positive assets that can reduce not only the risks of child maltreatment, but also influence the risks of other social and health problems, such as substance abuse, violence within families, and mental health disorders (particularly depression). The promise of prevention, yet to be realized, is to create opportunities and resources that can strengthen parenting practices and family resiliency in multiple settings and prepare families to deal with stressful conditions. Prevention efforts also point to a hope that such investments will generate significant savings in treatment costs, particularly those associated with CPS investigations and foster care placement.

In their introduction to the special 2009 issue of The Future of Children that focused on the prevention of child abuse and neglect, Paxson and Haskins (2009, p. 11) state:

Most researchers and CPS workers believe that prevention holds the key to reducing child maltreatment in the United States and to bringing down its well-documented long-term costs, both human and financial.

Prevention services are frequently categorized by the characteristics of the populations served. At present, research is focusing on the array of factors associated with vulnerable populations of families who are most susceptible to maltreatment (i.e., children who are prior victims of abuse or neglect; children of parents with mental health or behavioral disorders, especially depression or substance abuse; first-time unmarried mothers; low-birthweight infants). Preventive interventions focus on enhancing protective factors for these families through services such as parenting education and support, therapeutic or treatment services for the parents that include a parenting component, or home visiting services that offer an array of health programs as well as referrals to community resources.

Universal prevention services (also known as primary prevention) are offered to a general population through community-based efforts that attempt to reach all families, regardless of their risk status. Such efforts may include media and social marketing campaigns (e.g., the “Don’t Shake the Baby” efforts launched by pediatricians) or community-based programs housed in settings such as child care centers, hospitals (especially programs directed toward expectant or new parents), pediatric offices, or public health agencies.

Selective prevention interventions (also known as targeted or secondary prevention) focus on populations that exhibit certain risk characteristics generally associated with child maltreatment, such as poverty, first-time pregnancies, or unmarried mothers. Other selective interventions strive to add a prevention element to interventions focused on specific risk characteristics that frequently contribute to child abuse and neglect, such as parents with substance use disorders, mental health problems (usually depression), domestic violence, or child conduct disorders (Barth, 2009). Such efforts may include home visiting services that are offered to parents (usually mothers) who match one or more of these risk factors, parenting programs offered in treatment facilities that are focused on substance abuse or domestic violence, or components that are

embedded in child care centers for children whose families fit one or more of these risk factors.

Indicated prevention interventions are also known as tertiary prevention services. They are designed to prevent the recurrence of abuse and neglect or to mitigate adverse consequences among children who have already experienced maltreatment. These services are often provided to families or children who are in contact with CPS or child welfare agencies.

Research on the status of evidence-based programs in each of these three areas is presented below.

Universal Prevention

Community-Based Prevention Programs

The 2009 special issue of The Future of Children included a review of community-based programs prepared by Daro and Dodge (2009). Their article explored how “attention is shifting toward creating environments that facilitate a parent’s ability to do the right thing” (p. 67), drawing on research findings that offer insights into the ways in which neighborhoods influence child development and support parenting.

Daro and Dodge (2009) highlight five major initiatives that strive to reduce the incidence of child maltreatment, noting that these five examples are “representative of efforts underway in many states to reduce maltreatment risk or enhance child development” (p. 68). The five examples are:

1.  Triple P-Positive Parenting Program (Sanders, 1999; Sanders et al., 2003)

2.  Strengthening Families (Langford, 2007)

3.  Durham Family Initiative (Dodge et al., 2009)

4.  Strong Communities (Melton and Holaday, 2008)

5.  Community Partnerships for Protecting Children (CSSP, 1996, 1997, 2001)

The authors review the evidence base and theoretical framework that supports each initiative, highlight common and unique elements, and explore the challenges associated with the evaluation of such widescale community-based efforts that may produce evolutionary change within a general population over an extended period of time. On this latter point, the authors observe that “traditional evaluation methods that use random assignment to treatment and control conditions and assume a ’fixed’ intervention that adheres to a standardized protocol over time are of limited

utility in determining an initiative’s efficacy or in producing useful implementation lessons” (Daro and Dodge, 2009, p. 74). Though attractive in theory, few of these models, apart from the Triple P program, have demonstrated positive outcomes on child maltreatment rates in welldesigned evaluation studies. It is useful to note that the Triple P model focuses on changing parenting practices within specific contexts, rather than attempting to shift social norms around parenting at the scale of an entire neighborhood or community. One additional observation may be informative in examining the research findings associated with these five initiatives. Daro and Dodge note that, historically, prevention efforts sought to focus attention on particular populations at risk of maltreatment and provide them with “knowledge, skill-building opportunities, and assistance to overcome their personal limitations” (p. 68).

This approach, thought to be a more efficient use of public resources, has had only limited success. More recently, some communities have changed their approaches by developing prevention strategies that focus on building family strengths within an entire community rather than targeting at-risk populations. First, some communities strive “to expand public services and resources available in a community by instituting new services, streamlining service delivery processes, or fostering greater collaboration among local service providers” (Daro and Dodge, 2009, p. 68). These efforts are designed to offer coordinated and integrated services that can help families when they experience stressful circumstances.

“Other strategies focus on altering the social norms that govern personal interactions among neighbors, parent-child relationships, and personal and collective responsibility for child protection” (Daro and Dodge, 2009, p. 68). This approach builds on the theoretical frameworks associated with interactions among social capital, social organization, and rates of child maltreatment in selected neighborhoods. “In each case, the goal is to build communities with a rich array of formal and informal resources and a normative cultural context that is capable of fostering positive child and youth development” (p. 68).

Abusive Head Trauma Prevention Efforts

In addition to the community-based prevention programs reviewed by Daro and Dodge, other primary, universal preventive efforts have been initiated in healthcare settings, including prenatal classes, hospitalbased maternity wards, and pediatric offices. These efforts are focused on educating parents of newborns and infants about the risks of abusive head trauma (AHT), also known as shaken baby syndrome.

The goals of interventions are to describe typical infant behaviors and developmental processes for new parents, to highlight important areas of infant safety, and to explain the specific risks associated with head trauma and shaking an infant. The parent education materials may consist of one or more of the following items: a brief leaflet (e.g., Prevent Shaken Baby Syndrome prepared by the AAP); longer booklets (e.g., The Period of Purple Crying material developed by the National Center on Shaken Infant Syndrome); wall posters displayed in health care settings (e.g., Never, Never, Never, Never Shake an Infant prepared by SBS Prevention Plus, Groveport, Ohio); videotapes (e.g., Portrait of Promise: Preventing Shaken Baby Syndrome produced by Midwest Children’s Resource Center, St. Paul, Minnesota); and DVDs (e.g., excerpts from the child safety video produced by I Am Your Child Foundation) (Barr et al., 2009; Dias et al., 2005). The educational materials may be distributed in the healthcare setting by nurses in delivery wards, by office staff in other care settings, or by research assistants in study demonstration sites. In one study, both parents were also asked to indicate their receipt and understanding of the educational materials by signing a voluntary commitment statement (Dias et al., 2005).

In separate studies of The Period of Purple Crying Intervention, parents were given a time diary (The Baby’s Day Diary) to record infant states (e.g., awake, alert, fussing, crying, unsoothable crying, sleeping, feeding) as well as caregiver behaviors (e.g., carrying, holding, walking away) to assess the impact of the intervention on parental and other caregiver behaviors when the infant was cared for at home (Barr et al., 2009).

Evaluations of the impact of the parental education materials are limited at this time and are focused primarily on assessment of parent knowledge and behaviors as a result of the intervention (Barr et al., 2009). A regional study by Dias et al. (2005) examined outcomes associated with abusive head trauma in western New York following an extensive intervention that included multiple parent educational materials (as well as the commitment statement). During the 5.5-year study period (which began in 1998), the incidence of abusive head injuries decreased by 47 percent, from 41.5 cases per 100,000 live births (during the 6-year control period that preceded the study) to 22.2 cases; an average of 3.8 cases per 100,000 live births per year during the study period, compared with 8.2 during the prior control period (Dias et al., 2005).

Selective Prevention

In some cases, universal prevention programs with proven effectiveness have been adapted to serve families that exhibit particular risk factors for abuse and neglect. One example of this approach is the five levels of the Triple P system developed by Sanders (1999), Triple P was initially designed as a treatment intervention for children with oppositional defiant behavior problems, and was subsequently expanded into a five-level program that includes both primary prevention and maltreatment areas. In this model, it offers a comprehensive population-level system of parenting and family support, with interventions “of increasing intensity and narrowing population reach” (Prinz, 2009, p. 58). While level 1 consists largely of a media and communication strategy to reach all parents in a population, subsets of families receive more intensive and targeted sessions designed to enhance parental skills, prevent dysfunctional parenting practices, and promote teamwork between partners, thereby reducing those family risk factors commonly associated with child maltreatment.

The impact of the entire Triple P system was tested and evaluated through a population-wide sample that randomized 18 counties in a southern state, a study known as the U.S. Triple P System Population Trial (Prinz and Sanders, 2006, 2007). Three main indicators were measured to assess the impact of the intervention on child maltreatment-related events: (1) child out-of-home placements, (2) child injuries related to maltreatment (hospitalizations and emergency room visits), and (3) child maltreatment cases. As reported by Prinz et al. (2009), “significant prevention effects with large effect sizes” in all three population indicators were observed in the countries that received the Triple P intervention, following 2 continuous years of implementation.

Home visitation models are one of the best-known prevention strategies and have acquired extensive reviews in the research literature (Astuto and Allen, 2009; Daro, 2006; Isaacs, 2007; Olds et al., 1997, 1999; Paulsell et al., 2010). Originally developed as an intervention to foster healthy pregnancies and birth outcomes, early studies of the longterm outcomes of different home visitation programs in Hawaii and New York prompted the U.S. Advisory Board on Child Abuse and Neglect (1990) to recommend a universal system of home visitation for newborns and their parents, especially for first-time, low-income mothers. In response to the 2010 Affordable Care Act, the Maternal and Child Health Bureau, in collaboration with HRSA and ACF within the U.S. Department of Health and Human Services (HHS), created the Maternal, Infant,

and Early Childhood Home Visiting (MIECHV) Program. Under the MIECHV Program, HRSA awarded $91 million in FY 2010 grants to each of the 56 state and territorial entities. The grant recipients are now in the midst of developing state plans and evaluation strategies to determine their effectiveness in meeting the needs of high-risk families.

The MIECHV Program also contracted with Mathematica Policy Research (project known as “HomVEE”) to determine evidence of effectiveness of selected home visiting models in promoting child well-being (Paulsell et al., 2010). After identifying an initial list of 300 home visiting program models from the research literature, the Mathematica team ranked models by the quality of rigorous research evaluation studies on their effectiveness, implementation, and impact. This process yield 22 “prioritized” models of home visiting services for use in the state grants. HomVEE subsequently reviewed 174 impact studies and 179 implementation studies for these 22 models (Paulsell et al., 2010, p. 5). This analysis yielded nine home visiting models that meet the HHS criteria for an evidence-based, early childhood home visiting service delivery model: (1) Child FIRST, (2) Early Head Start-Home Visiting, (3) Early Intervention Program for Adolescent Mothers, (4) Family Check-Up, (5) Healthy Families America, (6) Healthy Steps, (7) Home Instruction for Parents of Preschool Youngsters, (8) Nurse Family Partnership (NFP), and (9) Parents as Teachers.⁵ “All of them have at least one high-or moderatequality study with at least two favorable, statistically significant impacts in two different domains or two or more high-or moderate-quality studies using non-overlapping analytic study samples with one or more statistically significant, favorable impacts in the same domain” (Paulsell et al., 2011, p. 8).

The Mathematica HomVEE study also examined the extent to which each of the selected home visiting models measured primary outcomes in each of eight primary domains of interest to the Maternal and Child Health Bureau (MCHB) initiative. While most of the models had favorable impacts on primary measures of child development, school readiness, and positive parenting practices, only five models specifically measured the impact of the home visitation program on child maltreatment. Within this group, only three models (Child First, Healthy Families America [HFA], and NFP), demonstrated positive impacts on child maltreatment as a primary outcome measure, with NFP showing the highest number of

______________________

⁵Key references describing each of these models, along with their study ratings, are available on the HOMVEE/ACF website: http://www.homvee.acf.hhs.gov/studyratings.aspx?rid=1&sid=10&mid=6&oid=1#High.

favorable impacts in this domain. NFP also had the greatest breadth of favorable primary findings, with favorable impacts on primary measures in six outcome domains. But measured reductions in downstream child maltreatment rates have sometimes been present in some variations and evaluations of the model, but absent in the majority of evaluations (HFA), or found in early studies but with uncertain replication in later studies (NFP). Other outcomes, apart from maltreatment prevention, have been found more consistently.

Efforts have now begun in selected regions to add services within the prioritized home visiting (HV) models that can address specific risk factors for abuse and neglect, such as maternal depression. One example of such interventions is a program that offers in-home cognitive behavioral therapy (CBT) for depressed mothers who are already receiving home visitation services (Ammerman et al., 2011). While this study does not measure child maltreatment reports as a primary outcome, the authors do report significant reductions in maternal depression, a primary risk factor for maltreatment. “There was a significantly greater reduction in depressive symptoms in the in home-cognitive behavioral therapy group relative to their counterparts who did not receive the treatment. Results from pre-and postcomparisons showed that treated mothers had decreased diagnosis of major depression, lower reported stress, increased coping and social support, and increased positive views of motherhood at posttreatment” (Ammerman et al., 2011, p. 1333).

One study has raised concern about the extent to which home visitation services are able to prevent the recidivism of physical abuse or neglect for families where maltreatment had already occurred (MacMillan et al., 2005). Other authors have noted that “to date, evaluations of whether home visitation services can alter the future life-course development in infants or children who have been maltreated have yet to be conducted” (Cicchetti and Toth, 2005).

Indicated Prevention

It is difficult to establish a clear boundary between preventive interventions and treatment services for families already in contact with child protective or child welfare agencies. In many cases, the interventions discussed in the Treatment Services section of this paper are designed to prevent the recurrence of maltreatment or to mitigate the adverse consequences of victimization. Other service delivery distinctions deserve attention; for example, families already in the child welfare system (indicated prevention) may be compelled by the courts to accept services,

whereas primary prevention services are normally voluntary. There also may be important service research implications—for example, indicated services prioritize models with large effect sizes, whereas primary prevention may emphasize reach and penetration.

Community-Based Interventions

Few efforts have been made to examine the effects of communitybased prevention interventions on child maltreatment outcomes for families that have already come to the attention of CPS. One study that has specifically examined the impact of preventive interventions on recidivism rates of child maltreatment is the Durham Family Initiative (DFI), which sought to “translate a science-based social-ecological model of how within-home child maltreatment develops, along with knowledge of public policy and practice, into a preventive system of care to reduce the population rate of child maltreatment” (Dodge et al., 2009, p. 68). DFI engaged multiple stakeholders (service providers, researchers, policy makers, and others) into a collaborative and coordinated effort that would achieve the following goals: “(1) identify families at risk for maltreatment.. . ; (2) make available evidence-based services that the community could deliver; and (3) connect families at risk with community services” (Dodge et al., 2009, p. 73). The evaluation study used a casecomparison design (not randomized) to assess the effects of the intervention in one county (Durham County, North Carolina) over time compared to five other counties that shared similar demographic characteristics with Durham County.

The maltreatment rate in Durham County declined by 49 percent as a result of the DFI, compared with a 22 percent decline for the average of the five comparison counties. The rate of decline was notable for children younger than 1 year of age (45 percent decline in Durham County compared to 12 percent average decrease in the five matched counties) (Dodge et al., 2009).

The DFI study also examined changes in recidivism rates, namely the rate at which those children who had been assessed for possible maltreatment by the Division of Social Services received a reassessment within 6 months. The results indicated that “the reassessment rate in Durham Country … decreased by 27 percent.… In contrast, the rate for the mean of the five … matched counties over the same period … decreased by 17 percent” (Dodge et al., 2009, p. 76).

Improving Developmental Outcomes for Maltreating Families

Other interventions strive not only to reduce the recurrence of maltreatment, but also to improve the life-course outcomes of young children who have already experienced maltreatment. A study by Cicchetti et al. (2006) of 1-year-old infants and their mothers examined the relative effects of selected interventions for 137 families reported for maltreatment. Their approach compared the relative strengths of different interventions in terms of their capacity “to prevent the compromised developmental attainments that accompany maltreatment and that are precursors to later maladaptation and psychopathology” (p. 624). The Cicchetti et al. (2006) study focused specifically on alternative strategies to strengthen the maternal-child relationship, to alter disruptions in attachment organization (building from the work on attachment theory by Ainsworth et al., 1978), and to promote an adaptive developmental course for infants in families that have already experienced maltreatment. The maltreating families in their study were randomly assigned to one of three treatment groups: (1) psychotherapeutic intervention, termed infant-parent psychotherapy (IPP) (dyadic parent-child therapy sessions designed to improve the parent-child attachment relationship by altering the influence of negative maternal representational models on parent-child interaction); (2) psychoeducational parenting intervention (PPI) (providing mothers with didactic training in child development, parenting skills, coping strategies for managing stresses in the immediate environment, and assistance in developing social support networks); and (3) standard community service (CS) control. A fourth group of infants from 52 nonmaltreating families provided an additional low-income normative comparison (NC) group. Assessments of the infants and mothers when the infants were approximately 26 months of age (the conclusion of the interventions) showed “dramatic changes in attachment classification” (Cicchetti and Toth, 2006, p. 643): the rate of secure attachment increased in both treatment intervention groups to 60.7 percent (from 3.1 percent preintervention) for the IPP group and 54.5 percent (from 0.0 percent preintervention) for the PPI group. The rate of secure attachment in the CS group remained extremely low (termed “virtually nonexistent” at 1.9 percent postintervention compared to 0 percent preintervention), and the rate of secure attachment in the NC group (38.6 percent compared to 32.7 percent in the post-and preintervention groups) continued to surpass the CS group. Both the IPP and PPI intervention treatment groups showed substantial gains in establishing secure attachment organization, suggesting that both types of interventions could alter the pre-

dominantly insecure attachment organizations of infants in maltreating families. The authors note that while they had initially hypothesized that “the IPP intervention would be more successful in improving attachment security than would the PPI intervention,” future studies may detect differences of these interventions during the preschool period (Cicchetti et al., 2006).

Services Provided by Child Welfare Agencies

Interest in the possible moderating impact of services provided through child welfare on later child outcomes is relatively new. Perhaps the most developed literature in this area has looked at the potential moderating effect of foster care and out-of-home placement on later outcomes with findings varying by study (Berger et al., 2009; Jonson-Reid and Barth, 2000; Ryan and Testa, 2005). Relatively little is known about the impact of lower intensity, in-home services (O’Reilly et al., 2010). This is complicated by regional variations in services and the fact that outcomes may be influenced by services that are accessed after referral by child welfare.

TREATMENT INTERVENTIONS⁶

A broad array of treatment interventions are available to serve families that are in contact with CPS, child welfare agencies, or health providers as a result of reports of abuse and neglect. Beyond the reporting, investigation, and case assessment services described in a preceding section, these interventions may include medical treatment of initial injuries or health disorders (e.g., head trauma injury in infants or growth retardation in young children), referrals to counselors or therapists for individual or group therapy, parenting education, and family support programs (which may include in-home family preservation or family reunification services). In cases that involve extreme forms of maltreatment, the child may be placed in temporary or permanent out-of-home care.

______________________

⁶As noted in the Introduction, this paper does not review the literature on the effects of different law enforcement or judicial interventions on the prevention, recurrence, or outcomes associated with child maltreatment. Such interventions may include differences in arrest policies, court practices and sentencing decisions, deterrence practices, and other interventions offered in these settings.

The scope, intensity, and duration of treatment services often depend on the initial characteristics of the reported abuse (including prior history of maltreatment), the age of the child, the complexity of family dysfunction (i.e., whether substance abuse or intimate partner violence is involved), the capacity of the health and human services agencies to provide effective treatment services, and the level of family engagement with the available services (which may be voluntary or mandatory).

A detailed description and evaluation of treatment models in each of these areas is beyond the scope of this paper. What follows is a brief description of the research base that includes selected treatment interventions, particularly those that have been shown to meet evidence-based standards in support of their effectiveness and widespread use.

Treatment of Abusive Head Trauma in Infants

In most cases, the treatment of injuries or health disorders that result from abuse and neglect follows general guidelines of the medical profession. However, particular attention has been directed toward the treatment of abusive head trauma in infants and young children, which is frequently associated with severe and sometimes fatal cerebral, spinal, and cranial injuries. In 2009, the American Academy of Pediatrics revised an earlier (2001) policy statement to clarify the terminology to be used in describing inflicted head trauma on infants and young children, often as a result of shaking and/or blunt impact (Christian et al., 2009). The term “abusive head trauma” was recommended as a replacement for “shaken baby syndrome,” which is frequently used by physicians and the public to describe the types of injuries associated with this form of maltreatment. The new AAP policy statement clarifies that “advances in the understanding of the mechanisms and clinical spectrum of injury associated with abusive head trauma compel us to modify our terminology to keep pace with our understanding of pathologic mechanisms.… The use of broad terminology that is inclusive of all mechanisms of injury, including shaking, is required” (Christian et al., 2009, p. 1409).

The 2009 AAP report notes that pediatricians should be especially alert to signs of AHT because they carry a particular burden in recognizing and responding to this condition (Christian et al., 2009). At the same time, they need to recognize the possibility of alternative causes of the head injury. Consultation with key subspecialists, child abuse experts, and social services personnel is particularly recommended:

A medical diagnosis of AHT is made only after consideration of all the clinical data. On some occasions, the diagnosis is apparent early in the course of the evaluation, because some infants and children have injuries to multiple organ systems that could only be the result of inflicted trauma. On other occasions, the diagnosis is less certain, and restraint is required until the medical evaluation has been completed. However, as physicians, we have an obligation to make a working diagnosis, as we do with many other diagnoses, and take the legally mandated steps for further investigation when indicated. Pediatricians often find it helpful to consult a subspecialist in the field of child abuse pediatrics to ensure that the medical evaluation has been complete and the diagnosis is accurate. Subspecialists in radiology, ophthalmology, neurosurgery, neurology, and other fields should also be consulted when necessary to ensure a complete and accurate evaluation. (Christian et al., 2009, p. 1410)

Special studies are currently under way to broaden public and professional awareness of the risks associated with abusive head trauma in infants and young children. These prevention efforts are discussed under the Prevention Efforts section of this appendix.

Evidence-Based Reviews of Treatment Interventions

Multiple reviews of the research literature have consistently noted the lack of well-designed treatment trials for children who have experienced maltreatment (Chaffin and Friedrich, 2004; Stevenson, 1999). Horwitz et al. (2010) note that “studies on health services use in general and mental health services use in particular have lagged behind studies documenting need” (p. 280). They observe that since the early 1990s, studies have gradually emerged that document patterns and predictors of mental health services for children in the child welfare system. For the most part, the content, quality, and outcomes of existing mental health, counseling, parenting education, and family support services have not been evaluated.

Recent efforts have emerged, however, to develop systematic reviews that classify existing treatment interventions according to the quality of the evidence base that supports them. These review efforts are based on scientific criteria (e.g., the soundness of the theory base and strength of empirical support, including the availability of controlled and randomized studies) as well as clinical feasibility and acceptance of the intervention in child welfare settings (e.g., the extent to which the inter-

vention is feasible to use in different clinical settings with families served by child welfare or child protection agencies). More recently, some rating groups have added a third criterion in the assessment of treatment interventions, namely, the extent to which they are consistent with family and child values. These three factors—best research evidence, best clinical experience, and consistency with patient values—support the definition of evidence-based practice as defined by the IOM (2001).

In 2004, two expert panels published reports that documented the extent to which evidence-based treatment interventions for child victims of abuse or neglect as well as the adult offenders were available to CPS and child welfare agencies. The first review was conducted by a group convened by the Office of Victims of Crime (OVC) in the U.S. Department of Justice (Saunders et al., 2004). This review was supplemented by a second review sponsored by the Kauffman Foundation to identify interventions that showed particular promise even if they did not yet achieve the highest standard of empirical support (Kauffman Best Practices Project, 2004). Chaffin and Friedrich (2004) subsequently summarized the OVC and Kauffman reviews within two categories of treatment interventions: (1) mental health services and (2) “other classes of services” (p. 1106) that include family preservation or reunification models as well as broad ecologically based interventions (e.g., parenting services or other in-home programs that address specific aspects of multiproblem families served by child welfare agencies).⁷

A third review effort was later initiated by an expert group convened by the state of California, which formed the California Evidence-Based Clearinghouse for Child Welfare (CEBC), an ongoing activity led by the California Department of Social Services (CDSS). CDSS selected the Chadwick Center for Children and Families-Rady Children’s Hospital-San Diego, in cooperation with the Child and Adolescent Services Research Center, to create the CEBC, which provides guidance and Webbased rating tools on evidence-based child welfare practices to statewide agencies, counties, public and private organizations, and individuals.

The treatment interventions described below are those that have received the ratings in the top two categories used by these three groups, which are generally described as “well established” or “supported.” In a few cases, additional attention is given to treatment models that do not

⁷The review by Chaffin and Friedrich (2004) also discussed prevention models, which were discussed in an earlier section of this paper.

fall within these categories, but are viewed as especially significant in the treatment of child abuse and neglect. Sorting the interventions according to their review criteria is a challenging task, however, because there is not common agreement among the three groups in rating similar treatment models.

Mental Health Interventions for Child Victims of Abuse and Neglect

Expert review groups (including OVC, and, subsequently, a collaborative effort funded by the Kauffman Foundation of St. Louis) have identified three treatment models for child victims of abuse and neglect that meet selected standards of evidence (standards include criteria such as soundness of the theory base, extent of general clinical acceptance, and strength of empirical support, among others). The review groups scored existing treatment models based on these standards and highlighted those that deserve further uptake in routine practice settings (Kauffman Best Practices Project, 2004; Saunders et al., 2004). The models that met the highest standards of evidence according to these reviews include

•   Trauma-Focused Cognitive Behavioral Therapy (TF-CBT) (Cohen and Mannarino, 1997, 2008; Cohen et al., 2006; Deblinger et al., 1996);

•  Abuse-Focused Cognitive-Behavioral Therapy (AF-CBT) (Kolko and Swenson, 2002); and

•  Parent-Child Interaction Therapy (PCIT) (Chaffin and Friedrich, 2004; Hembree-Kigin and McNeil, 1995; Urquiza and McNeil, 1996).

A later review by CEBC also scored TF-CBT and PCIT at the highest level (“1-well-established”). However, the CBEC assigned a lower score to AF-CBT (“3-promising”), stating that although a random controlled trial reported that parents receiving AF-CBT showed positive changes on most risk and outcome indicators, “the changes were similar to those found with a less clearly defined family therapy study arm delivered in the laboratory setting, and both were superior to usual services. Interpretation of study comparisons is limited by the unknown effectiveness of the family therapy comparison and by the non-random allocation to usual services” (Damashek and Chaffin, in press, p. 961).

Each model uses a well-supported, empirically validated treatment protocol that addresses specific psychological symptoms or behavioral

disorders. TF-CBT uses the principles of gradual exposure and cognitive restructuring to address the psychological impact and memory of abuse. PCIT involves a live-coached parent training program that focuses on early childhood behavioral problems.

The OVC and Kauffman review panels noted that a fourth treatment model—Multisystemic Therapy (MST) (Curtis et al., 2004)—also showed significant initial promise, and its use with adolescent sexual offenders is highlighted below.

Family and Parenting Support Services

The other class of interventions subjected to empirical tests involves family and parenting support services that focus on the recognition and change of specific behaviors that contribute to maltreatment. These interventions go beyond traditional parenting education classes and often involve in-home services that may include mental health components as well as role playing, coaching and mentoring, self-assessment questionnaires, and quasi-experimental studies to establish the empirical validity of specific components of selected programs.

The OVC and Kauffman reviews highlighted selected examples of such evidence-based service models. Later reviews by CEBC gave the highest rating (“1-well-established”) to two programs that are focused on the treatment of physical abuse (The Incredible Years and Triple P) and the second highest rating (“2-supported”) to the use of the Homebuilders treatment model for families reported for child neglect:

•  The Incredible Years is a parenting and child behavior management skill training program initially designed as a treatment for child behavior disorders as well as a school-based behavior problem prevention program for high-risk populations. The program has been adapted for parents from high-risk contexts, such as Head Start, that sometimes have high rates of future child welfare involvement (Webster-Stratton, 1998; Webster-Stratton and Reid, 2003).

•  Triple P is a program designed to promote positive parenting skills in managing child behavior problems. Parenting materials address specific developmental periods from infancy through adolescence, across five levels of intensity, with higher levels designed for progressively higher risk populations. Originally developed in Australia, Triple P has been tested as a prevention model in the United States to compare population-level child

maltreatment outcomes among 18 counties (9 randomly assigned to receive multilevel Triple P versus 9 randomized to services as usual) (Prinz et al., 2009).

•  Homebuilders was designed specifically as an intensive, shortterm, home-based program to prevent out-of-home child placement or to help reunify children in the foster care system with their biological parents. This approach is different from the first two models, which modify services designed for general child problem behaviors with adaptations appropriate for families reported for or at risk of child maltreatment (Fraser et al., 1996).

Other programs highlighted by the earlier OVC and Kauffman reviews as efforts that were acquiring greater use within the child welfare system and showed promise of empirical support for initial controlled studies included the following:

•  Project 12-Ways/Safe Care (Lutzker et al., 2001)

•  Family Connections model (Thomas et al., 2003)

•  Parent Management Training (Patterson et al., 2002)

SafeCare and Family Connections both received a “promising” (level 3) rating from CBEC as well. A recent large-scale (n = 2,175) statewide controlled trial of the SafeCare model (Chaffin et al., in press) has reported significantly reduced downstream child maltreatment outcomes among families in child welfare, which may improve the earlier ratings by the OVC and Kauffman reviews.

Adult Treatment of Child Sex Molesters

In addition to identifying TF-CBT as an evidence-based treatment intervention for victims of physical and sexual abuse, the OVC-convened expert panel gave a high rating to the adult treatment program for child sex molesters (Saunders et al., 2004). The program description indicates that the category of child sex molesters is a “heterogeneous group who may have a variety of psychological and behavioral problems or show no psychopathology beyond their sexual interest in a particular child” (p. 96). For those offenders who could be diagnosed as psychopathological, “there is currently no known effective treatment” (p. 96), and those who exhibit signs of sexual deviancy are at high risk of recidivism.

Despite these observations, the Association for the Treatment of Sexual Abusers (ATSA, 2001) has indicated that the sexual molestation of children is a treatable behavior problem. The OVC expert panel gave

their second highest rating to a model for child molester treatment that uses cognitive behavioral and adjunctive therapies. The model aims to help child sexual offenders develop the motivation and skills to stop sexual offending by replacing harmful thinking and behaviors with healthy thoughts and the skills to make choices that will reduce risk.

Cognitive behavioral approaches are currently considered the most effective methods of treatment, with pharmacological, educational, skills-building, self-help, and other methods used as adjuncts to treatment (ATSA, 2001). Alexander’s (1999) analysis of 79 sexual offender treatment studies, including Relapse Prevention and other CBT models, found that the cognitive behavioral interventions had recidivism rates of 8.1 percent, while other treatment approaches had a recidivism rate of 18.3 percent and untreated molesters showed a recidivism rate of 25.8 percent.

In a more recent report, CEBC assigned a high rating to the use of MST for adolescent sexual abusers in June 2011, awarding this model a provisional rating of “1” or “Established,” which is the highest category used by the Clearinghouse (Damashek and Chaffin, in press). CEBC notes that MST has been recommended as a model program for general delinquency by two organizations: the Blueprints for Violence Prevention Project (Mihalic et al., 2001) and the U.S. Surgeon General’s report on youth violence (USPHS, 2001). Three randomized trials have supported MST with adolescent sex offenders.

One area that bridges treatment of maltreatment victims and perpetrators is treatment of preadolescent children with aggressive sexual behavior problems, which can be both a sequelae of the child’s own abuse and a behavior that often targets other children. Short-term, parentinvolved programs have been developed and tested for these children, with good long-term results (Carpentier et al., 2006) and low rates of subsequent sexual offenses.

Social Service and Other Child Welfare Interventions

The treatment interventions described above are generally offered through contract services to state-or county-based human services agencies in responding to families reported for child maltreatment. The procedures used by these agencies to detect, report, investigate, and substantiate reports of abuse and neglect are discussed in a separate section of this appendix.

The following section of this appendix, Services and System-Level Issues, describes highlights from research in areas such as intake and

referrals, investigations and risk assessment, alternative or differential responses to reports of child abuse and neglect, and case management practices.

SERVICES AND SYSTEMS-LEVEL ISSUES

Reviews of the quality and effectiveness of the individual service and decision-making components of CPS and child welfare agencies have identified several key themes that deserve research attention in the evaluation of overall system performance. These include the use of data and evidence-based approaches in the following areas:

•  Intake and investigation

•  Risk assessment decision making

•  Referrals for services (including differential response)

•  The impact of class-action litigation on child welfare policies and practices

Rigorous studies of the evidence base are limited and empirical research is lacking in most of these areas. However, the research base is expanding, as illustrated in the following sections.

Intake and Referrals

The traditional practice of CPS agencies is to respond to reports of abuse and neglect, to identify actions or circumstances that may be harmful to the child, and to provide services and resources that can ensure child safety and well-being. The intake process is the “front end” of the CPS system: It involves the actions associated with the initial receipt of and response to a complaint of child abuse and neglect, also called a “referral.” The referrals may involve one or more children, they may involve one or more types of abuse and neglect, they may be a single event or part of a recurring pattern of maltreatment, they may or may not fall within the statutory guidelines of the CPS agency, and the complaint may involve actions that require attention by other social service, health, or law enforcement agencies. Limited knowledge is available about the provision and acceptance of these referrals, in part because the linked data systems needed to track such information are largely nonexistent (Jonson-Reid and Drake, 2008). Most of the admittedly few studies that have focused explicitly on the intake process are concerned with the screening-in/screening-out decision making that occurs at this time, no-

ting that such decisions can result in both the overinclusion of children who are not at significant risk of maltreatment as well as the underinclusion of children who require protection and support (Pecora et al., 2009; Waldfogel, 1998).

The HHS (2011) report Child Maltreatment 2010, based on NCANDS data from 45 states,⁸ reported that the participating states received a total of 2.6 million referrals in 2010. From this total, 60.7 percent of the initial complaints were screened in (ranging from 25.2 to 98.7 percent among the individual states) and 39.3 percent were screened out (ranging from 1.3 to 74.8 percent among the states) (HHS, 2011, p. 5). When applied to the national population of all 50 states, the District of Columbia, and Puerto Rico, CPS agencies received an estimated 3.3 million referrals (including 5.9 million children) during FY 2010. This is estimated to be a national average rate of 43.8 referrals per 1,000 children.

Risk Assessment and Investigation

Traditionally, both research and state services systems relied heavily on system categories of “substantiation” or “indication” as classifications of risk. Yet several studies suggest that the practice of indication or substantiation as a means of identifying risk has little or no relationship to the actual future risk of harm (Fallon et al., 2010; Hussey et al., 2005; Kohl et al., 2009). “Substantiation” of abuse and neglect in law and common language implies a specificity or distinction that frequently does not appear to exist in fact or in practice. Studies have demonstrated that cases with unsubstantiated reports appear to be equally at risk and have problems equivalent to those with substantiated reports, including being equally at risk for future substantiated maltreatment (Drake et al., 2003; Kohl et al., 2009). The risk-level comparability between substantiated and unsubstantiated cases therefore raises concern about limiting maltreatment intervention or prevention services only to those cases classified as “substantiated.”

As a result, substantial interest has emerged in developing effective risk assessment tools that can be tied to service provision. Several research studies have compared the merits of using different models of risk assessment in examining child maltreatment cases. The traditional model is a consensus-based protocol, which relies on a consensus judgment of

______________________

⁸States that did not provide referral data included Hawaii, Illinois, New Jersey, New York, North Carolina, and Pennsylvania.

experts to identify and assess specific client characteristics. This assessment provides a basis for the caseworker’s clinical judgment about the risk of future abuse and neglect (Baird et al., 1999). The second approach draws on an actuarial model, which uses data from longitudinal research and empirically validated instruments to estimate the probability of future maltreatment.

One rigorous analysis, which compared the strengths of two consensus-based models with one actuarial system, reported that “actuarialbased systems are more accurate than consensus-based or expert systems and, therefore, have the potential to improve CPS decision making” (Baird and Wagner, 2000, p. 868). The conclusion was based on differences in the rates of subsequent investigations, substantiations, and placements that were computed over an 18-month period for cases classified at low-, moderate-, and high-risk levels in each model.

In 2003 the Center for Child Welfare Policy in Columbus published the results of an extensive initiative to examine the use of risk assessment models in child welfare decision making (Rycus and Hughes, 2003). The report concluded that while most child welfare agencies had adopted some form of risk assessment to guide the resolution of case-specific and system-related practice problems, fundamental problems continued to challenge these efforts. They identified key themes that deserved attention:

•  “There is a lack of agreement regarding the proper scope and purpose of risk assessment technology in child welfare assessment and case planning activities.

•  Fundamental concepts, premises, terminology, and measures have not always been well defined or articulated, are often applied in an idiosyncratic manner, are highly inconsistent among risk models, and in some cases, are simply inaccurate. This creates ambiguity, confusion, and contradiction, and greatly increases the likelihood of error and bias in risk ratings and subsequent practice decisions.

•  There are serious methodological problems in the design and development of many risk assessment technologies and models, and also in much of the research designed to evaluate and validate them. This not only impacts the reliability and validity of the models, but results in the communication of inaccurate information about their methodological soundness to the practice field.

•  A variety of systemic, bureaucratic, and individual barriers impede the large-scale implementation of formal risk assessment technologies by child welfare agencies.

•  It is often expected that formal risk assessment activities should serve a variety of administrative, political, and systemic functions in child welfare organizations that have little to do with making accurate protective decisions for children.

•  A number of ethical and legal issues related to risk assessment have not been fully addressed.” (Rycus and Hughes, 2003, p. 7)

The 2003 report also noted that one of the major contributions of risk assessment models is to identify those families that have a high likelihood of continuing recurrence of child maltreatment. They are, in essence, “safety” assessment models. This type of risk assessment during the intake process allows the caseworkers to classify families into different categories of low, moderate, and severe risk and to determine what types of referrals or investigations are most appropriate for each set of family circumstances. Other risk assessment models are designed to provide an ongoing form of data collection and monitoring throughout the case, to prioritize the types of services that are appropriate for groups of families, and to best determine the workload for each case worker. The 2003 report observes that risk assessment has thus expanded to include not only a “point-in-time” evaluation tool, but also an overarching case management strategy that requires ongoing assessment of risk and family needs (Rycus and Hughes, 2003, p. 10)

Referrals for Services (Including Differential Response)

Once a referral is “screened in” by a CPS agency and the initial risk assessment has been completed, caseworkers decide whether the referral involves serious and immediate harm to the child, which warrants an investigatory process to substantiate the abuse or neglect and to invoke action against the perpetrator, or whether other types of risk are involved that could benefit from supportive interventions. In cases involving serious risk, CPS will conduct the investigation and prepare a report for further action by other agency or law enforcement personnel. The traditional CPS response to child maltreatment cases (sometimes referred to as a forensic response) is appropriate for those circumstances that present desperate situations or immediate threats to safety or injury to a child, such as sexual abuse, imminent harm, or abuse by caregivers in a state or county residential facility.

However, intake efforts in child protective services have identified many families that do not exhibit severe forms of maltreatment that threaten child health or safety, but still constitute endangerment and require support and services to help parents and other family members care for their children during difficult times, such as circumstances that involve substance abuse or domestic violence. In other circumstances, insufficient evidence of harm may be present, yet the family may be willing on a voluntary basis to cooperate with child protective services personnel in order to improve the well-being of their children.

As a result, multiple states have developed legislative reforms and flexible service strategies that provide opportunities to offer a “differential response” to CPS investigations of accepted reports of child abuse and neglect allegations (this approach may also be termed an alternative response, family assessment response, dual track, or multiple track). The differential response allows agencies to focus on the whole family unit as opposed to restricting their intervention to the child who was the subject of the initial complaint (Flynn et al., 2011).

The 2010 Child Maltreatment report (HHS, 2011) noted the increasing number of children served by differential or alternative response programs between 2006 and 2010. Nearly 10 percent of children (of the initial 3 million who were the subject of initial referrals) received an alternative response in 2010, nearly twice the number who received such a disposition in 2006.

The differential response approach considers numerous factors, “such as the type and severity of the maltreatment, number and sources of previous reports, and willingness of the family to participate in services” (NQIC, 2009, p. 1). In some cases, county agencies will further distinguish between a differential response for cases involving domestic violence (which require specialized referrals), and those that involve a family assessment process that identifies other areas of child safety, stability, or well-being that require attention. A report of one county agency in Minnesota, for example, noted that about 62 percent of families receiving services via CPS during the period 1999-2004 were served by the differential response, and about one-third of these families involved services for domestic violence (Sawyer and Lohrbach, 2005).

In launching a differential response approach, several states have supported demonstration and evaluation efforts to determine the feasibility, outcomes, and effectiveness this alternative to traditional investigations and services. The National Quality Improvement Center (NQIC) indicated that Illinois, Minnesota, Missouri, North Carolina, Ohio, Ten-

nessee, and Virginia have supported “pilot projects with specified time frames and demonstration sites; statewide expansion is/has been dependent upon results from the demonstrations” (NQIC, 2010, p. 4). A recent three-state study of the use of the differential response system (DRS) in Minnesota, Nevada, and North Carolina reported that

the core of [a] Differential Response System is a framework based on the values of family strengths, collaboration, respect, and community connections. Within this framework, DRS provides flexibility for states to shape the model to fit their own population, culture, and government structure. This system is not a “one-size-fits-all” model for families, or for states. (Flynn et al., 2011, p. 105)

Recognizing the multiple dimensions of the differential response model, the HHS Children’s Bureau has launched a cooperative agreement with the NQIC on Differential Response in Child Protective Services to develop a 5-year project to expand the knowledge base on differential response. The project consists of two studies that include a national needs assessment (Phase 1) and the implementation of the differential response in three demonstration sites—Colorado, Illinois, and Ohio—and dissertation research (HHS, 2011).

Case Management Practices

Significant efforts and investments have included directed family meetings as a standard CPS response. These include models such as Family Groups Conferencing, Family 2 Family, Wrap Around, Family Team Decision-Making, and others. The time and resources associated with family meetings may be quite expensive in terms of family, social services, and community professionals.

Few rigorous studies have been conducted to compare the outcomes associated with different types of family meeting strategies. One study, requested by the Washington State Legislature, examined the effects of Family Team Decision-Making (FTDM) on child placement outcomes, using a pre/postadoption method as the study design (Miller, 2011). The authors concluded that the implementation of FTDM “had no overall significant impact on rates of placement following CPS referrals, time to permanency, or new accepted CPS referrals after an exit to a permanent placement” (p. 8). However, some differential impacts were observed on some outcomes, depending on race.

The Impact of Class-Action Litigation on Child Welfare Policies and Practices

The desire to introduce more flexibility into the responses of CPS and child welfare agencies to reports of child maltreatment has also emerged in judicial treatment of various class-action suits that strive to establish standards and procedures that would foster greater accountability of the agencies to the children and families they serve. Noonan et al. (2009) provide an overview of the historical practices that have guided judicial oversight in such cases, noting that such lawsuits have successfully challenged the child welfare system in about two thirds of the states, involving “demonstrations or concessions of massive noncompliance with federal requirements—failure to take action in response to indications of abuse and neglect; arbitrary removal of children without reasonable reunification efforts; and placement of children in inappropriate, often dangerous, settings without substantial consideration or review” (Noonan et al., 2009, p. 530).

The Noonan et al. (2009) overview highlights how courts in Alabama and Utah have sought to navigate between the historic extremes of governing bureaucratic agencies by explicit rules or rigorous standards, setting a new course of injunctive relief that supports “a collaborative process for specifying norms through analysis of cases,” as well as “a form of norm elaboration through peer review that engages all levels of the system, as well as outside experts” (p. 545). The monitoring procedure adopted by the courts as the central measure of compliance in decisions to terminate court supervision in these two examples relies on a distinctive and innovative diagnostic monitoring process called the Quality Service Review (QSR), which complements and strengthens the customizing and collaborative features of traditional social work practice. “The QSR preserves the traditional social work commitment to forms of supervision that respect the complex contextuality of frontline decisions and encourage workers to respond to clients as concrete individuals” (Noonan et al., 2009, p. 542).

The authors note that rather than imposing a uniform set of standards on decision-making procedures that involve changing circumstances and different contexts for each case, the courts sought to develop remedies that would allow caseworkers to “do the right thing” in response to the cases presented to them. They further observed:

Although it is useful to speak of the Alabama and Utah reforms in terms of a single model, the model does not have a canonical defini-

tion. Many participants see the core of the reform as the reconception of frontline case work as contextual and collaborative judgment, sometimes called in Utah and Alabama “the Practice Model” (and elsewhere, the “problem-solving model”) (Huntington, 2006, pp. 674-687). Others put greater weight on central facilitation of diagnostic monitoring, especially through the QSR. Despite these differences, there is a consensus that both elements are crucial. What we call the Alabama-Utah model is a heuristic that explains how the integration of collaborative casework with diagnostic monitoring makes it possible for administration to learn from local practice while correcting its mistakes. (Noonan et al., 2009, p. 538)

The QSR was initially developed in Alabama, and has also been applied to child welfare programs in 11 other states, including Utah.

SOCIAL POLICY

A detailed review of the key social policy issues associated with the identification, assessment, treatment, and prevention of child abuse and neglect is far beyond the scope of this paper. However, it is useful to highlight some of the major themes that have emerged since the publication of the 1993 NRC report to frame the types of questions that could inform future studies and discussion. These include

•  policy objectives in protecting children and deterring perpetrators;

•  purpose and scope of risk assessment for maltreating families;

•  developing trauma-informed treatment interventions for children and families;

•  strengthening families and neighborhoods under adverse conditions to prevent abuse and neglect;

•  interactions of abuse and neglect with other health, regulatory, and cognitive functions;

•  integration of data sources and measures—ethical and legal issues; and

•  moving research into policy and practice.

Each of these topics, and others, has generated significant discourse in research and policy settings. Brief summaries are offered below as a first step in framing these issues for future discussion.

Policy Objectives in Protecting Children and Deterring Perpetrators

Given the complexity of the nature, scope, and consequences of child abuse and neglect, public agencies and legislators must consistently balance multiple objectives that may at times conflict with each other. These include (1) safeguarding and protecting children, (2) preserving and strengthening families, (3) protecting confidential and private information, and (4) making efficient use of public resources. Unilateral efforts to advance one objective (e.g., the early efforts focused on family preservation for children victimized by abuse and neglect) often raise substantial questions about their impact on other social policy goals.

The lack of progress in developing effective policy responses to achieve these multiple goals is not unique to the United States. For example, researchers in Northern Ireland have termed child abuse as a “wicked problem” and called the current conceptualization of child abuse “flawed” (Devaney and Spratt, 2009). In considering policy developments in the United Kingdom relating to children and families experiencing multiple adversities, they argue that

in adopting a rational technical approach to the management of child abuse, there is a tendency to focus on shorter term outcomes for the child, such as immediate safety, that primarily reflect the outputs of the child protection system. However, by viewing child abuse as a wicked problem, that is complex and less amenable to being solved, then child welfare professionals can be supported to focus on achieving longer term outcomes for children that are more likely to meet their needs. The authors argue for an earlier identification of and intervention with children who are experiencing multiple adversity, such as those living with parents misusing substances and exposed to intimate partner violence. (p. 1)

At the same time, multiple policy initiatives are under way to accelerate aggressive responses that can protect children from sexual abuse and other fatal injuries. Such initiatives include various legislative initiatives that require the public identification and notice of sexual offenders as well as the prosecution of institutional officials who fail to report information about child abuse and neglect.

Purpose and Scope of Risk Assessment for
Maltreating Families

As noted in a prior section of this appendix, risk assessment initiatives have increasingly relied on standardized protocols to guide caseworker decision making. However, there continues to be ambiguity as to whether risk assessment should be viewed as a one-time effort to determine child safety needs, or whether it should be part of an ongoing examination of the needs and circumstances of families that experience difficult circumstances (which may include inadequate housing, substance abuse, and intimate partner violence) (Rycus and Hughes, 2003). The report from the Center for Child Welfare Policy captures much of the uncertainty facing administrators and caseworkers in this area:

Since large-scale change has historically been so difficult for many organizations, it may ultimately be easier to support ineffective, even potentially harmful, technologies rather than change them, both because of the financial investment already made, and because an overburdened workforce cannot sustain another large-scale change. Unfortunately, perhaps because of the many other seemingly intractable problems facing the child welfare field, we appear to have a collective vulnerability to the promises of untested and unproven risk assessment models and technology. (Rycus and Hughes, 2003, p. 30)

Development of Trauma-Informed Treatment Services for
Children and Families

Advances in research on the effects of stress and trauma on diverse biological and psychological systems have sparked interest in developing methods to incorporate these findings into treatment interventions for children who have been maltreated as well as those who have witnessed severe violence (Harris et al., 2004; Shenk et al., 2010). Efforts to build evidence-based interventions in medical and judicial settings as well as other centers that serve high-risk populations of children and their families may yield important insights into the ways in which children of varying ages may respond to the effects of different forms of trauma, including maltreatment.

Strengthening Families and Neighborhoods Under Adverse
Conditions to Prevent Abuse and Neglect

The nature of categorical funding for federal programs and the hodgepodge of national and state legislation that addresses child and family needs create significant barriers to collaborative or innovative efforts among programs that frequently serve the same families. In many cases, families reported for maltreatment experience multiple other problems, such as substance abuse, intimate partner violence, mental health disorders, poverty, inadequate housing, poor schools, violent neighborhoods, among others. While major initiatives exist throughout the federal government to address each of these problems, the efforts are splintered across multiple authorization bills, cabinet departments and agencies, and programmatic budgets. This compartmentalization affects the research endeavor as well, as one agency may launch a major initiative for prevention studies that has the potential to examine the impact of their initiative on child maltreatment, but financial resources (or authority) are lacking to add this specific topic to the primary outcomes of interest.

One example highlights this difficulty. The Obama Administration has launched the Promise Neighborhoods initiative, a nationwide effort that will help 20 high-poverty neighborhoods in implementing comprehensive preventive interventions. In a collaborative effort, the National Institute on Drug Abuse is also supporting a Promise Neighborhood Research Consortium (PNRC) (http://www.promiseneighborhoods.org/about) to provide the technical expertise, measurement, and data collection resources that will be necessary to support an extensive evaluation of the impact of the Promise Neighborhoods initiative. The consortium has multiple goals, including identifying evidence-based prevention and treatment interventions (strategies, practices, programs, and policies) that communities can adopt and implement; assisting local communities in implementing measures of well-being and of risk and protective factors in order to assess whether prevention and treatment interventions are achieving their intended benefits; and building a series of communitybased research initiatives that can examine the impact of evidence-based policies, programs, and practices when implemented in high-poverty communities.

The model developed by the PNRC identifies a series of major influences on child and youth outcomes, including family and neighborhood poverty, social isolation, and access to health care, which are equally relevant to risk factors for child maltreatment. Attention to issues related to child maltreatment and family violence are absent from the consorti-

um’s materials, although emphasis is placed on peer violence and bullying, which are relevant to educational settings.

It is certainly understandable that each major federal initiative to support vulnerable families and neighborhoods cannot address all dimensions of the problems that affect their lives. But opportunities may exist to raise the significance of the problem of child maltreatment, in terms of its impact on the health and well-being of so many children, and to support multiagency collaborative efforts as part of these comprehensive prevention initiatives.

Interactions of Abuse and Neglect with Other Health,
Regulatory, and Cognitive Functions

In a similar vein, studies of the precursors to many adult health disorders are now striving to understand the fundamental mechanisms and influences that disrupt multiple physiological systems that regulate health and well-being. These studies have focused attention on diverse influences and mechanisms related to stress and trauma, such as allostatic load (McEwen, 2000), toxic stress (Shonkoff et al., 2012), and the role of telomere length in the aging process (Drury et al., 2011). These and other studies also examine how adverse experiences (e.g., abuse and neglect) may influence certain perceptual and decision-making systems related to executive function, memory, and pattern recognition, among other areas.

The ability to image brain structures and functioning in infants, children, and youth is still in its early stages of development, and normative standards and processes are not yet reliable for the general or special populations. But this field offers an opportunity to integrate studies of abuse and neglect with other forms of stress and trauma in order to discern the relative contribution of key variables that are poorly understood. Secure attachment, for example, remains an important construct for both animal and human studies, and requires rigorous studies to examine the ways in which the timing, severity, and duration of disruptions in the parent-child relationship status may influence later health and behavioral outcomes.

In some cases, studies of early neglect and deprivation with animal models may be available that offer opportunities to demonstrate key interactions that inform our understanding of the causes and consequences of child maltreatment, including transgenerational processes that affect behavior (Champagne et al., 2003; Kaufman et al., 2000; Suomi, 1997). In other cases, longitudinal research studies of non-U.S. populations such as the Romanian orphan or Dunedin studies can yield important findings

about the interactions among social stressors, physiological disruptions, and brain function (Chugani et al., 2001). All too often, however, studies are conducted with small populations of samples drawn from clinical settings and involve families or children who are struggling with multiple adversities, only some of which may include maltreatment. These barriers challenge research scientists and the agencies that support them to identify innovative ways to combine efforts, collaborate in the development of common theory building, measure development, and build consortium efforts that can yield productive interdisciplinary approaches to major research questions.

Integration of Data Sources and Measures—Ethical and
Legal Issues

While the desire for collaborative and interdisciplinary studies provides opportunity for creative endeavors, these initiatives require attention to the particular ethical and legal issues associated with the conduct of research on child maltreatment. Efforts to identify the scope and experience with child maltreatment within the general population are especially sensitive because many individuals may be reluctant to disclose their own circumstances or may not be able to recall the timing, duration, or severity of specific events. While efforts to document adverse experiences may be appropriate as part of a clinical intervention, soliciting such information in the absence of therapeutic services raises basic issues of fairness and invasion of privacy. Establishing valid and reliable sources for self-reports requires privileged access to confidential social service and health records.

Many researchers have developed appropriate ways to protect sensitive information, including the removal of identifiers in survey samples and longitudinal studies. More challenging issues are emerging on the horizon, however, with the advent of electronic health systems and efforts to document experience with stress and adversity as part of personal health records.

Moving Research into Policy and Practice

Researchers in the field of child abuse and neglect are consistently pressed to translate their findings into clinical applications as well as recommendations for policy and practice. At present, the infrastructure to support the dissemination and translation of basic research findings into practice and policy is limited. Notable exceptions include the new na-

tional home visiting initiative recently launched by HRSA and the child fatality review teams that are now common in many states and localities.

Several research centers are striving to promote interdisciplinary exchanges among diverse research investigators and scholars through the use of consortium meetings and special training programs. Examples include efforts by the Translational Research on Child Neglect Consortium (http://www.trcnconsortium.com/index.htm) and the National Data Archive on Child Abuse and Neglect (http://www.ndacan.cornell.edu). Numerous clearinghouse efforts and national centers have been created to address specific topics in the field of child maltreatment studies and to distribute knowledge through online websites, publications, training efforts, and technical assistance.

What is lacking, however, is the creation of dedicated and ongoing efforts that could accomplish multiple goals: (1) identify key priority areas of policy and programmatic needs in the field of child abuse and neglect, (2) synthesize major research findings into white papers and other working materials that address those priority areas, (3) offer opportunities for practitioners and researchers to discuss and critique the strengths and limitations of the current research base in meeting priority needs, and (4) develop research initiatives to address those shortcomings in research studies. A striking example of an effort to address the first three objectives is the 2003 white paper on issues of risk assessment in child protective services, produced by the North American Resource Center for Child Welfare (Rycus and Hughes, 2003).

Key systems-level factors to consider in improving the implementation and study of evidence-based practices include workforce education and training issues, as well as the need for cost-effective and collaborative mechanisms that can strengthen and extend existing informationsharing and technical assistance efforts (i.e., clearinghouse approaches).

CONCLUSION

Since the publication of the 1993 NRC report Understanding Child Abuse and Neglect, significant advances have occurred in multiple areas of research interest in this field. These advances are yielding new insights into the types of methods, measures, and data sources that are likely to provide significant improvements in the prevention and treatment of child maltreatment in the coming years. At the same time, research advances are highlighting the tremendous complexity of the problem of

child maltreatment. As the knowledge base grows, multiple theoretical frameworks and approaches have emerged that each provide some significant insights. Yet an integrated and robust explanation of the causes and consequences of abuse and neglect remains elusive.

One major and persistent challenge within the field is the need to identify those types of abuse and neglect that present immediate threats to the child and thus require a swift and legalistic approach to ensure the protection and safety of the child. At the same time, many social services agencies recognize that the bulk of the cases referred to their attention require more family-oriented approaches in which fostering child health and well-being may be best achieved by meeting the multiple needs of several family members. Many agencies believe they lack the resources to provide effective and sustainable remedies that can foster family stability and support parents in their efforts to care for their children, especially during times of hardship and difficult circumstances.

Insights into the long-term adverse effects of even minor but chronic forms of abuse and neglect point to the need for collaborative efforts that can raise the visibility of this topic in areas such as public health, early learning, neighborhood development, and crime prevention. Future syntheses and critiques of research advances in selected fields such as those sketched briefly in this paper can inform these efforts and provide the basis for evidence-based practices. Such critiques can also help to shape future partnerships as well as guidelines for policy and practice in social services, health care systems, population health efforts, and law enforcement.

REFERENCES

AHRQ (Agency for Healthcare Research and Quality). 2011. HCUP databases. Healthcare cost and utilization project (HCUP). http://www.hcupus. ahrq.gov/kidoverview.jsp (accessed January 19, 2012).

Ainsworth, M. D. S., M. C. Blehar, E. Waters, and S. Wall. 1978. Patterns of attachment: A psychological study of the strange situation. Hillsdale, NJ: Lawrence Erlbaum Associates.

AlEissa, M. A., J. D. Fluke, B. Gerbaka, L. Goldbeck, J. Gray, N. Hunter, B. Madrid, B. V. Puyenbroeck, I. Richards, and L. Tonmyr. 2009. A commentary on national child maltreatment surveillance systems: Examples of progress. Child Abuse & Neglect 33(11):809-814.

Alexander, M. 1999. Sexual offender treatment efficacy revisited. Sexual Abuse: A Journal of Research and Treatment 11(2):101-116.

Ammerman, R., F. Putnam, J. Stevens, N. Bosse, J. Short, A. Bodley, and J. Van Ginkel. 2011. An open trial of in-home CBT for depressed mothers in home visitation. Maternal and Child Health Journal 15(8):1333-1341.

Anda, R. F., A. Butchart, V. J. Felitti, and D. W. Brown. 2010. Building a framework for global surveillance of the public health implications of adverse childhood experiences. American Journal of Preventive Medicine 39(1):93-98.

Andersen, S. L., A. Tomada, E. S. Vincow, E. Valente, A. Polcari, and M. H. Teicher. 2008. Preliminary evidence for sensitive periods in the effect of childhood sexual abuse on regional brain development. Journal of Neuropsychiatry and Clinical Neurosciences 20(3):292-301.

Astuto, J., and L. Allen. 2009. Home visitation and young children: An approach worth investing in? Social Policy Report 23(4):3-21.

ATSA (Association for the Treatment of Sexual Abusers). 2001. Practice standards and guidelines for members of the Association for the Treatment of Sexual Abusers. Beaverton, OR: ATSA.

Baird, C., and D. Wagner. 2000. The relative validity of actuarial-and consensus-based risk assessment systems. Children and Youth Services Review 22(11-12):839-871.

Baird, C., D. Wagner, T. Healy, and K. Johnson. 1999. Risk assessment in child protective services: Consensus and actuarial model reliability. Child Welfare 78(6):723-748.

Barr, R. G., F. P. Rivara, M. Barr, P. Cummings, J. Taylor, L. J. Lengua, and E. Meredith-Benitz. 2009. Effectiveness of educational materials designed to change knowledge and behaviors regarding crying and shaken-baby syndrome in mothers of newborns: A randomized, controlled trial. Pediatrics 123(3):972-980.

Barth, R. P. 2009. Preventing child abuse and neglect with parent training: Evidence and opportunities. The Future of Children 19(2):95-118.

Bauer, A. M., J. A. Quas, and W. T. Boyce. 2002. Associations between physiological reactivity and children’s behavior: Advantages of a multisystem approach. Developmental and Behavioral Pediatrics 23(2):102-113.

Belsky, J. 1980. Child maltreatment: An ecological integration. American Psychologist 35:320-335.

Belsky, J., and J. Vondra. 1989. Lessons from child abuse: The determinants of parentings. In Child maltreatment: Theory and research on the causes and consequences of child abuse and neglect, edited by D. Cicchetti and V. Carlson. New York: Cambridge University Press. Pp. 153-202.

Berger, L. M., S. K. Bruch, E. I. Johnson, S. James, and D. Rubin. 2009. Estimating the “impact” of out-of-home placement on child well-being: Approaching the problem of selection bias. Child Development 80(6):1856-1876.

Blumstein, A., and J. Wallman. 2006. The crime drop in America, 2nd ed. Cambridge, UK: Cambridge University Press.

Bos, K. J., N. Fox, C. H. Zeanah, and C. A. Nelson. 2009. Effects of early psychosocial deprivation on the development of memory and executive function. Frontiers in Behavioral Neuroscience 3(16):1-7.

Bronfenbrenner, U. 1979. Six theories of child development: Revised formulations and current issues. In Annals of child development, vol. 6. Greenwich, CT: JAI Press. Pp. 185-246.

Bronfenbrenner, U., and P. Morris. 2006. The bioecological model of human development. In Handbook of child psychology: Theoretical models of human development, vol. 1, edited by W. Damon and R. M. Lerner. Hoboken, NJ: John Wiley & Sons, Inc.

Carpentier, M. Y., J. F. Silovsky, and M. Chaffin. 2006. Randomized trial of treatment for children with sexual behavior problems: Ten-year follow-up. Journal of Consulting and Clinical Psychology 74(3):482-488.

Caspi, A., J. McClay, T. E. Moffitt, J. Mill, J. Martin, I. W. Craig, A. Taylor, and R. Poulton. 2002. Role of genotype in the cycle of violence in maltreated children. Science 297(5582):851-854.

CDC (Centers for Disease Control and Prevention). 2008. Nonfatal maltreatment of infants—United States. October 2005-September 2006. Morbidity and Mortality Weekly Report 57(13):336-339.

CDC and NCHS (Centers for Disease Control and Prevention and National Center for Health Statistics). 2009. International classification of diseases, 9th revision (ICD-9). http://www.cdc.gov/nchs/icd/icd9.htm (accessed February 2, 2012).

Chaffin, M., and B. Friedrich. 2004. Evidence-based treatments in child abuse and neglect. Children and Youth Services Review 26(11):1097-1113.

Champagne, F. A., D. D. Francis, A. Mar, and M. J. Meaney. 2003. Variations in maternal care in the rat as a mediating influence for the effects of environment on development. Physiology & Behavior 79(3):359-371.

Christian, C. W., and D. F. Schwarz. 2011. Child maltreatment and the transition to adult-based medical and mental health care. Pediatrics 127(1):139-145.

Christian, C. W., R. Block, and Committee on Child Abuse Neglect. 2009. Abusive head trauma in infants and children. Pediatrics 123(5):1409-1411.

Chugani, H. T., M. E. Behen, O. Muzik, C. Juhász, F. Nagy, and D. C. Chugani. 2001. Local brain functional activity following early deprivation: A study of postinstitutionalized Romanian orphans. NeuroImage 14(6):1290-1301.

Cicchetti, D., and R. Rizley. 1981. Developmental perspectives on the etiology, intergenerational transmission, and sequelae of child maltreatment. New Directions for Child Development 1981(11):31-55.

Cicchetti, D., and F. Rogosch. 2001. Diverse patterns of neuroendocrine activity in maltreated children. Development and Psychopathology 13(3):677-693.

Cicchetti, D., and S. Toth. 1995. Developmental psychopathology and disorders of affect. In Developmental psychopathology: Risk, disorder, and

adaptation. Vol. 2, edited by D. Cicchetti and D. Cohen. New York: John Wiley & Sons, Inc. Pp. 369-420.

Cicchetti, D., and S. L. Toth. 2005. Child maltreatment. Annual Review of Clinical Psychology 1:409-438.

Cicchetti, D., and K. Valentino. 2006. An ecological-transactional perspective on child maltreatment: Failure of the average expectable environment and its influence on child development. In Developmental psychopathology: Theory and method. Vol. 3, edited by D. Cicchetti and D. Cohen. New York: John Wiley & Sons, Inc. Pp. 129-201.

Cicchetti, D., F. Rogosch, and S. Toth. 2006. Fostering secure attachment in infants in maltreating families through preventive interventions. Development and Psychopathology 18(3):623-649.

Cohen, J. A., and A. P. Mannarino. 1997. A treatment study for sexually abused preschool children: Outcome during a one-year follow-up. Journal of the American Academy of Child & Adolescent Psychiatry 36(9):1228-1235.

Cohen, J., and A. P. Mannarino. 2008. Disseminating and implementing traumafocused CBT in community settings. Trauma, Violence & Abuse 9(4):214-226.

Cohen, J. A., A. P. Mannarino, and E. Deblinger. 2006. Treating trauma and traumatic grief in children and adolescents. New York: Guilford Press.

Colvert, E., M. Rutter, J. Kreppner, C. Beckett, J. Castle, C. Groothues, A. Hawkins, S. Stevens, and E. Sonuga-Barke. 2008. Do theory of mind and executive function deficits underlie the adverse outcomes associated with profound early deprivation? Findings from the English and Romanian adoptees study. Journal of Abnormal Child Psychology 36(7):1057-1068.

Coulton, C. J., and J. E. Korbin. 1995. Community level factors and child maltreatment rates. Child Development 66(5):1262-1276.

Coulton, C. J., D. S. Crampton, M. Irwin, J. C. Spilsbury, and J. E. Korbin. 2007. How neighborhoods influence child maltreatment: A review of the literature and alternative pathways. Child Abuse & Neglect 31(11-12):1117-1142.

Crume, T. L., C. DiGuiseppi, T. Byers, A. P. Sirotnak, and C. J. Garrett. 2002. Underascertainment of child maltreatment fatalities by death certificates, 1990-1998. Pediatrics 110(2):e18.

CSSP (Center for the Study of Social Policy). 1996. Community partnerships for protecting children. Washington, DC: CSSP.

CSSP. 1997. Strategies to keep children safe: Why community partnerships will make a difference. Washington, DC: CSSP.

CSSP. 2001. Building capacity for local decision-making: Executive summary. Washington, DC: CSSP.

Curtis, N., K. Ronan, and C. M. Borduin. 2004. Multisystemic treatment: A meta-analysis of outcome studies. Journal of Family Psychology 18(3):411-419.

Damashek, A. L., and M. J. Chaffin. In press. Child abuse and neglect. In Handbook of evidence based practice in clinical psychology, edited by P. Strumey and M. Hersen. New York: John Wiley & Sons, Inc. Pp. 947-985.

Daro, D. 2006. Home visitation: Assessing progress, managing expectations. Chicago, IL: Ounce of Prevention Fund and Chaplin Hall Center for Children.

Daro, D., and K. Dodge. 2009. Creating community responsibility for child protection: Possibilities and challenges. The Future of Children 19(2):67-93.

De Bellis, M. D. 2001. Developmental traumatology: The psychobiological development of maltreated children and its implications for research, treatment, and policy. Development and Psychopathology 13(3):539-564.

Deblinger, E., J. Lippmann, and R. Steer. 1996. Sexually abused children suffering posttraumatic stress symptoms: Initial treatment outcome findings. Child Maltreatment 1(4):310-321.

Devaney, J., and T. Spratt. 2009. Child abuse as a complex and wicked problem: Reflecting on policy developments in the United Kingdom in working with children and families with multiple problems. Children and Youth Services Review 31(6):635-641.

DeVooght, K., M. McCoy-Roth, and M. Freundlich. 2011. Young and vulnerable: Children five and under experience high maltreatment rates. Child Trends: Early Childhood Highlights 2(2):1-20.

Dias, M. S., K. Smith, K. deGuehery, P. Mazur, V. Li, and M. L. Shaffer. 2005. Preventing abusive head trauma among infants and young children: A hospital-based, parent education program. Pediatrics 115(4):e470-e477.

DiScala, C., R. Sege, G. H. Li, and R. M. Reece. 2000. Child abuse and unintentional injuries—a 10-year retrospective. Archives of Pediatrics & Adolescent Medicine 154(1):16-22.

Dodge, K., R. Murphy, K. O’Donnell, and C. Christopoulos. 2009. Communitylevel prevention of child maltreatment: The Durham family initiative. In Preventing child maltreatment: Community approaches, edited by K. Dodge and D. Coleman. New York: Guilford Press. Pp. 68-81.

Drake, B., and S. Pandey. 1996. Understanding the relationship between neighborhood poverty and specific types of child maltreatment. Child Abuse & Neglect 20(11):1003-1018.

Drake, B., M. Jonson-Reid, I. Way, and S. Chung. 2003. Substantiation and recidivism. Child Maltreatment 8(4):248-260.

Drake, B., J. M. Jolley, P. Lanier, J. Fluke, R. P. Barth, and M. Jonson-Reid. 2011. Racial bias in child protection? A comparison of competing explanations using national data. Pediatrics 127(3):471-478.

Drury, S. S., K. Theall, M. M. Gleason, A. T. Smyke, I. De Vivo, J. Y. Y. Wong, N. A. Fox, C. H. Zeanah, and C. A. Nelson. 2011. Telomere length and early severe social deprivation: Linking early adversity and cellular aging. Molecular Psychiatry. Published online before print May 17, 2011, doi:10.1038/mp.2011.53.

Dubowitz, H., R. R. Newton, A. J. Litrownik, T. Lewis, E. C. Briggs, R. Thompson, D. English, L.-C. Lee, and M. M. Feerick. 2005. Examination of a conceptual model of child neglect. Child Maltreatment 10(2):173-189.

Dubowitz, H., S. Feigelman, W. Lane, and J. Kim. 2009. Pediatric primary care to help prevent child maltreatment: The Safe Environment for Every Kid (SEEK) model. Pediatrics 123(3):858-864.

Egeland, B., D. Jacobvitz, and L. A. Sroufe. 1988. Breaking the cycle of abuse. Child Development 59(4):1080-1088.

Egeland, B., T. Yates, K. Appleyard, and M. van Dulmen. 2002. The long-term consequences of maltreatment in the early years: A developmental pathway model to antisocial behavior. Children’s Services 5(4):249-260.

Ertem, I. O., J. M. Leventhal, and S. Dobbs. 2000. Intergenerational continuity of child physical abuse: How good is the evidence? Lancet 356(9232):814-819.

Fallon, B., N. Trocmé, J. Fluke, B. MacLaurin, L. Tonmyr, and Y.-Y. Yuan. 2010. Methodological challenges in measuring child maltreatment. Child Abuse & Neglect 34(1):70-79.

Farrington, D. 2011. Families and crime. In Crime and public policy, edited by J. Wilson and J. Petersilia. New York: Oxford University Press. Pp. 130-157.

FIFCFS (Federal Interagency Forum on Child and Family Statistics). 2011. America’s children: Key national indicators of well-being, 2011. Washington, DC: U.S. Goverment Printing Office.

Finkelhor, D. 2007. Developmental victimology: The comprehensive study of childhood victimizations. In Victims of crime, 3rd ed., edited by R. C. Davis, A. J. Lurigio, and S. Herman. Thousand Oaks, CA: Sage Publications. Pp. 9-34.

Finkelhor, D. 2008. Childhood victimization: Violence, crime, and abuse in the lives of young people. New York: Oxford University Press.

Finkelhor, D., and L. Jones. 2006. Why have child maltreatment and child victimization declined? Journal of Social Issues 62(4):685-716.

Finkelhor, D., R. K. Ormrod, H. A. Turner, and S. L. Hamby. 2005. Measuring poly-victimization using the Juvenile Victimization Questionnaire. Child Abuse & Neglect 29(11):1297-1312.

Flynn, B., J. Furtado, T. Orbach, and J. Scott. 2011. Designing and implementing differential response systems in child protective services: A three state case study. Boston, MA: Department of Urban and Environmental Policy and Planning, Tufts University.

Fraser, M. W., E. Walton, R. E. Lewis, P. J. Pecora, and W. K. Walton. 1996. An experiment in family reunification: Correlates of outcomes at one-year follow-up. Children and Youth Services Review 18(4-5):335-361.

Garner, A. S., J. P. Shonkoff, B. S. Siegel, M. I. Dobbins, M. F. Earls, L. McGuinn, J. Pascoe, and D. L. Wood. 2012. Early childhood adversity, toxic stress, and the role of the pediatrician: Translating developmental science into lifelong health. Pediatrics 129(1):e224-e231.

Gilbert, R., J. Fluke, M. O’Donnell, A. Gonzalez-Izquierdo, M. Brownell, P. Gulliver, S. Janson, and P. Sidebotham. 2012. Child maltreatment: Variation in trends and policies in six developed countries. Lancet 379(9817):758-772.

Gunnar, M. R., P. A. Fisher, and the Early Experience, Stress, and Prevention Science Network.2006. Bringing basic research on early experience and stress neurobiology to bear on preventive interventions for neglected and maltreated children. Development and Psychopathology 18(3):651-677.

Hanson, R. F., S. Self-Brown, A. E. Fricker-Elhai, D. G. Kilpatrick, B. E. Saunders, and H. S. Resnick. 2006. The relations between family environment and violence exposure among youth: Findings from the National Survey of Adolescents. Child Maltreatment 11(1):3-15.

Harris, W. W., F. W. Putnam, and J. A. Fairbank. 2004. Mobilizing trauma resources for children. Paper presented at Johnson and Johnson Pediatric Institute: Shaping the Future of Children’s Health, San Juan, Puerto Rico, February 12-16.

Hembree-Kigin, T. L., and C. B. McNeil. 1995. Parent-child interaction therapy. New York: Plenum Press.

Herrenkohl, R., and T. Herrenkohl. 2009. Assessing a child’s experience of multiple maltreatment types: Some unfinished business. Journal of Family Violence 24(7):485-496.

HHS (U.S. Department of Health and Human Services). 2011. Child maltreatment 2010. Washington, DC: HHS.

Hicks, B. M., S. C. South, A. C. DiRago, W. G. Iacono, and M. McGue. 2009. Environmental adversity and increasing genetic risk for externalizing disorders. Archives of General Psychiatry 66(6):640-648.

Horwitz, S., M. Hurlburt, and J. Zhang. 2010. Patterns and predictors of mental health services use by children in contact with the child welfare system. In Child welfare and child well-being, edited by M. Webb, K. Dowd, B. Harden, J. Landsverk, and M. Testa. New York: Oxford University Press. Pp. 279-329.

Hunter, W., and E. Knight. 1998. LONGSCAN research briefs, volume 1. Chapel Hill, NC: LONGSCAN Coordinating Center.

Huntington, C. 2006. Rights myopia in child welfare. UCLA Law Review 53:637-700.

Hussey, J. M., J. M. Marshall, D. J. English, E. D. Knight, A. S. Lau, H. Dubowitz, and J. B. Kotch. 2005. Defining maltreatment according to substantiation: Distinction without a difference? Child Abuse & Neglect 29(5):479-492.

IOM (Institute of Medicine). 2001. Crossing the quality chasm: A new health system for the 21st century. Washington, DC: National Academy Press.

IOM. 2002. Confronting chronic neglect: The education and training of health professionals on family violence. Washington, DC: National Academy Press. Isaacs, J. B. 2007. Cost-effective investments in children. Washington, DC: Brookings Institution.

Jaffee, S. R., A. Caspi, T. E. Moffitt, M. Polo-Tomás, and A. Taylor. 2007. Individual, family, and neighborhood factors distinguish resilient from nonresilient maltreated children: A cumulative stressors model. Child Abuse & Neglect 31(3):231-253.

Jonson-Reid, M., and R. P. Barth. 2000. From maltreatment report to juvenile incarceration: The role of child welfare services. Child Abuse & Neglect 24(4):505-520.

Jonson-Reid, M., and B. Drake. 2008. Multisector longitudinal administrative databases. Child Maltreatment 13(4):392-399.

Jonson-Reid, M., N. Presnall, B. Drake, L. Fox, L. Bierut, W. Reich, P. Kane, R. D. Todd, and J. N. Constantino. 2010. Effects of child maltreatment and inherited liability on antisocial development: An official records study. Journal of the American Academy of Child & Adolescent Psychiatry 49(4):321-332.

Jonson-Reid, M., P. Kohl, and B. Drake. 2012. Child and adult outcomes of chronic child maltreatment. Pediatrics. Published online before print April 23, 2012. http://www.pediatrics.aappublications.org/content/129/5/X5.short (accessed May 24, 2012).

Kauffman Best Practices Project. 2004. Closing the quality chasm in child abuse treatment: Identifying and disseminating best practices, the findings of the Kauffman Best Practices Project to help children heal from child abuse. San Diego, CA: Chadwick Center for Children and Families, Children’s Hospital-San Diego and the National Call to Action.

Kaufman, J., B. Birmaher, J. Perel, R. E. Dahl, P. Moreci, B. Nelson, W. Wells, and N. D. Ryan. 1997. The corticotropin-releasing hormone challenge in depressed abused, depressed nonabused, and normal control children. Biological Psychiatry 42(8):669-679.

Kaufman, J., P. M. Plotsky, C. B. Nemeroff, and D. S. Charney. 2000. Effects of early adverse experiences on brain structure and function: Clinical implications. Biological Psychiatry 48(8):778-790.

Kempe, C. H., F. N. Silverman, B. F. Steele, W. Droegemueller, and H. K. Silver. 1962. The battered-child syndrome. Journal of the American Medical Association 181(1):17-24.

Kilpatrick, D. G., and B. E. Saunders. 1999. Prevalence and consequences of child victimization: Results from the National Survey of Adolescents, final report. Charleston, SC: National Crime Victims Research and Treatment Center, Department of Psychiatry and Behavioral Sciences, Medical University of South Carolina.

Kilpatrick, D. G., and B. E. Saunders. 2009. National Survey of Adolescents in the United States, 1995. Ann Arbor, MI: Inter-university Consortium for Political and Social Research.

Kilpatrick, D. G., K. J. Ruggiero, R. Acierno, B. E. Saunders, H. S. Resnick, and C. L. Best. 2003. Violence and risk of PTSD, major depression, substance

abuse/dependence, and comorbidity: Results from the National Survey of Adolescents. Journal of Consulting and Clinical Psychology 71(4):692-700.

Knopf, D., M. J. Park, and T. P. Mulye. 2008. The mental health of adolescents: A national profile, 2008. San Francisco, CA: National Adolescent Health Information Center, University of California, San Francisco.

Kohl, P. L., M. Jonson-Reid, and B. Drake. 2009. Time to leave substantiation behind. Child Maltreatment 14(1):17-26.

Kolko, D. J., and C. C. Swenson. 2002. Assessing and treating physically abused children and their families: A cognitive-behavioral approach. Thousand Oaks, CA: Sage Publications.

Korbin, J. E., C. J. Coulton, S. Chard, C. Platt-Houston, and M. Su. 1998. Impoverishment and child maltreatment in African American and European American neighborhoods. Development and Psychopathology 10(02):215-233.

Kotch, J. B., D. C. Browne, C. L. Ringwalt, V. Dufort, E. Ruina, P. W. Stewart, and J.-W. Jung. 1997. Stress, social support, and substantiated maltreatment in the second and third years of life. Child Abuse & Neglect 21(11):1025-1037.

Langford, J. 2007. Strengthening families through early care and education. Washington, DC: Center for the Study of Social Policy.

Lanier, P., M. Jonson-Reid, M. J. Stahlschmidt, B. Drake, and J. Constantino. 2010. Child maltreatment and pediatric health outcomes: A longitudinal study of low-income children. Journal of Pediatric Psychology 35(5):511-522.

Leeb, R. T., L. J. Paulozzi, C. Melanson, T. R. Simon, and I. Arias. 2008. Child maltreatment surveillance uniform definitions for public health and recommended data elements, version 1.0. Atlanta, GA: Centers for Disease Control and Prevention.

Leventhal, J. M., K. D. Martin, and J. R. Gaither. 2012. Using U.S. data to estimate the incidence of serious physical abuse in children. Pediatrics. Published online before print February 6, 2012, doi: 10.1542/peds.2011-1277. http://www.pediatrics.aappublications.org/content/early/2012/02/01/peds.2011-1277.abstract.

Litrownik, A. J., A. Lau, D. J. English, E. Briggs, R. R. Newton, S. Romney, and H. Dubowitz. 2005. Measuring the severity of child maltreatment. Child Abuse & Neglect 29(5):553-573.

Lott, D. 2011. National Child Traumatic Stress Network: Changing the course of children’s lives. Rockville, MD: U.S. Department of Health and Human Services.

Lutzker, J. R., A. J. Tymchuk, and K. M. Bigelow. 2001. Applied research in child maltreatment: Practicalities and pitfalls. Children’s Services 4(3):141-156.

Lyons-Ruth, K. 2003. Dissociation and the parent-infant dialogue: A longitudinal perspective from attachment research. Journal of the American Psychoanalytic Association 51(3):883-911.

MacMillan, H. L., B. H. Thomas, E. Jamieson, H. A. Walsh, M. H. Boyle, H. S. Shannon, and A. Gafni. 2005. Effectiveness of home visitation by publichealth nurses in prevention of the recurrence of child physical abuse and neglect: A randomised controlled trial. Lancet 365(9473):1786-1793.

Manly, J. T., J. E. Kim, F. Rogosch, and D. Cicchetti. 2001. Dimensions of child maltreatment and children’s adjustment: Contributions of developmental timing and subtype. Development and Psychopathology 13(04):759-782.

McCrory, E., S. A. De Brito, and E. Viding. 2010. Research review: The neurobiology and genetics of maltreatment and adversity. Journal of Child Psychology and Psychiatry 51(10):1079-1095.

McEwen, B. S. 2000. Effects of adverse experiences for brain structure and function. Biological Psychiatry 48(8):721-731.

McMillen, J. C., B. T. Zima, L. D. Scott, Jr., W. F. Auslander, M. R. Munson, M. T. Ollie, and E. L. Spitznagel. 2005. Prevalence of psychiatric disorders among older youths in the foster care system. Journal of the American Academy of Child & Adolescent Psychiatry 44(1):88-95.

Melton, G. B., and B. Holaday. 2008. Strong communities as safe havens for children. In Family and community health, edited by G. B. Melton and B. Holaday. Vol. 31, no. 2. Hagerstown, MD: Wolters Kluwer Health, Lippincott Williams & Wilkins.

Mihalic, S., A. Fagan, K. Irwin, D. Ballard, and D. Elliott. 2001. Blueprints for violence prevention. Washington, DC: Office of Juvenile Justice and Delinquency Prevention.

Miller, M. 2011. Family team decision-making: Does it reduce racial disproportionality in Washington’s child welfare system? Olympia, WA: Washington Institute for Public Policy.

Moyer, V., M. LeFevre, and A. Siu. 2011. First annual report to Congress on high-priority evidence gaps for clinical preventive services. Rockville, MD: U.S. Preventive Services Task Force.

Nelson, C. A., C. H. Zeanah, N. A. Fox, P. J. Marshall, A. T. Smyke, and D. Guthrie. 2007. Cognitive recovery in socially deprived young children: The Bucharest Early Intervention Project. Science 318(5858):1937-1940.

Noonan, K. G., C. F. Sabel, and W. H. Simon. 2009. Legal accountability in the service-based welfare state: Lessons from child welfare reform. Law & Social Inquiry 34(3):523-568.

NQIC (National Quality Improvement Center on Differential Response in Child Protective Services). 2009. Differential response in child protective services: A literature review. (accessed May 24, 2012). NQIC. 2010. Differential response approach in child protective services: An analysis of state legislative provisions. (accessed May 24, 2012).

NRC (National Research Council). 1993. Understanding child abuse and neglect. Washington, DC: National Academy Press.

Olds, D. L., J. Eckenrode, C. R. Henderson, H. Kitzman, J. Powers, R. Cole, K. Sidora, P. Morris, L. M. Pettitt, and D. Luckey. 1997. Long-term effects of home visitation on maternal life course and child abuse and neglect. Journal of the American Medical Association 278(8):637-643.

Olds, D. L., C.R. Henderson, Jr., H. J. Kitzman, J. J. Eckenrode, R. E. Cole, and R. C. Tatelbaum. 1999. Prenatal and infancy home visitation by nurses: Recent findings. The Future of Children 9(1):44-65.

O’Reilly, R., L. Wilkes, L. Luck, and D. Jackson. 2010. The efficacy of family support and family preservation services on reducing child abuse and neglect: What the literature reveals. Journal of Child Health Care 14(1):82-94.

Ouellet-Morin, I., C. L. Odgers, A. Danese, L. Bowes, S. Shakoor, A. S. Papadopoulos, A. Caspi, T. E. Moffitt, and L. Arseneault. 2011. Blunted cortisol responses to stress signal social and behavioral problems among maltreated/bullied 12-year-old children. Biological Psychiatry 70(11):1016-1023.

Palusci, V. J. 2011. Risk factors and services for child maltreatment among infants and young children. Children and Youth Services Review 33:1374-1382.

Paolucci, E. O., M. L. Genuis, and C. Violato. 2001. A meta-analysis of the published research on the effects of child sexual abuse. Journal of Psychology 135(1):17.

Patterson, G. R., J. B. Reid, and J. M. Eddy. 2002. A brief history of the Oregon model. In Antisocial behavior in children and adolescents: A developmental analysis and model for intervention, edited by J. B. Reid, G. R. Patterson, and J. J. Snyder. Washington, DC: American Psychological Association. Pp. 3-20.

Paulsell, D., S. Avellar, E. Sama Martin, and P. Del Grosso. 2010. Home visiting evidence of effectiveness review: Executive summary. Washington, DC: U.S. Department of Health and Human Services.

Paxson, C., and R. Haskins. 2009. Introducing the issue. The Future of Children 19(2):3-17.

Pecora, P. J., J. K. Whittaker, A. N. Maluccio, R. P. Barth, and R. D. Plotnick. 2009. The child welfare challenge: Policy, practice, and research. New Brunswick, NJ: Aldine Transaction.

Prinz, R. J. 2009. Toward a population-based paradigm for parenting intervention, prevention of child maltreatment, and promotion of child wellbeing. In Preventing child maltreatment: Community approaches, edited by K. Dodge and D. Coleman. New York: Guilford Press. Pp. 55-67.

Prinz, R. J., and M. R. Sanders. 2006. Testing effects on parenting at a broad scale: The U.S. Triple P system population trial. In Familien starken: Evidenz-basierte anstze zur unterstutzung der psychischen gesundheit von kindern (Strengthening families: Evidence-based approaches to support child mental health), edited by N. Heinrichs, K. Hahlweg, and M. Dopfner. Munster, Germany: Verlag fur Psychotherapie. Pp. 483-511.

Prinz, R. J., and M. R. Sanders. 2007. Adopting a population-level approach to parenting and family support interventions. Clinical Psychology Review 27(6):739-749.

Prinz, R., M. Sanders, C. Shapiro, D. Whitaker, and J. Lutzker. 2009. Population-based prevention of child maltreatment: The U.S. Triple P system population trial. Prevention Science 10(1):1-12.

Riggins-Caspers, K. M., R. J. Cadoret, J. F. Knutson, and D. Langbehn. 2003. Biology-environment interaction and evocative biology-environment correlation: Contributions of harsh discipline and parental psychopathology to problem adolescent behaviors. Behavior Genetics 33(3):205-220.

Ryan, J. P., and M. F. Testa. 2005. Child maltreatment and juvenile delinquency: Investigating the role of placement and placement instability. Children and Youth Services Review 27(3):227-249.

Rycus, J. S., and R. C. Hughes. 2003. Issues in risk assessment in child protective services. Columbus, OH: Center for Child Welfare Policy, North American Resource Center for Child Welfare.

Sanders, M. R. 1999. Triple P-Positive Parenting Program: Towards an empirically validated multilevel parenting and family support strategy for the prevention of behavior and emotional problems in children. Clinical Child and Family Psychology Review 2(2):71-90.

Sanders, M. R., C. Markie-Dadds, and K. M. T. Turner. 2003. Theoretical, scientific and clinical foundations of the Triple P-Positive Parenting Program: A population approach to the promotion of parenting competence, Parenting research and practice monograph no. 1. Brisbane, Queensland, Australia: The Parenting and Family Support Centre.

Saunders, B. E., L. Berliner, and R. F. Hanson (eds.). 2004. Child physical and sexual abuse: Guidelines for treatment (revised report: April 26, 2004). Charleston, SC: National Crime Victims Research and Treatment Center.

Sawyer, R., and S. Lohrbach. 2005. Differential response in child protection: Selecting a pathway. Protecting Children: A Professional Publication of American Humane 20(2&3):44-54.

Schnitzer, P. G., P. L. Slusher, R. L. Kruse, and M. M. Tarleton. 2011. Identification of ICD codes suggestive of child maltreatment. Child Abuse & Neglect 35(1):3-17.

Seckl, J. R. 1998. Physiologic programming of the fetus. Clinics in Perinatology 25(4):939-962, vii.

Sedlak, A. J., K. McPherson, and B. Das. 2010a. Supplementary analyses of race differences in child maltreatment rates in the NIS-4. Washington, DC: U.S. Department of Health and Human Services.

Sedlak, A. J., J. Mettenburg, M. Basena, I. Petta, K. McPherson, A. Greene, and S. Li. 2010b. Fourth National Incidence Study of child abuse and neglect (NIS-4): Report to Congress. Washington, DC: U.S. Department of Health and Human Services, Administration for Children and Families.

Shenk, C. E., J. G. Noll, F. W. Putnam, and P. K. Trickett. 2010. A prospective examination of the role of childhood sexual abuse and physiological asymmetry in the development of psychopathology. Child Abuse & Neglect 34(10):752-761.

Shields, A., and D. Cicchetti. 1998. Reactive aggression among maltreated children: The contributions of attention and emotion. Journal of Clinical Child Psychology 27(4):381.

Shipman, K. L., and J. Zeman. 2001. Socialization of children’s emotion regulation in mother-child dyads: A developmental psychopathology perspective. Development and Psychopathology 13(02):317-336.

Shonkoff, J. P., A. S. Garner, the Committee on Psychosocial Aspects of Child and Family Health, Committee on Early Childhood Adoption, and Dependent Care, and Section on Developmental and Behavioral Pediatrics, B. S. Siegel, M. I. Dobbins, M. F. Earls, A. S. Garner, L. McGuinn, J. Pascoe, and D. L. Wood. 2012. The lifelong effects of early childhood adversity and stress. Pediatrics, published online before print December 26, 2011. http://pediatrics.aappublications.org/content/129/1/e232.abstract (accessed May 24, 2012).

Slack, K.S., J. L. Holl, M. McDaniel, J. Yoo, and K. Bolger. 2004. Understanding the risks of child neglect: An exploration of poverty and parenting characteristics. Child Maltreatment 9(4):395-408.

Stahmer, A. C., L. K. Leslie, M. Hurlburt, R. P. Barth, M. B. Webb, J. Landsverk, and J. Zhang. 2005. Developmental and behavioral needs and service use for young children in child welfare. Pediatrics 116(4):891-900.

Stevenson, J. 1999. The treatment of the long-term sequelae of child abuse. Journal of Child Psychology and Psychiatry and Allied Disciplines 40(1):89-111.

Strathearn, L., P. H. Gray, M. J. O’Callaghan, and D. O. Wood. 2001. Childhood neglect and cognitive development in extremely low birth weight infants: A prospective study. Pediatrics 108(1):142-151.

Stroufe, L. A., B. Egeland, E. A. Carlson, and W. A. Collins. 2005. The development of the person: The Minnesota study of risk and adaptation from birth to adulthood. New York: The Guilford Press.

Sullivan, P. M., and J. F. Knutson. 2000. Maltreatment and disabilities: A population-based epidemiological study. Child Abuse & Neglect 24(10):1257-1273.

Suomi, S. J. 1997. Early determinants of behaviour: Evidence from primate studies. British Medical Bulletin 53(1):170-184.

Taylor, A., and J. Kim-Cohen. 2007. Meta-analysis of gene-environment interactions in developmental psychopathology. Development and Psychopathology 19(04):1029-1037.

Teicher, M. H., S. L. Andersen, A. Polcari, C. M. Anderson, C. P. Navalta, and D. M. Kim. 2003. The neurobiological consequences of early stress and childhood maltreatment. Neuroscience and Biobehavioral Reviews 27(1-2):33-44.

Thomas, D., C. Leicht, C. Hughes, A. Madigan, and K. Dowell. 2003. Emerging practices in the prevention of child abuse and neglect. Washington, DC: U.S. Department of Health and Human Services.

Trickett, P. K., L. Allen, C. J. Schellenbach, and E. F. Zigler. 1998. Violence against children in the family and the community. In Violence against children in the family and the community, edited by P. K. Trickett and C. J. Schellenbach. Washington, DC: American Psychological Association. Pp. 419-437.

Urquiza, A. J., and C. B. McNeil. 1996. Parent-child interaction therapy: An intensive dyadic intervention for physically abusive families. Child Maltreatment 1(2):134-144.

U.S. Advisory Board on Child Abuse and Neglect. 1990. Child abuse and neglect: Critical first steps in response to a national emergency. Washington, DC: U.S. Department of Health and Human Services.

USPHS (U.S. Public Health Service). 2001. Youth violence: A report of the Surgeon General. Rockville, MD: USPHS, Office of the Surgeon General.

Waldfogel, J. 1998. Rethinking the paradigm for child protection. The Future of Children 8(1):104-119.

Waldrop, A. E., R. F. Hanson, H. S. Resnick, D. G. Kilpatrick, A. E. Naugle, and B. E. Saunders. 2007. Risk factors for suicidal behavior among a national sample of adolescents: Implications for prevention. Journal of Traumatic Stress 20(5):869-879.

Webster-Stratton, C. 1998. Parent training with low-income families: Promoting parental engagement through a collaborative approach. In Handbook of child abuse research and treatment, edited by J. R. Lutzker. New York: Plenum Press. Pp. 183-212.

Webster-Stratton, C., and M. J. Reid. 2003. Treating conduct problems and strengthening social and emotional competence in young children. Journal of Emotional and Behavioral Disorders 11(3):130-143.

Weder, N., B. Z. Yang, H. Douglas-Palumberi, J. Massey, J. H. Krystal, J. Gelernter, and J. Kaufman. 2009. Maoa genotype, maltreatment, and aggressive behavior: The changing impact of genotype at varying levels of trauma. Biological Psychiatry 65(5):417-424.

WHO (World Health Organization). 1999. International statistical classification of diseases, 9th revision. Geneva, Switzerland: WHO. WHO. 2003. International statistical classification of diseases and related problems, 10th revision. Geneva, Switzerland: WHO.

Widom, C. S. 1989. Does violence beget violence? A critical examination of the literature. Psychological Bulletin 106(1):3-28.

Widom, C. S. 1999. Posttraumatic stress disorder in abused and neglected children grown up. American Journal of Psychiatry 156(8):1223-1229.

Widom, C. S., and L. M. Brzustowicz. 2006. MAOA and the “cycle of violence”: Childhood abuse and neglect, MAOA genotype, and risk for violent and antisocial behavior. Biological Psychiatry 60(7):684-689.

Widom, C. S., K. DuMont, and S. J. Czaja. 2007. A prospective investigation of major depressive disorder and comorbidity in abused and neglected children grown up. Archives of General Psychiatry 64(1):49-56.

Wulczyn, F. 2009. Epidemiological perspectives on maltreatment prevention. The Future of Children 19(2):39-66.

Yanos, P., S. Czaja, and C. Widom. 2010. A prospective examination of service use by abused and neglected children followed up into adulthood. Psychiatric Services 61:796-802.

Zeanah, C., C. Nelson, N. Fox, A. Smyke, P. Marshall, S. Parker, and S. Koga. 2003. Designing research to study the effects of institutionalization on brain and behavioral development: The Bucharest early intervention project. Development and Psychopathology 15(04):885-907.

Zeanah, C. H., A. T. Smyke, S. F. Koga, E. Carlson, and The Bucharest Early Intervention Project Core Group. 2005. Attachment in institutionalized and community children in Romania. Child Development 76(5):1015-1028.

This page intentionally left blank.