New Directions in Child Abuse and Neglect Research (2014)

2


Describing the Problem

Child abuse and neglect is well established as an important societal concern with significant ramifications for the affected children, their families, and society at large (see Chapter 4). A critical step in devising effective responses is reasonable agreement on the definition of the problem and its scope. Yet achieving clarity in the area of child abuse and neglect has been an ongoing challenge. Legal definitions vary across states; researchers apply diverse standards in determining incidence and prevalence rates in clinical and population-based studies; and substantial obstacles hamper learning about the experiences of children, especially young children, with caregiver-inflicted abuse or neglect. As a result, definitions of the characteristics of the problem and determinations of its scope will differ depending on the data source used for analysis. This challenge was articulated in the 1993 National Research Council (NRC) report (NRC, 1993) and continues to impede a full understanding of the nature of the child abuse and neglect problem. The purpose of this chapter is to describe briefly what is known about the problem from current data sources and to highlight issues that remain problematic, as well as identify areas in which advances have been made. The chapter addresses, in turn, definitions of child abuse and neglect, incidence rates and the problem of underreporting, trends in the incidence of child abuse and neglect, and how cases are determined by medical and mental health professionals and the legal system. The final section presents conclusions.

DEFINITIONS

A key definition of child abuse and neglect is contained in Section 3 of the Child Abuse Prevention and Treatment Act (CAPTA)¹:

At a minimum, any recent act or set of acts 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 is especially important because it is enshrined in federal legislation. To be eligible to receive funding under Section 106² of the act, states must, at a minimum, include the conduct described in Section 3 in their state child abuse and neglect authorizing legislation. All 50 states, as well as American Samoa, the Commonwealth of Puerto Rico, the Commonwealth of the Northern Mariana Islands, the District of Columbia, Guam, and the Virgin Islands, have mandatory child abuse and neglect reporting laws that define the terms slightly differently for their jurisdiction and lay out the requirements for mandatory reporting (CWIG, 2011). Federal law defines child abuse and neglect and identifies reporting requirements on tribal lands³ (see CWIG, 2012b, for further information) and on military installations⁴ (see Military OneSource, n.d., for further information); in some circumstances, state laws on child abuse and neglect reporting also apply to tribal lands and military installations. The Victims of Child Abuse Act⁵ (also see Chapter 8) lays out requirements for reporting child abuse that occurs on federal lands and in federal facilities.

The National Child Abuse and Neglect Data System (NCANDS) is the official government data source to which all states must contribute information about child abuse and neglect reports. To collect data on reported and confirmed cases of child abuse and neglect uniformly from all states, NCANDS provides the following somewhat more comprehensive definition of child abuse and neglect:

An act or failure to act by a parent, caregiver, or other person as defined under State law that results in physical abuse, neglect, medical neglect, sexual abuse, emotional abuse, or an act or failure to act which presents an imminent risk of harm to a child. (ACF, 2012)

Many states, reflecting the words “at a minimum” in CAPTA, have more expansive definitions of the conduct that legally constitutes child

________________

¹42 U.S.C. § 5101 note.

²42 U.S.C. § 52016a.

³25 U.S.C. § 1169.

⁴10 U.S.C. § 1787.

⁵42 U.S.C. § 13001, et seq.

abuse and neglect for purposes of mandatory reporting. In some states, for example, only conduct by current caregivers is defined as reportable child abuse and neglect; in other states, the conduct must be reported regardless of the perpetrator’s relationship to the child. Pennsylvania, for example, considers only acts of abuse as reportable acts of maltreatment and uses a different mechanism for capturing neglect. CAPTA permits states to limit reporting to “recent” acts, but most states have no time limit on when the conduct occurred for the mandatory reporting requirement to be invoked. A summary of the differences in states’ child abuse and neglect reporting laws is available (CWIG, 2011).

How child abuse and child neglect are defined and who is obligated to report them are subject to changes in awareness or level of concern about possible abuse- and neglect-related hazards faced by children. It is common for a specific case, especially one involving an egregious situation not addressed by extant law, to prompt advocacy for legislative change (Gainsborough, 2010). Newly identified problem areas, changes in societal consensus about child protection, and revelations that certain groups of professionals are not included in mandatory reporting laws are typical scenarios for bringing about statutory reforms. In 2012, 107 bills addressing child abuse and neglect reporting were introduced in 30 states and the District of Columbia (NCSL, 2012). For example, a number of states expanded mandatory reporting to apply to university employees in response to the Penn State Sandusky scandal.

In some cases, such changes have unintended consequences. An example is the occasional inclusion of exposure to domestic violence as a statutorily specified form of reportable child abuse and neglect, a result of increasing awareness of the association between domestic violence and child abuse and neglect and concern for the welfare of children exposed to this violence, so that affected children would receive protection and services. The Minnesota state legislature instituted such a change in 1999. The result was a dramatic increase in the number of referrals, emanating mainly from law enforcement officials who responded to reports of domestic violence and, as mandated, reported the family to child protective services. Parents, primarily mothers, who themselves were victims of domestic violence thus became the subjects of neglect reports based on their alleged failure to protect their children from exposure to the violence. This was not the intent of the legislation, and the provision was quickly rescinded (Edleson et al., 2006).

Child abuse and neglect laws are for the most part concerned with parental behaviors of omission or commission that place children in jeopardy. Acts of omission usually are characterized as neglect. They include failing to provide adequate supervision; not protecting children from known dangers; and not providing for basic needs, such as proper medical care, adequate

food and clothing, safe/hygienic shelter, and school attendance. Child neglect reports may also be made in some states if a child is born affected by illegal drug or alcohol abuse by the mother or if a child is living where drugs are being manufactured and/or distributed.

Child abuse, on the other hand, refers to acts of commission by a caregiver. Physical abuse encompasses physical assaults that exceed permitted corporal punishment. States may define explicitly the types of behavior that fall in this category. In some cases for example, the age of the child may determine whether a behavior is acceptable discipline (e.g., slapping an infant versus an older child across the face). Sexual abuse generally includes the range of sexual behaviors that are defined by criminal statutes, including sexual exposure, sexual touching, rape, and sexual exploitation. Emotionally abusive behaviors include threatening, terrorizing, or deliberately frightening a child; rejecting, ridiculing, shaming, or humiliating behaviors; extreme isolating or restricting behaviors; and corruption or encouraging involvement in illegal behaviors. However, of the 48 states that mention emotional abuse in law, only Delaware identifies specific emotionally abusive caregiver behaviors; most states define emotional abuse by its impact on the child’s mental health (CWIG, 2011). Because the involvement of the child protection system focuses on caregivers, cases of abuse committed by non-family members or siblings may be classified as neglect. In those cases, it is the presumed or alleged failure of the caregiver to protect the child that drives the designation. For example, the majority of sexual abuse and a notable proportion of serious physical abuse cases involve non-family members as perpetrators (Finkelhor and Dziuba-Leatherman, 1994). Instances of abuse committed by a non-family member, a sibling, or another person regularly present in the household are classified as neglect if it is determined that the caregiver failed to protect the child victim from that individual.

As noted, child abuse and neglect laws also vary in how mandated reporters are defined. Some states define all adult citizens as mandated reporters, but most specify certain groups of professionals and others who work with children (CWIG, 2012a). State laws usually exempt from a reporting obligation priests acting in the role of receiving confession; states vary, however, as to whether reporting is required of priests or pastors acting in other capacities. Regardless of the groups specified, anyone not listed as a mandated reporter can still make a report. Both mandated reporters and others are legally protected for good faith reports, while mandated reporters who fail to report may be prosecuted for that failure. No evidence-based research has assessed whether the breadth of inclusion in mandatory reporting laws makes a difference in rates of reporting, although it may affect substantiation rates (McElroy, 2012; also see the discussion of mandatory reporting laws in Chapter 8).

Some acts of child abuse and neglect are also crimes. The specific statu-

tory definitions and names of those crimes vary by state, but in general, criminal statutes cover the same acts in all states. Sexual abuse is always a crime; most cases are classified as felonies. Physical abuse is a crime unless the behavior falls within the discipline exception for corporal punishment. Most cases of physical abuse are likely to be classified as misdemeanors unless a child is seriously injured or dies. A minority of neglect cases involve criminal conduct. When the failure to supervise, protect, or provide care for a child rises to a certain level of negligent treatment, it may meet the criteria for violation of criminal codes (e.g., child endangerment or criminal neglect) and can be prosecuted. Just because child abuse and neglect falls within the statutory definition of a crime, however, does not mean it will be fully investigated by law enforcement and prosecuted. Law enforcement investigations and prosecutions tend to focus on sexual abuse and on serious physical abuse and very serious neglect that have resulted in a child’s experiencing physical harm or death (e.g., starvation, inflicted medical trauma).

As with state laws, child abuse and neglect is defined in various ways for research purposes. The National Incidence Study (NIS)-4 (Sedlak et al., 2010a) applies two definitional standards: a harm standard and an endangerment standard. The harm standard is restricted to cases in which children have been harmed by child abuse and neglect, whereas the endangerment standard encompasses children who have not yet been harmed under certain circumstances. The numbers vary depending on which definition is used (NIS-4 harm standard = 1.25 million children; endangerment standard = 3 million children). Under both standards, alleged instances of abuse or neglect are classified according to eight major categories. Table 2-1 lists actions or failures to act that are representative of each type of abuse or neglect and, for the purposes of this chapter, provides examples of how these forms of maltreatment can be defined in a research setting.

A widely used method of defining child abuse and neglect in research is the classification scheme developed by Barnett and colleagues (1993). Many studies focused specifically on child abuse and neglect use these definitions rather than the officially reported labels (e.g., English et al., 2005). The Centers for Disease Control and Prevention (CDC) also has recommended a set of uniform definitions for public health purposes to allow for monitoring of incidence over time and detection of trends (Leeb et al., 2008). Notably, both the classification scheme developed by Barnett and colleagues and the CDC recommendations are designed for analysis of existing information from public sources, primarily child protective services case records.

Slack and colleagues (2003) note that research definitions developed for analysis of child protective services case records may not be applicable to survey research. They argue that these definitions may capture risk factors associated with the detection of child abuse and neglect rather than risk factors associated with the commission of child abuse and neglect. They

TABLE 2-1 National Incidence Study (NIS)-4 Abuse and Neglect Classifications

SOURCE: Sedlak et al., 2010a.

have built on the framework created by Barnett and colleagues (1993) to develop a set of research definitions for neglect that they intend for use in survey research.

Likewise, other investigators develop their own study-specific designations. These definitions vary in comprehensiveness and behavioral specificity. For example, a study not focused specifically on child abuse and neglect but interested in it as one of many independent variables may use a single general question to get at the construct.

Finding: Child abuse and neglect are defined differently for different purposes. Legal definitions at the state level are properly subject to the legislative process. In research, however, the variability in definitions

compromises learning the true scope and characteristics of the problem, understanding trends over time, and determining the relationship between child abuse and neglect and various outcomes.

Finding: State laws vary in what groups are specified as mandated reporters of child abuse and neglect. No evidence-based research has assessed whether the breadth of inclusion in mandatory reporting laws makes a difference in rates of reporting, although it may affect substantiation rates.

INCIDENCE RATES AND THE PROBLEM OF UNDERREPORTING

Determining the true incidence of child abuse and neglect is problematic for the same reason encountered in attempting to quantify any social problem: discrepancies between actual rates and the number of cases reported to authorities. It is well established that most crimes (the exception being homicide) are not reported (Langton et al., 2012). Data on the incidence of child abuse and neglect are derived from three primary sources: NCANDS, the official reporting system for cases of child abuse and neglect referred to state child protective services; two U.S. government surveys—the Uniform Crime Reporting (UCR) system, administered by the Federal Bureau of Investigation (FBI), and the National Crime Victimization Survey (NCVS), administered by the Bureau of Justice Statistics (BJS) to a large representative sample of U.S. citizens aged 12 and older; and the NIS, a study conducted every decade by the Department of Health and Human Services on a nationally representative sample that captures both cases of abuse and neglect reported to child protective services and unreported cases identified by professionals working with children.

National Child Abuse and Neglect Data System

Each state receiving a federal Basic State Grant for child abuse and neglect prevention and treatment programs is required to submit data annually to NCANDS.⁶ In fiscal year (FY) 2011, all states, the District of Columbia, and all territories contributed to NCANDS counts of the number of cases referred to child protective services, the number accepted for investigation, the number substantiated, the case characteristics, and the case outcomes. As previously noted, the definitions of child abuse and neglect used by child protective services vary by state, as do reporting requirements. Because NCANDS collects information from child protective services case files in each state, the data reflect inconsistencies in state-level definitions of

________________

⁶42 U.S.C. § 5106a(d).

types of maltreatment, reporting requirements, and procedures for responding to reports of child abuse and neglect.

NCANDS reports are issued annually. According to the FY 2011 NCANDS report (ACF, 2012), there were 3.4 million referrals involving 6.2 million children; some of the children were the subject of more than one referral. Nationally, more than three-quarters of these cases are classified as neglect, 18 percent as physical abuse, and 9 percent as sexual abuse. The specific rates vary among states but overall reflect the general pattern that a substantial majority of cases are neglect, with physical and sexual abuse representing much smaller groups.

Based on NCANDS, victims of child abuse and neglect are approximately evenly divided between males and females. The highest rates of child abuse and neglect occur among the very youngest children (see Table 2-2). Perpetrators are mainly parents (81 percent) and among parents are primarily biological parents (88 percent), which reflects the legal definition for reportable cases. Somewhat more than half of perpetrators are female (ACF, 2012). These demographic characteristics are also reflected in other data sources, such as the NIS-4 (Sedlak et al., 2010a).

In FY 2011, NCANDS reported 1,545 child fatalities resulting from abuse and neglect. Again, young children were at greatest risk: 80 percent of victims were less than 4 years old. Deaths were higher among boys than girls. About 70 percent of the fatalities are associated with neglect and nearly half are attributed to physical abuse, either exclusively or in combination. A Government Accountability Office (GAO, 2011) report notes that the NCANDS method relies only on cases reported to child protective services for these figures. The report states that not all child fatalities due to abuse and neglect are known to the child welfare system, suggesting that the actual figure is likely higher, although it acknowledges the difficulty of obtaining an accurate count.

An important limitation of NCANDS is that it does not capture accurate rates of child abuse and neglect among American Indian children. Only states submit information to NCANDS; there are no mechanisms for tribal child welfare systems to submit data to the system. American Indian and Alaska Native families and children whose cases are reported to and investigated by state child protection authorities and who self-identify as American Indian or Alaska Native are included in NCANDS. Children served by tribal child welfare systems, the Bureau of Indian Affairs, or the Indian Health Service are not. Thus, “it is estimated that 40 percent of all cases of child abuse and neglect among American Indian and Alaska Native children are not reported to NCANDS” (Cross and Simmons, 2008, p. 3; also see Earle and Cross, 2001). NCANDS is further limited in its ability to reveal the levels of abuse and neglect suffered by American Indian and Alaska Native children by the fact that state or county employees, rather than

TABLE 2-2 Child Maltreatment Cases/Victims, Rates per Thousand Population Ages 0-17,^a by Various Characteristics, 2002-2011

^aIncludes “substantiated” cases, in which investigation results in a disposition concluding that the allegation of maltreatment or risk of maltreatment was supported or founded according to state law or policy—the highest level of finding by a state agency. Also includes cases designated “indicated” or “reason to suspect,” which are those not substantiated by investigation but for which there is reason to suspect that the child may have been maltreated or was at risk of maltreatment. Not all states distinguish between substantiated and indicated dispositions. All data for 2009 and later represent “unique” cases—children who have experienced at least one instance of substantiated or indicated maltreatment (see definition above), with duplicate cases removed.

^bEstimates for specific racial groups have been revised to reflect the new Office of Management and Budget race definitions and include only those who are identified with a single race. Hispanics may be of any race.

^cA child may be a victim of multiple types of maltreatment and is counted once for each type(2007, when children were counted once only, was an exception).

SOURCES: CDC, 2003; Children’s Bureau, 1995-2011, reprinted with permission from Child Trends(2013).

tribal workers, collect the data reported to NCANDS. Therefore, not only does NCANDS lack data on many cases that occur on tribal lands, but the data it does include may be flawed because non-Native workers with American Indian or Alaska Native culture often are tasked with making determinations of abuse or neglect in such settings (Fox, 2004).

U.S. Government Surveys

The U.S. government uses the two surveys noted above to learn about crime rates. The UCR covers crimes reported to police, whereas the NCVS is a household survey of a large representative sample of individuals aged 12 and older that asks about both reported and unreported crimes. Self-reported rates of crime victimization frequently are several times the rates of official reports, with the discrepancies being especially high for sexual assault.

The ability of such surveys to capture cases accurately hinges, in part, on how the question is asked. Using official terminology or labels for acts of child abuse and neglect requires respondents to label their own experiences as abusive or neglectful. In some cases, respondents may not know the official definitions or exactly what they encompass. For example, many children and adults may consider hitting a child with a belt appropriate corporal punishment. In other cases, the victim may be reluctant to define what happened as abusive. For example, evidence suggests that labeling acts as intentionally abusive is associated with increased distress in children (Kolko et al., 2002).

These labeling considerations are particularly acute in cases of sexual assault. Asking a single question—such as “Have you ever been raped?”—yields far fewer responses than a series of behaviorally specific questions about acts that meet the legal definition of sexual abuse and rape. For example, rates of endorsement of child sexual abuse in self-report research vary substantially based on how the question is posed. A meta-analysis of studies that used self-report surveys to examine childhood sexual abuse experiences around the world found that differences in the way sexual abuse was defined and the specific questions asked produced dramatically different rates of sexual abuse prevalence (Stoltenborgh et al., 2011).

In addition to these survey design issues, the point in time and circumstances under which respondents provide information about child abuse and neglect are crucial. Surveys of adults about their childhood experiences may yield very different rates than surveys of children. For example, population-based telephone interviews of youth aged 10 and older provide extensive information about self-reported victimization and exposure to violence (Finkelhor, 2009; Kilpatrick and Saunders, 1995). However, the rates of intrafamilial sexual and physical abuse reported in these studies are

relatively low compared with the rates reported among adult samples when asked their childhood abuse experiences. Children may be less likely to report intrafamilial crimes when they are still children and are living at home.

Another method of learning about child abuse and neglect is asking adults about their behavior toward their children. Surveys using the Conflict Tactics Scale can provide a picture of self-reported corporal punishment and parental acts that would meet legal criteria for child physical abuse (Straus and Stewart, 1999; Straus et al., 1998; Theodore et al., 2005). This method has the obvious limitation, however, that even when responding to anonymous surveys, parents may underreport socially undesirable or illegal acts.

Discrepancies between official reports and child and adult self-reports can be in either direction. Children or adults may not define their experiences as child abuse and neglect because they do not know better or believe the conduct was deserved or acceptable, or because of the distress associated with reporting that caregivers are behaving abusively toward them. Adults may not define their own behavior as abusive or neglectful because of fears of being reported, social undesirability, or shame about the conduct. On the other hand, substantial evidence shows that careful and detailed questioning of children about their experiences produces substantially higher rates than official reports. For example, computer-assisted interviews were used to obtain self-reports of abuse and neglect from a sample of youth aged 12-13 enrolled in a prospective study of high-risk and abused children (Everson et al., 2008). This method yielded rates that were four to six times higher than those in the official child protective services records. At the same time, close to half of adolescents in the sample with confirmed child protective services reports failed to note that experience in the interview.

The National Incidence Study

The NIS is a congressionally mandated report on the incidence of child abuse and neglect that has been issued periodically since 1974 (OPRE, 2009). It estimates national rates of reported and unreported child abuse and neglect based on a representative sample of counties. The study uses official data and also collects information from “sentinels” representing community professionals who may encounter child abuse and neglect victims during the course of their work. The methodology of the NIS is explicitly designed to uncover child abuse and neglect that may not have been reported to authorities but was identified by professionals. The most recent report, issued in 2010, is based on data collected in 2005-2006 (Sedlak et al., 2010a). As noted above, the NIS defines child abuse and neglect differently from federal and state law, applying both a harm and an endan-

germent standard. All cases sampled in the study—both those identified by child protective services agencies and those reported by sentinels—are evaluated to determine whether they meet the definitional standards of the NIS for physical abuse, sexual abuse, emotional abuse, physical neglect, emotional neglect, and educational neglect. The NIS considers only abuse and neglect perpetrated or permitted by a parent or caregiver, aligning its definitions with those of child protective services.

The primary investigators of the NIS-4 note that findings of differential incidence rates for abuse and neglect of black and white children are limited by the range of risk factors available for analysis in multifactor risk models, which exclude such key elements as neighborhood characteristics, social isolation, substance use, and mental illness (Sedlak et al., 2010b). Likewise, many children’s records lacked information on socioeconomic status, and the socioeconomic status measures used classified black and white children differently, limiting the utility of the data for examining socioeconomic status as a risk factor for child abuse and neglect.

Reasons for Underreporting

It is well known that not all child abuse and neglect cases come to the attention of authorities at the time they happen. Retrospective reports from adults abused or neglected as children reveal that most cases are not reported to anyone, and fewer still are reported and investigated by child protection workers or law enforcement officials (e.g., Finkelhor, 1994; MacMillan et al., 2003). Adults abused or neglected as children give a variety of explanations for why they did not tell anyone or make an official report, including not realizing that what was happening was wrong, illegal, or a form of child abuse and having fears or concerns about what would happen if they reported the experience or attempted to seek help.

Child abuse and neglect can sometimes be identified without a child’s making a statement about it. Examples include certain types of injuries or medical conditions that are noticed by others or become known to a medical provider. Some types of neglect can also be detected through observable behaviors, such as young children found wandering the streets or coming to school unclean or very disheveled. But detection of many cases of physical abuse and neglect and almost all cases of sexual abuse depends largely on children making statements and adults acting on those statements. The statements may be made spontaneously or may be in response to adult inquiries about behaviors, circumstances, or injuries observed in the children. Once abuse or neglect has been detected, many variables can affect whether adults take action, including personal attitudes and beliefs about what will happen as a result of reporting, the relationship of the adult to the child or the caregiver who may have committed the abuse or neglect, the certainty of

the concern about maltreatment, and understanding of the child abuse reporting laws (Alvarez et al., 2004; Khan et al., 2005; Sedlak et al., 2010a).

Therefore, official reports do not capture all instances in which child abuse and neglect is suspected or even is detected and acted upon. For example, adults in a child’s life may learn about child abuse and neglect and take informal actions on behalf of the child without necessarily reporting to authorities. Although citizens are protected if they make a good faith report of suspected child abuse or neglect, there are many reasons why they might be hesitant about or deterred from making an official report even if strong evidence or suspicion exists. For example, they may fear retaliation or rejection by the abuser or negative consequences for the child or family. Indeed, despite the fact that relatives, neighbors, and friends are most likely to observe or hear about child abuse or neglect because of their proximity and involvement in children’s lives, they account for only a minority (18 percent) of reporters of cases to child protective services (ACF, 2012).

Professionals account for the other three-fifths of child abuse and neglect reports, with teachers (16 percent), law enforcement officials (17 percent), and social service providers (11 percent) making the majority of these reports (ACF, 2012). However, mandated reporters do not always make a report when they suspect child abuse or neglect. Among mandated reporters involved as sentinels in the NIS-4, a significant percentage have had suspicion and not made a report. Professionals identify a variety of reasons for not reporting their suspicions (Sedlak et al., 2010a). The most common reasons given are concerns that intervention by child protective services will be more harmful than helpful and the professionals’ belief that they can do a better job of addressing the suspected child abuse or neglect on their own without involving the authorities. Rates of reporting also may vary by profession and relationship with the family. In one state survey of pediatricians, only 10 percent had ever not reported a suspected case of abuse or neglect; the most common reason given was not feeling that the evidence for suspicion was strong enough or believing that the case could be better handled by the physician or family without the involvement of child protective services (Theodore and Runyan, 2006). For mental health providers, the dilemma may be more acute. For example, Steinberg and colleagues (1997) found that among psychologists who had made a report to child protective services, 27 percent stated that their client ended the therapy relationship because of the child abuse report.

In addition to the concerns of professionals about the consequences of reporting for themselves and their practice, a lack of clarity exists as to what constitutes reasonable suspicion as defined by the law. There is little dispute about suspicion when the basis for concern is clear-cut (e.g., the child makes a credible statement about being sexually abused or has hand print bruises on the cheek). In many cases, however, the information avail-

able to the reporter is vague, inconclusive, or only suggestive. Is it neglect when a child comes to school in dirty clothes and smelling bad? How young a child can be left alone at home? What if a child says, “I am afraid to go home”? If a child is engaging in highly sexualized behavior, is that indicative of abuse? There is a substantial gray area that is open to interpretation with respect to whether a statement or behavior meets criteria for triggering a legally mandated report of child abuse and neglect. A lack of consensus exists even among expert child abuse doctors. Levi and Crowell (2011) found no agreement among experts on how high child abuse and neglect would have to be on the list of differential diagnoses and how certain the provider would have to be that child abuse and neglect accounted for the child’s presentation to meet the reporting criterion of reasonable suspicion.

On the other hand, only about 60 percent of referrals to child protection authorities are accepted and screened in for some type of official response (ACF, 2012). Cases may be screened out because they do not meet the legal criteria for child abuse and neglect or state standards for accepting cases, or because information about the case is insufficient to enable completing a report. Among states, screen-in rates range from a low of 25 percent to a high of virtually all referrals (ACF, 2012). Thus citizens and professionals likely recognize many situations in which they suspect child abuse and neglect, but their suspicions do not meet the threshold of concern required by local statute to justify an investigation.

Disproportionality

Concerns have been raised about possible racial and ethnic bias in child abuse and neglect reporting and investigations because African American and American Indian children are referred to child protective services at higher rates than their representation in the population, whereas Asian American and Latino children are referred at lower rates. A recent study used a birth cohort methodology and linked vital statistics and child abuse report records for young children (Putnam-Hornstein, 2011). Prior child abuse reports were associated with an almost sixfold increase in the probability of intentional death and double the rate of unintentional fatal injury; the rates were higher for African American and American Indian children and lower for Asian American and Latino children relative to the general population. In other words, the racial/ethnic patterns of injury and death mirror the child abuse and neglect reporting rates by racial and ethnic group. Moreover, the overall underrepresentation of Latino children in referrals to the child welfare system masks significant differences between the experiences of Latino children of U.S.-born mothers and Latino children of foreign-born mothers, both in rates of referral (Putnam-Hornstein et al., 2013) and in type of abuse or neglect (Dettlaff and Johnson, 2011).

Authoritative commentators (Drake et al., 2011; Putnam-Hornstein, 2012; Putnam-Hornstein et al., 2013) agree that there are real group differences in the rates of child abuse and neglect and conclude that these differences reflect the higher burden of social ills borne by some groups. As Putnam-Hornstein concludes: “The findings suggest that the overrepresentation of black and Native American children in the child welfare system may be a manifestation of historical and contemporary racial inequities that place these minority children at a disproportionate risk of maltreatment” (2012, p. 171).

Disproportionality extends beyond referrals. Miller (2011) examined disproportionality in Washington state at both the referral point and key decisions points after cases had been screened in (e.g., risk rating, placement, length of time in care). As with other states, disproportional rates of referral were seen. When disproportionality from the point of referral was examined, virtually no differences were found among whites, Asians, and Latinos following case entry into the child welfare system. After case receipt, rates of disproportionality were reduced for African American families at most decision points, with the largest discrepancy remaining in length of time in care. For American Indian cases, the disproportionality continued at every decision point following case acceptance. These results suggest that the observed disproportionality may have a variety of causes, some that reflect larger social forces and others that may be more reflective of professional assumptions and local practices. Disproportionality is discussed further in Chapter 5 of this report.

Finding: According to NCANDS data from FY 2011, there were 3.4 million child abuse and neglect referrals involving 6.2 million children. Nationally, more than three-quarters of these cases are classified as neglect, 18 percent as physical abuse, and 9 percent as sexual abuse. The highest rates of child abuse and neglect occur among young children, specifically those less than 3 years old.

Finding: Tribal child welfare systems, the Bureau of Indian Affairs, and the Indian Health Service do not report to NCANDS and are therefore not included in the datasets, thus limiting the ability to determine levels of abuse and neglect among many American Indian and Alaska Native populations. Moreover, non-Native workers report on cases of child abuse and neglect without familiarity with or consideration of the culture in these communities.

Finding: Difficulties arise in determining rates of child abuse and neglect. When researchers attempt to identify instances of child abuse and neglect through survey instruments, results can vary based on the

types of questions asked and the point in time and circumstances under which respondents provide the information. Conducting retrospective surveys of childhood experiences, asking children about recent experiences, and surveying parents about their behaviors toward children all can yield different results.

Finding: African American and American Indian children are referred to child protective services at disproportionate rates relative to their representation in the general population.

INCIDENCE TRENDS

Questions about whether child abuse and neglect are increasing, decreasing, or being detected and reported more often have become prominent in recent years. At the time of the 1993 NRC report, there was a general consensus that child abuse and neglect was underreported. Since that time, substantial changes have occurred in the social climate with regard to awareness of child abuse and neglect, attitudes toward reporting it, and the availability of programs and services for children and families affected by it. These developments have explicitly been intended to increase reporting of child abuse and neglect by victims, the general public, and professionals. However, establishing whether changes in official reporting represent true changes in incidence is complicated by the limitations of the reporting systems discussed above, as well as the difficulties inherent in ascertaining rates of events that happen to children, many of whom are very young, and that occur mainly in the private context of family life. As revealed by the review below, discrepancies exist in some areas and considerable ambiguity in others regarding the conclusions to be drawn from the available trend data, suggesting outstanding questions that would benefit from more systematic empirical analyses of these trends over time.

Sexual abuse has shown the largest decline in reported rates. NCANDS reports a decline of 62 percent since 1992 (Finkelhor and Jones, 2012). The sharpest declines occurred during the late 1990s, but the downward trajectory has continued, with a 3 percent decline being reported between 2009 and 2010. This same pattern is demonstrated in the NIS-4, issued in 2010, which reported a 47 percent decline from the mid-1990s through 2005, when the data for that report were collected (Finkelhor and Jones, 2012).

Additional information on trends in sexual abuse is derived from surveys of youth. The NCVS documents a 68 percent decrease in reported and unreported sexual assault or rape of 12- to 17-year-olds between 1993 and 2010 (White and Lauritsen, 2012). In a national survey on sexual and reproductive activity, young women (aged 15-24) reported a 39 percent decline in sexual experiences with a partner 3 or more years older before

the age of 15 (Finkelhor and Jones, 2012). This survey follows the same pattern as NCANDS, with the declines being steepest in the 1990s and tapering off although still continuing in the 2000s. Finkelhor and colleagues (2010b) compare results from the National Survey of Children Exposed to Violence (NatSCEV) in 2003 and 2008 and find that reports of sexual assault declined from 3.3 percent of all children aged 2-17 in 2003 to 2.0 percent of children in 2008. In contrast, the National Survey of Adolescents (NSA), a survey of a large nationally representative sample of youth, found no decline in self-reported sexual assault between 1995 and 2005 (Finkelhor and Jones, 2012).

The trend data are more ambiguous with respect to physical abuse. NCANDS reports a decline of 56 percent in physical abuse reports from the early 1990s through 2010 (Finkelhor et al., 2010a). The decrease for physical abuse began somewhat later than that for sexual abuse but has followed the same slope, with steep declines in the late 1990s that tapered off by 2009. Likewise, the NIS-4 reported a 29 percent drop in endangerment-standard physical abuse starting in the early 1990s (Finkelhor and Jones, 2012).

Survey results produce a somewhat different picture. The NCVS reports a 69 percent decline in aggravated physical assaults on children (aged 12-17) from 1993 through 2008; however, these events are mainly peer and sibling assaults rather than physical abuse by parents (Finkelhor and Jones, 2012). Zolotor and colleagues (2011) compared results from a 2002 survey of parents in North Carolina (Carolina Survey of Abuse in the Family Environment) using the Parent-Child Conflict Tactics Scale with the findings of a Gallup survey completed in 1995 and the results of two National Family Violence Surveys, conducted in 1975 and 1985, that used the same scale. The results show a decline in parental reports of physical abuse. On the other hand, neither the NatSCEV nor the NSA found significant declines in youth-reported physical abuse by caregivers (Finkelhor and Jones, 2012).

Another source of data on physical abuse is admissions to a hospital for abuse-related injury. Physical abuse encompasses a broad range of acts. The most common is striking a child such that bruising results—ranging from relatively minor, temporary, and localized marks caused by pinching or slapping to significant marks caused by whipping or hitting with an object that may leave scars. These types of injuries do not typically entail admission to a hospital or even require medical care. On the other hand, a relatively small percentage of physical abuse cases involve injuries, such as fractures, burns, blunt trauma, and abusive head trauma (formerly known as shaken baby syndrome), that require medical care and possibly hospitalization (Zolotor and Shanahan, 2011). Approximately 1.4 percent of physical abuse cases are estimated to result in hospitalization (Leventhal et al., 2012).

A number of studies have investigated changes in rates of admission for head injuries resulting from child physical abuse—the most common reason for child abuse-related hospital admission. Leventhal and Gaither (2012) found a small but concerning increase in the rate of serious injuries as documented in coding on medical records in a series of children’s hospitals (from 6.1 to 6.4/100,000) from 1997 to 2009. Additional studies, attempting to show an association between economic indicators and child abuse, similarly have found increases in rates of injuries coded as child abuse occurring during the 2000s (Berger and Waldfogel, 2011; Berger et al., 2011; Wood et al., 2012). A national study conducted in Taiwan also found a significant increase from 1996 to 2007, but only for infants and largely accounted for by changes in coding practices since 2003 (Chiang et al., 2012).

Neglect reports show the most mixed trends picture. NCANDS neglect reports declined by 10 percent between 1990 and 2010 (Finkelhor et al., 2010a), but there was significant variability across states. From 1992 to 2010, for example, fluctuations ranged from a 90 percent decline in neglect in Vermont to a 189 percent increase in Michigan. These dramatic state variations are not mirrored in the sexual and physical abuse rates, which declined across almost all states over the same period. The NIS-4 found no decline in neglect cases (Sedlak et al., 2010a). Self-report survey data are not available for neglect to permit comparisons over time. In part, this is due to the fact that retrospective self-report surveys are poorly suited to gathering information about neglect involving very young children, which is the most frequent form of child abuse and neglect.

Child maltreatment–related fatalities include deaths caused by both physical abuse and neglect, with a majority being attributed to neglect. NCANDS reports an increase of 46 percent in abuse- and neglect-related fatalities between 1993 and 2007 (Finkelhor and Jones, 2012). In contrast, homicide rates for children fell by 43 percent during the same period, with a 26 percent decline for the youngest children (aged 0-5) (Finkelhor and Jones, 2012); between 1980 and 2008, 63 percent of murdered children aged 0-5 were killed by a parent (Cooper and Smith, 2011). It is unclear to what extent cases officially reported by law enforcement as homicides correspond to cases included in the NCANDS child abuse and neglect dataset, most of which, as noted, are attributed to neglect.

Trends in child abuse and neglect occur in the larger context of rates of crime and violence in the United States. The consensus is that crime has decreased substantially, although there are some year-to-year fluctuations and pockets where these results are not seen. Both official reports as reflected in the UCR and population-based counts of reported and unreported crime as determined by the NCVS reveal declines in virtually all crime categories since the mid-1990s (FBI, 2010; Truman and Planty, 2012). These declines extend to sexual assault and domestic violence, crimes that share character-

istics of child sexual and physical abuse and often involve people in close interpersonal relationships or family members. As with child abuse and neglect, extensive efforts have been undertaken to change the social climate around these crimes, encourage reporting, and expand service availability. The NCVS shows a 68 percent decline in the number of children aged 17 and younger living in households in which someone aged 12 and older was the victim of sexual assault or violent crime between 1993 and 2010 (Truman and Smith, 2012).

In sum, trends are inconsistent across types of child abuse and neglect, and in the case of neglect are inconsistent across states. Sexual abuse reporting appears to indicate a clear decline that is not reflected in only a single data source. Although most sexual abuse is not committed by immediate family members, the declines here appear to extend equally to family and nonfamily sexual assaults. It is worth noting that the declines in child sexual abuse began about the same time as general declines in crime and have followed a similar slope. Physical abuse presents a more complicated picture, with some official sources showing overall declines and several surveys not showing declines. Although physical assaults in general (e.g., nonfamily assault, bullying) appear to be down, it is not clear that these trends extend to intrafamilial physical abuse.

Increases in child abuse-related hospital admissions are especially concerning because these data represent the most severe assaults, even though they make up a very small subgroup of child abuse cases. There are several possible explanations for these increases. First, they may represent actual increases in serious injury. Several studies have directly examined the correlation between the increases in identified cases and larger economic forces (Berger and Waldfogel, 2011; Berger et al., 2011; Wood et al., 2012). Berger and colleagues (2011) hypothesize an association between the economic recession and rising rates of child abuse-related injury, citing increases in child abuse and neglect reports from the prerecession to the recession period. However, they find no association with local unemployment rates. Wood and colleagues (2012) link data on child abuse-related hospital admissions to mortgage delinquency, foreclosures, and unemployment rates between 2000 and 2009. They find increases in admission rates to be correlated with mortgage foreclosure and delinquency rates but not with unemployment rates. Another possibility is that the increases reflect greater awareness and willingness of health care providers to label injuries as child abuse. The increases coincide with the advent of growing use of hospital diagnostic and billing codes that specify child abuse as the injury cause and a period when a child abuse subspecialty was created in pediatrics. These changes may have contributed to greater willingness to identify child abuse as the cause of injury in official records. Now that abusive head trauma is being captured more accurately in administrative data, it could potentially account

for a decline in other forms of head injury (Leventhal and Gaither, 2012). It is also possible that caregivers who inflict severe injuries have more severe psychopathology or are otherwise different from the typical child abuser, and are therefore less amenable to the influences associated with general societal changes and less likely to accept offers of voluntary assistance.

The lack of a significant decline in child neglect and the large jurisdictional variations in this area remain the least understood. The past two decades have seen a growing emphasis on encouraging recognition of neglect as its deleterious effects have increasingly been documented. Awareness campaigns have been undertaken to encourage reporting of neglect, and in some cases its definition has been expanded to incorporate a variety of risky circumstances and conditions. For example, the relationship of parental substance abuse to child abuse and neglect has received widespread attention. These forces may have contributed to increased reporting of a broader spectrum of neglect cases. Greater awareness and expanded definitions may have offset any declines in reports of traditionally defined neglect.

Poverty often is considered a major contributor to neglect, yet there is little empirical support for a strong relationship between changes in indicators of poverty and neglect reporting rates. For example, there was a great deal of concern that welfare reform, especially the timelines for receiving Temporary Assistance for Needy Families (TANF), would produce an increase in cases of neglect as parents were forced off income support. However, no significant change in neglect rates was seen during this period. And as mentioned, two separate investigations failed to find a relationship between unemployment rates and child abuse and neglect reports.

A better understanding is needed of whether and why rates of physical and sexual abuse are declining while no change in neglect is being observed. Criminologists have focused on understanding the substantial declines in crime rates as well as the occasional fluctuations or stubborn persistence of high crime rates in a few areas. Multiple commentators have examined possible causes and explanations (Finkelhor et al., 2010b; Levitt, 2004; Oppel, 2011; Zimring, 2008, 2011). Other fields, such as medicine, would certainly have devoted extensive scientific inquiry to understanding an epidemiological phenomenon as significant and inconsistent across different forms of the same problem area. Yet there has been no similar focus in the field of child abuse and neglect. Attention to the topic has been limited to a few investigators who have repeatedly reported on trends (e.g., Finkelhor and Jones, 2012) and to targeted examinations of specific subareas, such as hospital admissions (e.g., Chiang et al., 2012; Leventhal and Gaither, 2012). A greater focus on understanding the fluctuations in child abuse reporting data and other indicators of child injury both nationally and within specific communities and populations could have important implications for the design and targeting of intervention and prevention efforts.

Finding: Strong evidence indicates that sexual abuse has declined substantially in the past two decades; the balance of evidence favors a decline in physical abuse, especially its more common and less serious forms. There is no evidence that neglect is declining overall; however, states vary significantly as to whether neglect is increasing, decreasing, or remaining constant. These disparate trends have important implications for understanding the nature of child abuse and neglect and the forces that potentially affect its trends. Social policy endeavors are hampered when insufficient attention is paid to understanding the various aspects of the problem.

Finding: Understanding is incomplete with respect to whether and why rates of physical and sexual abuse are declining while no change in neglect is being observed. Research on these trends has received inadequate attention given their important implications for intervention and prevention efforts.

DETERMINATION OF CHILD ABUSE AND NEGLECT

This section reviews the various methods of determining whether child abuse and neglect has occurred. The basis for the determination can range from a citizen’s or family member’s simply believing what a child says about being abused or neglected or being convinced by something observed, to a medical examination and diagnosis or the formation of a professional opinion, to the results of administrative or legal procedures. The process for making a determination by medical and mental health professionals is established by professional standards of practice, whereas legal standards of investigative practice, rules of evidence, and burdens of proof govern how legal determinations are made.

Determination by Medical and Mental Health Professionals

Medical determination or diagnosis is relevant in a small but very high-stakes minority of child abuse and neglect cases. A medical opinion is the only way to determine whether certain physical injuries—especially very serious injuries such as head injuries, fractures, and burns—are the result of child abuse and neglect in children who are too young to provide a verbal account of how the injury occurred. In certain cases involving children old enough to say what happened, a medical opinion may be necessary to distinguish accidental from nonaccidental injuries when the children’s or parents’ accounts are discrepant. In some neglect cases, such as those entailing malnutrition or failure to thrive, a medical opinion may be an essential component of the investigative process.

Taking a medical history is standard practice when medical professionals conduct a medical examination. In situations involving child abuse and neglect, especially when sexual abuse is suspected or the cause of an injury is in dispute, the child’s history may be the primary basis for a medical professional’s opinion or diagnosis. In such cases, although medical professionals may have specialized expertise in interviewing children, they, like other professionals and ordinary citizens, have no special ability to distinguish true from false or mistaken statements. However, statements made to a health care provider may be admissible in legal proceedings as an exception to the hearsay rule.

Overall, within the child abuse medical subspecialty, substantial consensus exists regarding the diagnostic criteria for forming a medical opinion about whether injuries or medical conditions are attributable to child abuse and neglect. However, there have been high-profile controversies about medical opinions in some child abuse cases. For example, questions have been raised about certain medical diagnoses, such as shaken baby syndrome, which as noted, is now called abusive head trauma. In some cases, child abuse experts have concluded that intentional injury has occurred, but other medical professionals have attributed the injuries to causes such as brittle bones or vitamin deficiencies. In large part, such conflicting opinions are due to the adversarial nature of the U.S. legal system. Opposing experts provide testimony to contradict a child abuse and neglect allegation and opine that alternative medical explanations account for the injuries, often, it has been argued, invoking scientifically unsupported assertions (Chadwick et al., 1998). Although there have been some salient scientific developments in terms of the causes of injuries, in most cases these disputes do not reflect significant scientific uncertainties.

Outstanding questions do remain about the types of tests and procedures that are most appropriate for making a determination of child abuse and neglect. For example, radiographic skeletal survey is the standard procedure for detecting clinically unsuspected fractures in possible child abuse victims since a certain percentage of children will have occult fractures. Standards for additional tests and their timing have not been definitively established. Absent consensus standards, practice shows considerable variability.

Other presentations for which a medical opinion is absolutely necessary include complex conditions such as Munchausen syndrome by proxy, or medical child abuse (Davis and Sibert, 1996; Fisher and Mitchell, 1995; Roesler and Jenny, 2008). While this condition is very rare (0.5/100,000 children), the potential consequences to children are extreme and severe (McClure et al., 1996). Parents repeatedly take their children to medical providers, often many different ones, with reports of multiple and sometimes extremely serious symptoms or conditions. In some cases, the child

has or had a legitimate underlying condition, and the parents have extreme anxiety and repeatedly seek out additional tests and procedures or exaggerate symptoms. In other cases, parents fabricate or cause the medical symptoms to obtain psychological gratification from the attention they receive in the role of concerned parent. Making a determination of medical child abuse in these cases is fraught with complications and frequently cannot occur until the child has suffered significant harm or endured unnecessary tests, procedures, and even surgeries. Suspicion does not even arise until the pattern of visits, procedures, and contacts with multiple providers emerges. Child abuse doctors face a daunting task in challenging the opinions and practices of other medical providers who may have been mistaken, but genuinely believed the child had a serious medical condition.

In sexual abuse cases, although medical assessment is the standard of care, medical diagnosis is relevant in only a small subset of cases. Physical signs or symptoms, such as genital changes or injuries, sexually transmitted diseases, pregnancy, or the presence of seminal fluids or sperm, are present in only about 4 percent of cases; the vast majority of children medically evaluated for sexual abuse have normal exams (Heger et al., 2002). Even when there are genital findings, most are nonspecific and cannot be linked conclusively to sexual assault (Heger et al., 2002). Cases with definitive medical evidence, such as the presence of semen or pregnancy, are exceedingly rare. Standards for making a medical determination of sexual abuse have been published (Kellogg and Committee on Child Abuse and Neglect, 2005).

There are two important reasons beyond medical diagnosis why medical assessment of children who may have been or report being sexually abused is the standard of care. One purpose is to allay the child’s and parents’ worries about the potential physical effects of sexual contact. A visit with a medical provider creates a nonstigmatizing opportunity for support and validation, psychoeducation about the impact of sexual abuse, and encouragement to engage in available treatment services. The second is that citizens, judges, and juries assume that medical findings will be present in sexual abuse cases, even though this frequently is not the case. Child protection and criminal legal professionals believe it is often necessary to have a medical exam and expert medical testimony primarily to counter this widespread misconception.

Mental health professionals may be asked by parents or other professionals to provide a professional opinion as to whether a child was abused. Most such requests involve concern about sexual abuse. A diagnosis is not made because sexual abuse is an event, not a medical or psychiatric condition. In many cases, the mental health professional’s opinion is sought in a forensic context when a report has been made to authorities or a legal action has been initiated, and the opinion is expected to help guide legal

decision making or provide the basis for expert testimony in a legal proceeding. In other cases, however, the opinion is sought to determine whether to initiate reporting or other legal actions.

Typically in these situations, mental health providers consider a range of information, including what the child says in an interview, what the child has told others, the circumstances of the discovery of abuse concerns, results of medical examinations, and the emotional and behavioral functioning of the child based on a psychosocial assessment or administration of a standardized checklist of tests. The degree of thoroughness and the formality of the process depend largely on the purpose the opinion will serve.

Whereas child abuse mental health professionals do bring specialized expertise, knowledge, and skills to the evaluation process, there are scientific limits on the conclusions that can be drawn about whether an event occurred based on psychosocial assessment. No psychological profile has sufficient specificity to permit conclusions about an event as the cause of a presentation (APA, 2013). In addition, the emotional and behavioral consequences of child abuse and neglect are varied and nonspecific (see Chapter 4). Conditions typically associated with child abuse and neglect, such as posttraumatic stress, anxiety, depression, and behavioral problems, are common mental health problems for children and have many other causes. The only behavioral problem that has a specific and significant relationship with child abuse and neglect is inappropriate sexual behavior. However, the majority of sexually abused children do not have sexual behavior problems, and there are other potential causes for sexual behavior in children (Friedrich, 1993; Friedrich and Trane, 2002; Friedrich et al., 1998, 2003).

To a large extent, professional opinions on child abuse and neglect rely heavily on determinations about the credibility of children’s statements. There is no reason to believe that children cannot give reliable and accurate information about events or that they are prone to making false complaints about abuse (Brown et al., 2007; Cederborg et al., 2008; Lamb et al., 2007; Lyon, 1999). On the other hand, it is well established that memory, especially in young children, is susceptible to error and distortion, and that children can form false beliefs that they have experienced events (Cederborg et al., 2008; Lyon, 1999). It turns out that the characteristics of true and untrue statements have many commonalities; some true statements are not very credible, and some untrue statements are highly detailed and convincing. Mechanisms devised for rating child reports about abuse and neglect and classifying them as accurate or inaccurate have not proven reliable (Hershkowitz et al., 2007). In other words, professionals have no special ability to detect truthfulness, nor is there a scientifically reliable method for doing so. This is why courts generally do not permit professionals to

opine about the credibility of witnesses, but reserve that function for the fact finder (Myers, 2012).

Standards have been established for conducting forensic assessments for purposes of providing an opinion about possible sexual abuse (e.g., Kuehnle and Connell, 2009; Sparta and Koocher, 2006). The standards cover the assessment process, interviewing approaches, the proper use of psychosocial information, and limits on the accuracy of opinions based largely on statements that cannot be verified and behaviors that are nonspecific. Unfortunately, the types of cases for which such assessments are sought are those that are most ambiguous and complex, such as when children are unable or unwilling to give a clear and credible history, they are very young, they have not made statements, their statements are vague or inconsistent, or they suffer from emotional and behavioral problems that affect their credibility.

Mental health professionals routinely form opinions on the basic truth of reports about historical events that are potentially relevant in explaining why clients present with emotional and behavioral problems. Mental health providers commonly inquire about a range of past events, such as child abuse and neglect; other forms of trauma; events and experiences such as divorce, family moves, and experiences at school or with peers; illness and hospitalization; and other relevant life experiences. This information is integrated with information derived from clinical observation and the results of assessment measures with respect to symptoms and behaviors. Except for what providers observe directly in session, nearly all the information that serves as the basis for an opinion about events and mental health problems is derived from self-reports. Reliance on self-reports, including reports of child abuse and neglect, is therefore a cornerstone of standard clinical practice.

Determination by the Legal System

Legal Investigations

Before a child abuse and neglect case arrives before a legal fact finder (judge or jury), an arm of the government investigates the case. Child protective services and law enforcement conduct the investigations that serve as the basis for the state’s actions regarding dependency or prosecution. In many cases, the parents or defendants come to an agreement with the government, and no actual fact-finding hearing takes place. If it does, the official legal determination is made by civil or criminal court.

Child protective services usually is responsible for investigating civil dependency cases; such cases are screened in by the child welfare system, and they fall under the jurisdiction of the juvenile court. Given that the

greatest number of reported cases involve neglect, and most do not involve criminal conduct, the child protective services investigation is the only process applied to making a determination about child abuse and neglect in the majority of cases. Caseworkers make home visits and observe the safety and hygiene status of the household; inspect bruises and injuries; and conduct interviews with children (when appropriate), caregivers, reporters, and others who may have relevant information (such as relatives, teachers, and health care providers). They then draw conclusions about whether the information and evidence thus obtained meet the legal standards for child abuse and neglect.

Law enforcement officials investigate crimes. They generally engage in the same activities as child welfare system caseworkers (e.g., interviewing victims and witnesses, examining home conditions); they may also collect evidence from crime scenes, undertake forensic analyses, and interrogate suspects. In many jurisdictions, child protective services and law enforcement officials conduct joint investigations (Cross et al., 2005).

A key activity in many child abuse and neglect dependency and criminal investigations, especially in cases involving sexual abuse and some involving physical abuse, is interviewing the child. Interviewing methods most likely to lead to accurate and complete reports have been extensively investigated (e.g., Cronch et al., 2006; Lamb et al., 2009; Larsson and Lamb, 2009; Saywitz et al., 2002). The protocol of the National Institute for Child Health and Development (NICHD) is the approach that has been the most researched in real-life settings and in laboratory analogue experiments, and serves as the model for the current standard of practice (Lamb et al., 2007). Other extant models, none of which has undergone the same level of empirical evaluation, share almost all the same procedures and practices as the NICHD protocol (Anderson et al., 2010).

Legal Determinations

A legal determination of child abuse and neglect is based on the weighing of admissible evidence that is collected following the accepted procedures for the specific legal arena. The common law legal system in the United States is adversarial and is based on principles that protect the due process rights of those who are accused and risk loss of liberty, access to their children, or assets. The legal contexts vary by whether they are criminal or civil, the intended outcomes of the case, and the standard of proof that applies.

The two primary legal systems that make determinations about child abuse and neglect are the child protection system and the criminal justice system (Myers, 2012). The child protection system carries out an administrative and civil justice process that involves the state’s seeking to intervene

in families, often but not always to assume temporary custody of children (e.g., establishing child abuse or neglect and then obtaining authority of the court for the child’s placement) or in a small fraction of cases to terminate parental rights. In these court cases (often called dependency cases), the standard of proof typically is more probable than not; in a case involving termination of parental rights, a higher standard of clear and convincing evidence has been set by the U.S. Supreme Court. The goals of the criminal justice system are to hold lawbreakers accountable and punish them, to bring justice for victims, and to protect the community. The standard of proof here is the highest (beyond a reasonable doubt) because the case involves the government’s restricting an adult’s liberty, including the possibility of incarceration. Child abuse and neglect also may be addressed in family court custody matters when it is alleged by one parent seeking to restrict the other parent’s access to the child. In addition, civil tort actions may be brought in which a child, or someone on his/her behalf, sues a caregiver, the government, or another entity for negligence, seeking monetary damages.

The large majority of both civil and criminal proceedings regarding child abuse and neglect do not progress to a formal fact-finding hearing or a trial. In many child protection cases, usually those not requiring a court order to remove a child from home against parental wishes, no formal legal process is even initiated; the family agrees to a voluntary service plan that is overseen by the state. Even when a dependency petition is filed in court, in the large majority of cases the parent reaches an agreement or case settlement regarding dependency, often without admitting to having committed an act of child abuse and neglect. On the criminal side, charges are not filed in many cases, even when prosecutors may believe a crime occurred, because of difficulties entailed in proving the case and in meeting the legal standard of proof of beyond a reasonable doubt. In the majority of cases when charges are filed, the accused pleads guilty to the crime or to a lesser crime.

Substantiation

The child protection system’s classification of a child abuse and neglect case as substantiated is an administrative procedure for making a formal recorded determination about the validity of a child abuse and neglect report. In most states, the result of an investigation of a report is classified as substantiated or unsubstantiated, although some states use other terminology (e.g., founded/unfounded) to describe the investigative outcome. In 2011, approximately 19 percent of screened-in cases were substantiated, or “indicated” (ACF, 2012). Substantiation can be legally disputed because the consequences of a substantiated report can be significant for caregivers

(e.g., job loss or being barred from certain professions or by certain employers) (CWIG, 2013; McCarthy et al., 2005).

No formal conclusion about whether child abuse and neglect occurred is recorded in cases that are referred for an alternative response (sometimes called a family assessment or differential response) and not formally investigated (CWIG, 2013). In 2011, about 10 percent of all cases reported to NCANDS received an alternative response (ACF, 2012), but that percentage is increasing. As of 2011, 17 states were implementing differential response at some level, and 6 states planned to implement it in the near future.

Rates of substantiation vary dramatically across states (ACF, 2012), and there is little consensus on what accounts for this variation. Overall, every method used to determine the accuracy of child abuse and neglect allegations has weaknesses and cannot be considered definitive. To some extent, this does not matter as long as the victims are safe and receive needed services. For example, most crimes will not be reported or prosecuted or result in conviction of the perpetrator; however, crime victims will still have access to many services designed to help them recover from the effects of the crime, and most can take at least some steps toward protecting themselves from the perpetrator. Although child abuse victims are dependent on caregivers for future protection, many parents can and do take steps to protect their children from known perpetrators or correct their own neglectful or abusive behavior. In terms of access to needed services, what happens officially in a case is unrelated to receipt of services in the child welfare system. The National Survey of Child and Adolescent Well-Being, a large longitudinal study of a nationally representative sample of cases reported to child protective services, produced illustrative results. Comparisons of cases that were closed or kept open, or were substantiated or not, revealed no difference in key variables related to services or outcomes (Hussey et al., 2005; Kohl et al., 2009).

The difficulty of ascertaining the validity of cases using official reporting or procedural outcomes may have more of an effect on research and interpretation of findings than on the lives of children who enter the child welfare system. For example, Kohl and colleagues (2009) argue that if substantiation does not discriminate true from untrue cases of child abuse and neglect, it is not a meaningful or accurate way of learning about the characteristics of actual abuse and neglect and its relationships to outcomes since the comparison group of unsubstantiated cases will contain many true cases. Therefore, child abuse research may benefit if consensus is achieved not only on definitions, but also on the meaning of different classification mechanisms for child abuse and neglect reports.

Finding: Significant advances have been made in dealing with children who may have been abused and neglected when they come in contact

with medical, mental health, or social services professionals. It has become more common for these professionals to screen children routinely for abuse and other traumatic experiences. The children’s accounts are generally accepted, at least for purposes of meeting the “reasonable suspicion” standard for making a child abuse report, except when there is significant evidence for coercion or contamination of their statements. Children who are suspected of being abused are commonly referred for specialized assessment, as well as clinical and support services.

Finding: Overall, substantial improvements have been achieved in the assessment and investigative procedures for determining whether child abuse and neglect has occurred since the 1993 NRC report was issued. Widely accepted standards for proper interviewing have been adopted by child protective services, law enforcement officials, and forensic evaluators and are well known even among general health, mental health, and other professionals (Lamb et al., 2007).

Finding: Rates of substantiation of child abuse and neglect allegations by child protective services vary dramatically across states, and there is little consensus on what accounts for this variation. Overall, every method of determining the accuracy of child abuse and neglect allegations has weaknesses, and no method can be considered definitive. This limits the substantiation classification as a meaningful way to learn about the characteristics of actual abuse and neglect and their relationships to outcomes.

CONCLUSIONS

Child abuse and neglect is a pervasive societal problem, with recent NCANDS data indicating that 3.4 million child abuse and neglect referrals involving 6.2 million children were made in a single year across the United States and its territories. As will be discussed in Chapter 4, these incidents of child abuse and neglect entail a substantial risk for deleterious consequences that can hinder child development and lead to problems that persist across the life course.

Cases of child abuse and neglect are referred to child protective services based on mandatory reports by professionals such as teachers, law enforcement officials, social service providers, and physicians, as well as good-faith reports by citizens. Not all cases of child abuse and neglect are reported, and standards for reasonable suspicion of abuse and neglect are not always clear-cut. Therefore, official reports do not capture all cases in which child abuse and neglect is suspected or even is detected and acted upon. For research purposes, then, sole reliance on referral data from child protective services

cannot capture the full scope of child abuse and neglect. Incorporating data from additional sources is necessary to determine the true incidence of the problem.

In addition, child abuse and neglect are defined differently for the varying purposes for which related information is collected, confounding attempts to portray the scope of the problem accurately or examine the surrounding circumstances. Results across studies based on surveys also may vary according to the survey methodology employed. Movement toward a reasonable degree of standardization in these areas is therefore needed.

Difficulties in ascertaining the scope of child abuse and neglect have contributed to uncertainties regarding whether the incidence of the problem is increasing or decreasing or cases are being detected and reported more frequently. Available trend data provide strong evidence that sexual abuse has declined substantially in the past two decades; the balance of evidence favors a decline in physical abuse, especially its more common and less serious forms. There is no evidence that neglect is declining overall. However, states vary significantly as to whether neglect is increasing, decreasing, or remaining constant. Discrepancies and ambiguity across analyses of different data sources highlight a need for more systematic empirical analyses of these trends over time. Research is needed to learn more about trends in child abuse and neglect and the variables that may account for decreases in the incidence of the problem or the lack thereof.

REFERENCES

ACF (Administration for Children and Families). 2012. Child maltreatment, 2011 report. Washington, DC: U.S. Department of Health and Human Services, ACF.

Alvarez, K. M., M. C. Kenny, B. Donohue, and K. M. Carpin. 2004. Why are professionals failing to initiate mandated reports of child maltreatment, and are there any empirically based training programs to assist professionals in the reporting process? Aggression and Violent Behavior 9(5):563-578.

Anderson, J., J. Ellefson, J. Lashley, A. Lukas Miller, S. Olinger, A. Russell, J. Stauffer, and J. Weigman. 2010. Cornerhouse forensic interviewing protocol: RATAC. Thomas M. Cooley Journal of Practical and Clinical Law 12:193-331.

APA (American Psychiatric Association). 2013. DSM-5 diagnostic and statistical manual of mental disorders, 5th ed. Arlington, VA: APA.

Barnett, D., J. T. Manly, and D. Cicchetti. 1993. Defining child maltreatment: The interface between policy and research. Child Abuse, Child Development, and Social Policy 8:7-73.

Berger, L. M., and J. Waldfogel. 2011. Economic determinants and consequences of child maltreatment. OECD social, employment and migration working papers, no. 111. Paris, France: OECD Publishing.

Berger, R. P., J. B. Fromkin, H. Stutz, K. Makoroff, P. V. Scribano, K. Feldman, L. C. Tu, and A. Fabio. 2011. Abusive head trauma during a time of increased unemployment: A multicenter analysis. Pediatrics 128(4):637-643.

Brown, D. A., M. E. Pipe, C. Lewis, M. E. Lamb, and Y. Orbach. 2007. Supportive or suggestive: Do human figure drawings help 5- to 7-year old children to report touch? Journal of Consulting and Clinical Psychology 75(1):33-42.

CDC (Centers for Disease Control and Prevention). 2003. 2000 and 2001 population estimates for calculating vital rates. http://www.cdc.gov/nchs/about/major/dvs/popbridge/popbridge.htm (accessed November 14, 2013).

Cederborg, A. C., D. La Rooy, and M. E. Lamb. 2008. Repeated interviews with children who have intellectual disabilities. Journal of Applied Research in Intellectual Disabilities 21(2):103-113.

Chadwick, D. L., R. H. Kirschner, R. M. Reece, L. R. Ricci, R. Alexander, M. Amaya, J. A. Bays, K. Bechtel, R. Beltran-Coker, C. D. Berkowitz, S. D. Blatt, A. S. Botash, J. Brown, M. Carrasco, C. Christian, P. Clyne, D. L. Coury, J. Crawford, N. Cunningham, M. D. DeBellis, C. Derauf, J. de Triquet, B. P. Dreyer, H. Dubowitz, K. W. Feldman, M. A. Finkel, E. G. Flaherty, L. Frasier, L. Gari, J. Glick, P. Grant, G. Fortin, S. Halpert, R. A. Hicks, D. Huyer, C. Jenny, M. Joffe, S. W. Kairys, K. M. Kaplan, M. Kaufhold, K. J. Kemper, E. J. Krane, H. Krous, M. Lorand, J. McCann, M. Mian, K. Moran, L. M. Osborn, V. Palusci, M. A. Radkowski, M. E. Rimsza, D. Runyan, M. Ryan, M. D. Sadof, C. Schubert, R. Sege, R. A. Shapiro, B. Siegel, A. Sirotnak, W. Smith, R. Socolar, D. Soter, S. P. Starling, C. Stashwick, R. D. Steiner, J. Stirling, N. Sugar, T. Truman, D. Turkewitz, C. Wang, J. M. Whitworth, and J. A. Zenel. 1998. Shaken baby syndrome—a forensic pediatric response. Pediatrics 101(2):321-323.

Chiang, W. L., Y. T. Huang, J. Y. Feng, and T. H. Lu. 2012. Incidence of hospitalization due to child maltreatment in Taiwan, 1996-2007: A nationwide population-based study. Child Abuse & Neglect 36(2):135-141.

Child Trends. 2013. Child maltreatment: Indicators on children and youth. Bethesda, MD: Child Trends DataBank.

Children’s Bureau. 1995-2011. Child maltreatment. http://www.acf.hhs.gov/programs/cb/research-data-technology/statistics-research/child-maltreatment (accessed December 3, 2013).

Cooper, A., and E. L. Smith. 2011. Homicide trends in the United States, 1980-2008. Washington, DC: U.S. Department of Justice.

Cronch, L. E., J. L. Viljoen, and D. J. Hansen. 2006. Forensic interviewing in child sexual abuse cases: Current techniques and future directions. Aggression and Violent Behavior 11(3):195-207.

Cross, T. L., and D. Simmons. 2008. Child abuse and neglect and American Indians. Overview and policy briefing. Portland, OR: National Indian Child Welfare Association.

Cross, T. P., D. Finkelhor, and R. Ormrod. 2005. Police involvement in child protective services investigations: Literature review and secondary data analysis. Child Maltreatment 10(3):224-244.

CWIG (Child Welfare Information Gateway). 2011. Definitions of child abuse and neglect. Washington, DC: U.S. Department of Health and Human Services, Children’s Bureau.

CWIG. 2012a. Mandatory reporters of child abuse and neglect. Washington, DC: U.S. Department of Health and Human Services, Children’s Bureau.

CWIG. 2012b. Tribal-state relations. Washington, DC: U.S. Department of Health and Human Services, Children’s Bureau.

CWIG. 2013. How the child welfare system works. Washington, DC: U.S. Department of Health and Human Services, Children’s Bureau.

Davis, P. M., and J. R. Sibert. 1996. Munchausen syndrome by proxy or factitious illness spectrum disorder of childhood. Archives of Disease in Childhood 74(3):274-275.

Dettlaff, A. J., and M. A. Johnson. 2011. Child maltreatment dynamics among immigrant and U.S. born Latino children: Findings from the National Survey of Child and Adolescent Well-Being (NSCAW). Children and Youth Services Review 33(6):936-944.

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.

Earle, K., and A. C. Cross. 2001. Child abuse and neglect among American Indian/Alaska Native children: An analysis of existing data. Seattle, WA: Casey Family Programs and National Indian Child Welfare Association.

Edleson, J. L., J. Gassman-Pines, and M. B. Hill. 2006. Defining child exposure to domestic violence as neglect: Minnesota’s difficult experience. Social Work 51(2):167-174.

English, D. J., R. Thompson, J. C. Graham, and E. C. Briggs. 2005. Toward a definition of neglect in young children. Child Maltreatment 10(2):190-206.

Everson, M. D., J. B. Smith, J. M. Hussey, D. English, A. J. Litrownik, H. Dubowitz, R. Thompson, E. D. Knight, and D. K. Runyan. 2008. Concordance between adolescent reports of childhood abuse and child protective service determinations in an at-risk sample of young adolescents. Child Maltreatment 13(1):14-26.

FBI (Federal Bureau of Investigation). 2010. Table 1: Crime in the United States by volume and rate per 100,000 inhabitants, 1991-2010. Washington, DC: U.S. Department of Justice.

Finkelhor, D. 1994. Current information on the scope and nature of child sexual abuse. The Future of Children 4(2):31-53.

Finkelhor, D. 2009. The prevention of childhood sexual abuse. Future Child 19(2):169-194.

Finkelhor, D., and J. Dziuba-Leatherman. 1994. Children as victims of violence: A national survey. Pediatrics 94(4):413-420.

Finkelhor, D., and L. Jones. 2012. Have sexual abuse and physical abuse declined since the 1990s? Durham: University of New Hampshire, Crimes Against Children Research Center.

Finkelhor, D., L. Jones, and A. Shattuck. 2010a. Updated trends in child maltreatment, 2010. Durham: University of New Hampshire, Crimes Against Children Research Center.

Finkelhor, D., H. Turner, R. Ormrod, and S. L. Hamby. 2010b. Trends in childhood violence and abuse exposure: Evidence from 2 national surveys. Archives of Pediatrics & Adolescent Medicine 164(3):238-242.

Fisher, G. C., and I. Mitchell. 1995. Is Munchausen-syndrome by proxy really a syndrome? Archives of Disease in Childhood 72(6):530-534.

Fox, K. 2004. Are they really neglected? A look at worker perceptions of neglect through the eyes of a national data system. First Peoples Child and Family Review 1(1):73-82.

Friedrich, W. N. 1993. Sexual victimization and sexual-behavior in children: A review of recent literature. Child Abuse & Neglect 17(1):59-66.

Friedrich, W. N., and S. T. Trane. 2002. Sexual behavior in children across multiple settings—commentary. Child Abuse & Neglect 26(3):243-245.

Friedrich, W. N., J. Fisher, D. Broughton, M. Houston, and C. R. Shafran. 1998. Normative sexual behavior in children: A contemporary sample. Pediatrics 101(4):E9.

Friedrich, W. N., W. H. Davies, E. Feher, and J. Wright. 2003. Sexual behavior problems in preteen children—developmental, ecological, and behavioral correlates. Sexually Coercive Behavior: Understanding and Management 989:95-104.

Gainsborough, J. F. 2010. Scandalous politics: Child welfare policy in the states (American governance and public policy series). Washington, DC: Georgetown University Press.

GAO (Government Accountability Office). 2011. Child maltreatment: Strengthening national data on child fatalities could aid in prevention. Washington, DC: GAO.

Heger, A., L. Ticson, O. Velasquez, and R. Bernier. 2002. Children referred for possible sexual abuse: Medical findings in 2,384 children. Child Abuse & Neglect 26(6-7):645-659.

Hershkowitz, I., S. Fisher, M. E. Lamb, and D. Horowitz. 2007. Improving credibility assessment in child sexual abuse allegations: The role of the NICHD investigative interview protocol. Child Abuse & Neglect 31(2):99-110.

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):479-452.

Kellogg, N., and Committee on Child Abuse and Neglect. 2005. The evaluation of sexual abuse in children. Pediatrics 116(2):506-512.

Khan, A., D. H. Rubin, and G. Winnik. 2005. Evaluation of the mandatory child abuse course for physicians: Do we need to repeat it? Public Health 119(7):626-631.

Kilpatrick, D. G., and B. E. Saunders. 1995. National survey of adolescents in the United States, 1995. ICPSR 2833. Washington, DC: U.S. Department of Justice.

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

Kolko, D. J., E. J. Brown, and L. Berliner. 2002. Children’s perceptions of their abusive experience: Measurement and preliminary findings. Child Maltreatment 7(1):41-53.

Kuehnle, K., and M. Connell. 2009. The evaluation of child sexual abuse allegations: A comprehensive guide to assessment and testimony. Hoboken, NJ: John Wiley & Sons, Inc.

Lamb, M. E., Y. Orbach, I. Hershkowitz, P. W. Esplin, and D. Horowitz. 2007. A structured forensic interview protocol improves the quality and informativeness of investigative interviews with children: A review of research using the NICHD investigative interview protocol. Child Abuse & Neglect 31(11-12):1201-1231.

Lamb, M. E., Y. Orbach, K. J. Sternberg, J. Aldridge, S. Pearson, H. L. Stewart, P. W. Esplin, and L. Bowler. 2009. Use of a structured investigative protocol enhances the quality of investigative interviews with alleged victims of child sexual abuse in Britain. Applied Cognitive Psychology 23(4):449-467.

Langton, L., M. Berzofsky, C. P. Krebs, and H. Smiley-McDonald. 2012. Victimizations not reported to the police, 2006-2010. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.

Larsson, A. S., and M. E. Lamb. 2009. Making the most of information-gathering interviews with children. Infant and Child Development 18(1):1-16.

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. Atlanta, GA: CDC, National Centers for Injury Prevention and Control.

Leventhal, J. M., and J. R. Gaither. 2012. Incidence of serious injuries due to physical abuse in the United States: 1997-2009. Pediatrics 130(5):1-6.

Leventhal, J. M., K. D. Martin, and J. R. Gaither. 2012. Using us data to estimate the incidence of serious physical abuse in children. Pediatrics 129(3):458-464.

Levi, B. H., and K. Crowell. 2011. Child abuse experts disagree about the threshold for mandated reporting. Clinical Pediatrics 50(4):321-329.

Levitt, S. D. 2004. Understanding why crime fell in the 1990’s: For factors that explain the decline in six that do not. Journal of Economic Perspectives 18(1):163-190.

Lyon, T. D. 1999. The new wave in children’s suggestibility research: A critique. Cornell Law Review 84(4):1004-1087.

MacMillan, H. L., E. Jamieson, and C. A. Walsh. 2003. Reported contact with child protection services among those reporting child physical and sexual abuse: Results from a community survey. Child Abuse and Neglect 27(12):1397-1408.

McCarthy, J., A. Marshall, J. Collins, G. Arganza, K. Deserly, and J. Milon. 2005. A family’s guide to the child welfare system. Washington, DC: National Technical Assistance Center for Children’s Mental Health.

McClure, R. J., P. M. Davis, S. R. Meadow, and J. R. Sibert. 1996. Epidemiology of Munchausen syndrome by proxy, non-accidental poisoning, and non-accidental suffocation. Archives of Disease in Childhood 75(1):57-61.

McElroy, R. 2012. Analysis of state laws regarding mandated reporting of child maltreatment with appendix. Washington, DC: State Policy Advocacy and Reform Center.

Military OneSource. n.d. Legislation. http://www.militaryonesource.mil/abuse/service-providers?content_id=267333 (accessed July 15, 2013).

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

Myers, J. E. B. 2012. “Nobody’s perfect”—partial disagreement with Herman, Faust, Bridges, and Ahern. Journal of Child Sexual Abuse 21(2):203-209.

NCSL (National Council of State Legislatures). 2012. Mandatory reporting of child abuse and neglect: 2012 introduced state legislation. http://www.ncsl.org/issues-research/human-services/2012-child-abuse-mandatory-reporting-bills.aspx (accessed February 11, 2013).

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

Oppel, R. A. 2011. Steady decline in major crime baffles experts. New York Times, May 23.

OPRE (Office of Planning Research and Evaluation). 2009. National Incidence Study of Child Abuse and Neglect (NIS-4), 2004-2009. http://www.acf.hhs.gov/programs/opre/research/project/national-incidence-study-of-child-abuse-and-neglect-nis-4-2004-2009 (accessed April 22, 2013).

Putnam-Hornstein, E. 2011. Report of maltreatment as a risk factor for injury death: A prospective birth cohort study. Child Maltreatment 16(3):163-174.

Putnam-Hornstein, E. 2012. Preventable injury deaths: A population-based proxy of child maltreatment risk in California. Public Health Reports 127(2):163-172.

Putnam-Hornstein, E., B. Needell, B. King, and M. Johnson-Motoyama. 2013. Racial and ethnic disparities: A population-based examination of risk factors for involvement with child protective services. Child Abuse & Neglect 37(1):33-46.

Roesler, T. A., and C. Jenny. 2008. Medical child abuse: Beyond Munchausen syndrome by proxy. Elk Grove, IL: AAP Press.

Saywitz, K. J., G. S. Goodman, and T. D. Lyon. 2002. Interviewing children in and out of court. In The ASPAC handbook on child maltreatment, edited by J. E. Myers, L. Berliner, J. Briere, C. T. Hendrix, C. Jenny, and T. A. Reid. Thousand Oaks, CA: Sage Publications, Inc. Pp. 349-378.

Sedlak, A. J., J. Mettenburg, M. Basena, I. Petta, K. McPherson, A. Greene, and S. Li. 2010a. Fourth National Incidence Study of Children Abuse and Neglect (NIS-4): Report to Congress. Washington, DC: U.S. Department of Health and Human Services, ACF.

Sedlak, A. J., K. McPeherson, and B. Das. 2010b. Supplementary analyses of race differences in child maltreatment rates in the NIS-4. Rockville, MD: Westat, Inc.

Slack, K. S., J. Holl, L. Altenbernd, M. McDaniel, and A. B. Stevens. 2003. Improving the measurement of child neglect for survey research: Issues and recommendations. Child Maltreatment 8(2):98-111.

Sparta, S. N., and G. P. Koocher, eds. 2006. Forensic mental health assessment of children and adolescents. New York: Oxford University Press.

Steinberg, K. L., M. Levine, and H. J. Doueck. 1997. Effects of legally mandated child-abuse reports on the therapeutic relationship: A survey of psychotherapists. American Journal of Orthopsychiatry 67(1):112-122.

Stoltenborgh, M., M. H. van Ijzendoorn, E. M. Euser, and M. J. Bakermans-Kranenburg. 2011. A global perspective on child sexual abuse: Meta-analysis of prevalence around the world. Child Maltreatment 16(2):79-101.

Straus, M. A., and J. H. Stewart. 1999. Corporal punishment by American parents: National data on prevalence, chronicity, severity, and duration, in relation to child and family characteristics. Clinical Child and Family Psychology Review 2(2):55-70.

Straus, M. A., S. L. Hamby, D. Finkelhor, D. W. Moore, and D. Runyan. 1998. Identification of child maltreatment with the parent-child conflict tactics scales: Development and psychometric data for a national sample of American parents. Child Abuse & Neglect: The International Journal 22(4):249-270.

Theodore, A. D., and D. K. Runyan. 2006. A survey of pediatricians’ attitudes and experiences with court in cases of child maltreatment. Child Abuse & Neglect 30(12):1353-1363.

Theodore, A. D., J. J. Chang, D. K. Runyan, W. M. Hunter, S. I. Bangdiwala, and R. Agans. 2005. Epidemiologic features of the physical and sexual maltreatment of children in the Carolinas. Pediatrics 115(3):e331-e337.

Truman, J. L., and M. Planty. 2012. Criminal victimization, 2011. Washington, DC: Bureau of Justice Statistics.

Truman, J. L., and E. L. Smith. 2012. Prevalence of violent crime among households with children, 1993-2010. Washington, DC: Bureau of Justice Statistics.

White, N., and J. L. Lauritsen. 2012. Violent crime agains youth, 1994-2010. Washington, DC: U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.

Wood, J. N., S. P. Medina, C. Feudtner, X. Luan, R. Localio, E. S. Fieldston, and D. M. Rubin. 2012. Local macroeconomic trends and hospital admissions for child abuse, 2000-2009. Pediatrics 130(2):e358-e364.

Zimring, F. E. 2008. The great American crime decline. New York: Oxford University Press.

Zimring, F. E. 2011. How New York beat crime. Scientific American 305:74-79.

Zolotor, A. J., and M. Shanahan. 2011. Epidemiology of physical abuse. In Child abuse and neglect: Diagnosis, treatment and evidence, edited by C. Jenny. St. Louis, MO: Saunders. Pp. 10-15.

Zolotor, A. J., A. D. Theodore, D. K. Runyan, J. J. Chang, and A. L. Laskey. 2011. Corporal punishment and physical abuse: Population-based trends for three-to-11-year-old children in the United States. Child Abuse Review 20(1):57-66.

This page intentionally left blank.