Child abuse and neglect, also known as child maltreatment (Cronholm & Witherspoon, 2016), are a national and international public health epidemic with devastating consequences (Horner, 2013; Jackson, Kissoon, & Greene, 2015; Salvagni & Wagner, 2006). If the abuse remains unidentified and untreated, the long-term consequences of abuse and neglect can carry into adulthood (Chen et al., 2010; Felitti et al., 1998; Wilson, 2010). Child abuse often occurs in more than one form (Child Welfare Information Gateway, 2013; Finkelhor, Ormrod, & Turner, 2007; Trickett, Mennen, Kim, & Sang, 2009) and is defined as "polyabuse."
Cronholm and Witherspoon (2016) categorize child abuse and neglect as acts of commission (abuse) and omission (neglect). Abuse and neglect prevention is the optimal goal, but given the current challenge of achieving this goal, it is imperative that all healthcare providers working with children be equipped to identify, halt, and intervene when signs of abuse are observed. Comprehensive screening is necessary to detect physical, sexual, psychological, and polyabuse and neglect (Borg, Snowdon, & Hodes, 2014; Centers for Disease Control and Prevention, 2014; Chen et al., 2010; Estroff, Foglia, & Fuchs, 2015; Felitti et al., 1998; Hornor, 2014, 2015; Wilson, 2010).
The World Health Organization (2001) estimates that as many as 40 million children are abused in any given year. In 2014, the number of U.S. child abuse victims was estimated at 702,000, and a probable 1,580 U.S. children died from abuse and neglect (Administration for Children and Families, Administration on Children, Youth and Families, Children's Bureau, U.S. Department of Health & Human Services [USDHHS], 2016). This estimate is considered low due to underreporting and injury deaths falsely classified as accidental or not recognized as neglect related (Child Welfare Information Gateway, 2016). The consequences of child abuse and neglect are replete in the literature, but well-defined and effective methods for prevention and consistent detection of all but the most severe forms of child abuse and neglect are elusive (Bailhache, Leroy, Pillet, & Salmi, 2013; Woodman et al., 2010). As a society, we have the responsibility of safeguarding our children who are the most vulnerable within the population. To effectively do this, healthcare providers at every point of care must effectively identify and intervene on behalf of children who deserve a life free from abuse.
The Child Welfare Information Gateway (2016) reports that 3.2 million U.S. children were the subject of at least one protective service report in 2014. In addition, an estimated 702,000 children were victims of abuse and neglect, with the greatest number of children experiencing neglect or physical abuse. Children in the first year of life had the highest rate of victimization. Almost 75% of these fatalities were children under 3 years old. Boys had a higher fatality rate than girls, and nearly 90% of fatalities of children were of White, African American, and Hispanic backgrounds. Most (79%) of fatalities involved the action of at least one parent. These numbers do not reflect the deaths caused by other means related to child abuse, nor do they include deaths outside the United States. Although the United States does not have the highest rate of child abuse worldwide, it does have the third highest annual number of deaths due to child maltreatment for children under the age of 15 years out of 27 industrialized nations (behind Portugal and Mexico; UNICEF, 2003).
Experts believe that the number of reported deaths attributed to abuse and neglect is lower than the actual number, as cases go unreported, unrecognized, and unsubstantiated (Flaherty & Sege, 2005; Tiyyagura, Gawel, Koziel, Asnes, & Bechtel, 2015). According to the Centers for Disease Control and Prevention (2014), the financial cost of child abuse in the United States topped 124 billion dollars in 2008, rivaling the cost of other major health problems such as Type II diabetes and stroke in the adult population. These costs are only the current costs and do not take into account the individual lifelong problems associated with child abuse.
Children who endure repeated abuse and neglect may experience attachment disorder, cognitive challenges, social difficulties, and compromised mental health, and are more prone to substance abuse with increased negative contact with law enforcement and the judicial system (Administration for Children and Families, Administration on Children, Youth and Families, Children's Bureau, USDHHS, 2016). These children are also more likely to experience permanent physical impairment, decreased brain development and participate in risky sexual behaviors (Chen et al., 2010; Administration for Children and Families, Administration on Children, Youth and Families, Children's Bureau, USDHHS, 2016; Wilson, 2010). The emotional, societal, and monetary costs of the adverse physical, psychological, behavioral, and social effects of child abuse and neglect do not end with a damaged childhood but can be carried into adulthood.
In 1998, Felitti et al. published their findings from the Adverse Childhood Experiences (ACE) Study. The long-term effects of child abuse/child maltreatment and the consequences that reach into adulthood became undeniable. Adverse childhood experiences include psychological, physical, and sexual abuse; household dysfunction, categorized as substance abuse; mental illness; witnessing their mother treated violently; and/or having an imprisoned family member (Felitti et al., 1998). A strong relationship between childhood abuse and the leading health conditions that contribute to early morbidity and mortality in adulthood, such as heart, lung, and liver disease and cancer, was established. Risk factors associated with these health conditions include smoking, depressed mood, suicide attempts, physical inactivity, alcoholism, drug abuse, parenteral drug use, severe obesity, and multiple sexual partners with a history of sexually transmitted diseases (Felitti et al., 1998). The ACE score correlates with the number of adverse childhood events-the higher the score, the higher the potential for morbidity and mortality. The ACE study linked child abuse to a long-term burden for the victim with incalculable costs to society (Felitti et al., 1998).
Despite the known problems that child abuse and neglect cause, screening for it remains sporadic, with most screening conducted in emergency settings. The U.S. Preventive Services Task Force (2016) is an independent panel of experts in primary care and prevention that systematically reviews the evidence for effectiveness and develops recommendations for clinical preventive services. In 2013, the task force found that, although child abuse is a very serious problem, there is insufficient evidence to determine how primary care practices can prevent abuse or neglect and therefore assigned a rating for screening children for abuse at "I" for inconclusive. The burden is on healthcare professionals to identify the signs and symptoms of child abuse. Healthcare providers must be educated to recognize the signs and be prepared to report abuse at every point of delivery of care, not only in emergency settings. Identification of signs and symptoms of child abuse must be comprehensive to include all types of abuse, physical, sexual, psychological, polyabuse, and neglect (Kodner & Wetherton, 2013).
Review of the Literature
The purpose of this literature review was to identify screening tools currently available to screen for abuse and neglect. The literature was searched for articles that describe a screening instrument for healthcare providers to screen children (0-17 years old) for abuse and/or neglect at the point of care. The effectiveness and limitations of screening instruments for physical, sexual, and psychological abuse and neglect are reviewed. For the purpose of this article, screening is defined as a screening instrument (tool) implemented to identify indicators of child maltreatment.
Methodology
Multiple sources were searched using the search terms "child abuse screening," "child maltreatment screening," "child abuse screening instruments," and "child abuse detection." Articles included in this review focused on a specific child abuse screening instrument for the identification of child abuse in individual children, were limited to those in the English language, and were published between 1995 and 2016. The Cochrane Register of Controlled Trials was searched from 1995 to 2016. The Cochrane Database of Systematic Reviews was also searched from 1995 to 2016. Psych INFO, PubMed, and CINAHL were searched using variations of the key terms. In total, nine abuse screening instruments were identified in the literature search between 1995 and 2016. Of these instruments, five screen for physical abuse, two screen for sexual abuse, one screens for neglect, and one screens for polyabuse (see Table 1).
Physical Abuse Screening
Higginbotham et al. (2014) implemented a screening guideline in the emergency setting, for children younger than 12 months presenting with skeletal fracture unrelated to motor vehicle accidents. The tool is in the form of an algorithm that includes a head-to-toe assessment, a skeletal survey, urinalysis, and transaminase evaluation. The identification of physical abuse is based on the findings of these diagnostic screenings. The tool was effective in identifying young children with skeletal fracture related to physical abuse and eliminated disparities related to racial or social economic background (Higginbotham et al., 2014). Although this screening algorithm evaluates suspected nonaccidental trauma, it has not been empirically tested.
The work of Louwers et al. (2011, 2012, 2014) culminated in development of the "Escape Form," an empirically established screening protocol for the assessment of physical abuse and neglect in all children presenting for care in the emergency setting. The "Escape Form" assesses six domains: consistent history, unnecessary delay in seeking medical help, consistency of injury with child's developmental stage, appropriateness of childhood behavior and interaction with caregiver, head-to-toe findings consistent with the history, and any concerning signals that cause doubt about the child's safety (Louwers et al., 2014).
Louwers et al. (2014) assessed the reliability and validity of the "Escape Form." The sensitivity of the "Escape Form" was 0.80 (44/55) with 1 item positive; specificity was 0.98 (17,844/18,220). The positive likelihood of the "Escape Form" was 40 with a negative likelihood ratio of 0.20 (Louwers et al., 2014). The number of children identified in the study was 0.3% of the screened children and 0.1% of the unscreened children (those whose abuse was overtly apparent). The authors acknowledged that part of the reason for this comparatively low identification rate is that the "Escape Form" primarily identifies children with physical abuse and the 3% of abused children in the general population have experienced other abuse forms (Louwers et al., 2014).
Benger and Pearce (2002) developed a screening tool designed as a flowchart to assess intentional injury in children 0-5 years old who presented to the emergency setting with an injury. Four screening domains include delay in seeking treatment, consistent history over time, other unexplained injuries noted on physical examination, and whether the child's behavior and interactions with the caregiver are appropriate. Any positive finding warrants further assessment for abuse. An increase in the number of screenings was observed when the screening flowchart was present and flagged in the chart. No psychometric testing is provided for the screening flowchart.
Sittig et al. (2011) implemented a tool to screen for physical abuse. Used by emergency department staff, it assesses for abuse in children presenting to the emergency setting with physical injury. The SPUTOVAMO-R screening tool evaluates five domains: injury is compatible with the history and age of the child, history is consistent when repeated, delay in arriving at the emergency setting without a satisfactory explanation, suspicion after a head-to-toe examination, other unexplained injury in the history, and appropriate interaction between child and parents (Hoytema van Konijnenburg, Teeuw, Zwaard, van der Lee, & van Rijn, 2014; Sittig et al., 2011). The screening is considered positive when one or more responses are positive. All positive screenings are referred for follow-up and review by the multidisciplinary child abuse team. All information gathered for each child was presented to an expert panel of child abuse experts, who were blinded to the conclusion reached by the frontline emergency staff. Sittig (2015) reported that, although the screening protocol identified the children who were abused, a significant number of false-positive screenings were included. The interrater agreement was poor for injury caused by neglect, and the author concludes that the SPUTOVAMO-R is not an accurate instrument for abuse screening due to these limitations (Sittig, 2015).
Pierce et al. (2010) developed a screening tool, the TEN-4 Bruising Clinical Decision Rule (TEN-4) from a case-control study of children younger than 48 months (n = 95), who were admitted to a pediatric intensive care unit and screened because of trauma. TEN-4 represents bruising in children 4 years and younger in the (T) torso, (E) ears, and (N) neck and any bruising in infants 4 months and younger (Pierce et al., 2010). Of those positive screening, 75% (n = 71) of the patients had bruising, and the characteristics were modeled. Modeling is any bruising on the ears, neck, and torso in children <= 4 years old and any bruising in infants <= 4 months old and is characteristic for abuse. The TEN-4 Bruising Clinical Decision Rule has a sensitivity of 97% and a specificity of 84% for predicting abuse in young children and infants (Pierce et al., 2010).
Sexual Abuse Screening
Salvagni and Wagner (2006) developed a screening tool administered to parents to identify children who have been sexually abused. The tool is called "questionnaire for evaluating behavior, physical and emotional symptoms of children 2-12 years old." The five questions addressed: (1) sudden emotional and/or behavioral changes; (2) fear of being alone with a specific person; (3) unusual interest in sex or genitals, (4) changes in recreational activities; and (5) presence of anal or genital lesions (Salvagni & Wagner, 2006). The Cronbach's alpha for internal consistency was .71. The limitations of their study included a relatively small, purposive sample and generalizable uncertainty. Further research on this instrument was not identified in the literature search. There is no further evidence of psychometric testing of this instrument in other settings. Despite the discriminant validity reported for this instrument, the criteria for sexual abuse identified in the questions are not reliable or definitive indicators of child sexual abuse (International Association of Forensic Nurses, 2016), casting doubt on the tool's validity.
Rogstad and Johnston (2015) developed a screening tool to identify children and adolescents at risk for child sexual abuse. The tool, "Spotting the Signs," was developed and implemented in clinics where youth seek sexual health services. A series of direct questions comprise the tool and allow care providers to determine if sexual abuse is occurring. The protocol was piloted with youth and practitioners and evaluated by participants after protocol completion. It was highly rated by clinicians for child sexual abuse information accuracy, tool usability, and question acceptability. However, "Spotting the Signs" has not been empirically tested.
Psychological Abuse Screening
No tools were found in the literature to screen for psychological abuse.
Neglect Screening
Srivastava and Polnay (1997) developed the Graded Care Profile (GCP) to measure the degree of guardian neglect based on four parameters: esteem, love, safety, and physical needs. An almost perfect level of interrater agreement was found in the areas of physical care (k = 0.899, 95% CI [0.850, 0.948]), safety (k = 0.894, 95% CI [0.854, 0.933]), and esteem (k = 0.877, 95% CI [0.808, 0.946]), with a substantial degree of interrater agreement in the area of love (k = 0.785, 95% CI [0.720, 0.849]). The study found that use of the GCP by child care professionals to identify potential child neglect results in similar conclusions about the presence/absence of neglect. However, no empirical testing of the GCP is reported in this article.
Polyabuse Screening
The Children's Trauma Assessment Center at the Western Michigan University developed two comprehensive screening checklists: "A Screening Checklist: Identifying Children at Risk Ages 0-5" and "A Screening Checklist: Identifying Children at Risk 6-18," both designed for parent completion (Henry Black-Pond, & Richardson, 2010). Henry et al. (2010) use the screening questionnaire in their work with children at the Children's Trauma Assessment Center. No evidence of psychometric testing of this instrument was found. The screening questionnaires are endorsed by the National Child Traumatic Stress Network to identify children who are at risk for multiple types of trauma exposure including physical, sexual, and psychological abuse; parental substance abuse; and domestic violence. The checklist instructions imply that indicating a symptom or behavior is not definitive for abuse but suggests the need for further evaluation for the possibility of trauma exposure.
Results
Of the five screening tools for physical abuse, two were psychometrically tested to reveal adequate sensitivity and specificity, one was designed for use in the emergency setting with children of any age, and the other was designed for the pediatric intensive care setting with children 48 months and younger. Only the "Escape Form" is appropriate for use in children of all ages; the usefulness of this instrument is limited by the physical abuse and neglect focus (omitting sexual and psychological abuse screening) and use only in emergency settings.
Salvagni and Wagner (2006) designed a screening instrument for identifying sexual abuse in children 2-12 years old. According to the International Association of Forensic Nurses (2016), the assessment domains are not definitive for child sexual abuse, eliminating the instrument as a valid and reliable screening tool. This was the only sexual abuse screening tool identified in the literature and yet is without empirical support. This finding further supports the need for a tool that screens not only for sexual abuse but also for all forms of child abuse with a sexual abuse screening component.
No screening tool was found to identify psychological abuse in children. The single screening tool designed to identify neglect in young children has no empirical support (Srivastava & Polnay, 1997). Psychological abuse often co-occurs with other abuse forms (Trickett et al., 2009); integrating a psychological abuse component into a comprehensive abuse screening tool again seems a desired goal if protecting children is a priority.
The two screening checklists for children 0-5 years and 6-18 years (Henry et al., 2010) identified in the review are assessment tools used to guide trauma treatment at the Western Michigan University's Children's Trauma Assessment Center (Henry et al., 2010). Evidence of empirical testing for these screening checklists were not found, yet the checklists are recommended to clinicians for abuse screening in pediatric care settings by at least one pediatric care provider (Hornor, 2015). Positive responses serve as an alert to the clinician that further assessment is needed to confirm or rule out child abuse (Henry et al., 2010).
The "Escape Form" (Louwers et al., 2014) shows the broadest empirical support of the nine instruments reviewed. However, the "Escape Form" is primarily designed to identify physical abuse and neglect in the emergency setting. Even if the instrument could be adapted to other points of care, it remains limited in its failure to screen for sexual and psychological abuse.
Notably absent in the literature is a comprehensive tool to assess all forms of child abuse and neglect at the point of care in the healthcare delivery system. There is currently no empirically established instrument available to screen for physical, sexual, and psychological abuse and neglect for all children under the age of 18 years at any point of care in the healthcare delivery system. Screening only for physical abuse and only in the emergency setting is inadequate for identifying children who are being abused by the other common forms of abuse (sexual, psychological, polyabuse, and neglect) with serious short- and long-term health consequences.
Limitations
The primary limitation of this review was the scarcity of screening tools present in the literature for identifying child abuse. Bailhache et al. (2013) report that scarce and low-quality evidence in the accuracy of instruments for identifying abused children is a problem. Only articles in English were reviewed, increasing the possibility that child abuse screening tools were missed. However, studies from the Netherlands do not cite screening instruments not found in this review. In the Netherlands, screening for child abuse is required by law and helps explain the higher quantity of research being conducted to develop abuse screening tools (Hoytema van Konijnenburg, et al., 2014). U.S. law does not require screening for child abuse. Most identified tools screen for physical abuse in the emergency setting. Children presenting for care in other healthcare settings are not routinely screened for child abuse.
This review focused on screening for children experiencing abuse and does not address child abuse prevention. Instruments such as the Safe Environment for Every Kid (SEEK) model (Dubowitz, Lane, Semiatin, & Magder, 2012), an intervention for identifying risk factors for abuse, were not reviewed. Although prevention is preferable for any health-compromising condition, identifying children who are currently being abused and experiencing an undisclosed crisis remained the focus of this review. The optimal approach to child abuse includes prevention efforts, but such discussion is beyond the scope of this work.
Implications for Future Research
The U.S. Preventive Services Task Force (2013) has called for research to be done to identify how primary care clinicians can effectively intervene to prevent abuse and neglect and screen for it. A need exists for screening protocols that assess for physical, sexual, and psychological abuse and neglect. Comprehensive screening protocols should be developed for all points of care in the healthcare system. Abused children do not always present in the emergency department. Health professionals are in a position to identify abused children in many different settings that serve the healthcare needs of children.
Jackson et al. (2015) assert that a screening tool that cannot detect abuse before it has occurred is not a screening tool. Bailhache et al. (2013) arrived at a similar conclusion, stating that the development of valid screening instruments is a prerequisite before considering screening programs (for child abuse). There is a concerted effort in the pediatric care community to make screening for abuse risk factors a high priority to prevent child abuse and neglect from occurring (Bailhache et al., 2013; Hornor, 2015). Given the high prevalence of child abuse and the number of unreported cases, even when signs are present, screening cannot be expected to be perfect before further attempts are made to safeguard children across points of care throughout the healthcare system. Experts agree that child abuse is most often identified when children already have serious consequences from the abuse. Potential signs of child abuse and neglect need to be identified early before severe morbidity or mortality occurs (Cronholm & Witherspoon, 2016; Hornor, 2012, 2013, 2014, 2015; Oranen, 2014; Tupola, Kallio, & Kivitie-Kallio, 2014). Until comprehensive screening for abuse is a standard care practice, providers must, at minimum, be identifying abused children as early as possible for recovery intervention.
Child abuse and neglect screenings should be adapted for all points of care including the emergency setting, inpatient and outpatient medical health, and inpatient and outpatient mental health. If substantiated cases of child abuse are the tip of the iceberg for all abused children, then it is reasonable to try to safeguard children in every healthcare setting. Child abuse screening is currently inconsistent across settings, types of abuse, age range, referral protocol, and follow-up care. Coordination of services that serve the needs of children and families is needed for better outcomes for the identified children and their families.
Conclusion
The literature reveals the signs and symptoms of abuse, in addition to the short- and long-term consequences of unidentified and untreated abuse for the child and society. There is a need to utilize the available data to develop an empirically sound comprehensive screening tool, useful for identifying children at all stages of development and all points of care. Our current system of inconsistent screening at primarily one point of care, the emergency setting, is not an effective approach to identifying abuse and neglect. The literature suggests that a comprehensive screening instrument, consistently used at all points of care for children 0-17 years old, is warranted. An empirically tested, comprehensive screening protocol should be a research priority. Abuse prevention is the optimal goal, but until that becomes a reality, early identification and intervention in cases of child abuse and neglect are imperative.
References