Child Sexual Abuse: Demography, Impact, and Interventions

Abstract

Because child sexual abuse (CSA) includes experiences from noncontact abuse to violent rape, the impact on victims varies greatly. Rape has the highest rates of posttraumatic stress disorder of any trauma. Depression and anxiety are associated with CSA, whereas sexualized behaviors, victim stigmatization, and shame are specific to it. The demography of CSA differs from other forms of child maltreatment by gender and social class, so that the risk factors for occurrence and prevention also differ. Evidence-based interventions for many kinds of childhood trauma and maltreatment were first established by CSA clinicians and researchers. Although well-established treatments for child and adult CSA victims now exist, evidence-based interventions are still needed for older adolescent CSA victims, especially those with polyvictimization histories.

Keywords:

• child sexual abuse
• CSA demography
• CSA impact
• CSA investigation
• CSA treatment
• CSA prevention

Child sexual abuse (CSA) is defined as any use of a child for sexual gratification by another person. It can be perpetrated by an adult, an older or more developmentally advanced child, or even a child of the same age if coercion is present. This broad definition encompasses a very wide range of experiences, from noncontact abuse (voyeurism, exhibitionism) to contact abuse that ranges from genital fondling to violent rape. The effects of CSA on victims and their families vary enormously, so that a search by CSA researchers in the 1980s to identify a “sexually abused child syndrome” similar to the “battered child syndrome” has been abandoned as unproductive (Olafson & Boat, 2000).

CSA often occurs in situations where children also suffer other adverse childhood experiences, such as child physical abuse; neglect; emotional abuse; and exposure to adult partner violence, animal cruelty, community violence, substance abusing family members, and mentally ill caregivers (Anda et al., 2006; Dong et al., 2004; Felitti et al., 1998; Finkelhor, Ormrod, & Turner, 2007). A recent multisite treatment study of 229 sexually abused children aged 8 to 14 reported a mean of 2.6 additional traumas experienced by these children. Seventy percent of these children had experienced a traumatic loss (e.g., the sudden death of a family member), 58% had witnessed domestic violence, 37% had been in serious accidents, and 26% reported having been physically abused (Cohen, Deblinger, Mannarino, & Steer, 2004). Because CSA is often part of a pattern of polyvictimization, it is difficult to isolate the effects of one type of trauma or abuse from the other. In this respect, CSA resembles child physical abuse and neglect and may differ from stressors or traumas such as disasters, ethnic violence, and war.

There are, of course, many cases in which children who experience either familial or nonfamilial sexual abuse live in homes that appear to be orderly and otherwise nonabusive and in which obvious risk factors for childhood victimization appear to be absent (Finkelhor et al., 2007). Familial sexual abuse in these cases of generally higher socioeconomic status (SES) can be disproportionally difficult to prevent, identify, and prosecute.

Although most perpetrators of CSA are known to their child victims, it differs from child physical abuse or neglect in that the majority of CSA perpetrators are not immediate family members or other relatives, a difference that is more marked for male than for female victims (Berliner, 2011; Hanson et al., 2006).

Demographic Characteristics and Challenges

Both retrospective surveys and official reports indicate that CSA has lower prevalence rates than child physical abuse and neglect. The National Child Abuse and Neglect Data System (NCANDS; Administration for Children & Families, 2007) compiled from data submitted by states for 2007 show that CSA comprised 7.6% of total cases (it was double that 10 years ago), with neglect the most common at 59.1% (NCANDS, 2007). However, because it appears that most cases of CSA are still not disclosed during childhood or are not reported to the authorities (Olafson & Lederman, 2006), the true prevalence of CSA remains unknown. Retrospective surveys offer additional data, but there is evidence that “survey reluctance” and inadequate questioning strategies may contribute to underreporting by former CSA victims who are now adult (Lyon & Ahern, 2011). Retrospective surveys across many countries have found rates of CSA from 7% to 36% for females and 3% to 29% for males (Finkelhor, 1994). One American review of 16 community sample surveys calculated CSA prevalence for women at 16.8% and 7.9% for men (Gorey & Leslie, 1997). The National Violence Against Women Survey conducted with a representative sample of 16,005 men and women found that 9% of the women and 2% of the men reported having been raped before the age of 18 (Tjaden & Thoennes, 2000).

Both class and gender characteristics specific to CSA inhibit effective public awareness and intervention. CSA differs from other forms of trauma and maltreatment (with the exception of domestic violence) because males are far more likely to be perpetrators and females to be victims. When victims are female, males are perpetrators in about 94% to 95% of cases, and when victims are male, males are found to be perpetrators in about 80% to 85% of cases (Seto, 2008; Whealin, 2004). Girls are two to three times more likely to be victimized by CSA than are boys (Gorey & Leslie, 1997; Putnam, 2003).

Issues of gender, class, and memory were central to the rediscovery of CSA in the late 20th century in ways that distinguish CSA from other forms of child maltreatment, such as child physical abuse (Olafson, 2002; Olafson, Corwin, & Summit, 1993). The emerging knowledge about the battered child in the 1960s came from physicians who were treating victimized children, that is, from credentialed professionals speaking on behalf of children. The caregivers identified by physicians as physically abusive to their children generally lacked access to economic resources, political power, and the media. A decade later, the rediscovery of CSA was initially sparked not primarily by professional meaning-makers but by adult women who reported having been sexually abused, often decades before (Olafson, 2002). Modern awareness of CSA thus emerged in the midst of the gender wars of the 1970s, when birth rates were plummeting and interest in and funding for children's issues in the United States was waning. The adults identified as alleged perpetrators of CSA were often wealthy and powerful and had access to excellent lawyers, influential policy makers, and the media (Olafson, 2002).

Indeed, there are significant ethnic and class differences in prevalence rates and risk factors that distinguish CSA from other forms of childhood maltreatment. Whereas it is well established that lower SES is a strong risk factor for both child physical abuse and neglect, community surveys suggest that SES does not appear to be related to CSA occurrence (Berliner, 2011; Putnam, 2003). Retrospective studies indicate that, unlike other forms of child maltreatment, CSA has prevalence rates that are roughly equivalent in families of all social classes, from the prosperous suburbs to the impoverished inner cities (Berliner, 2011; Putnam, 2003). Actual protective services reporting rates do not, however, reflect this reality; CSA reporting rates are disproportionally lower for children from the higher socioeconomic strata (Finkelhor, 1993; Putnam, 2003). Retrospective surveys do show minor ethnic differences; African American and Caucasian women report similar rates, whereas Native American women report somewhat higher and Asian women somewhat lower rates. Hispanic adolescent girls report significantly higher rates of CSA (Berliner, 2011).

Berliner (2011) lists the following risk factors for CSA: (a) girls are at greater risk of being sexually abused than are boys, (b) children with disabilities have almost double the reported incidence of CSA than do children with no disability, (c) both boys and girls who have lived without one of their natural parents are at increased risk, and (d) lower SES does not appear to be a risk factor.

Class and gender realities pose perhaps the most daunting challenge to sustained public policy attention and funding for CSA awareness, prevention, and intervention. The “backlash” against the late 20th century rediscovery of CSA, which was fully evident by the mid-1990s (Myers, 1994), was strengthened by the class and gender realities of this crime against children. As middle and upper class men (and a few women) with privileged access to the media, to policy makers, and to government officials found themselves accused of sexual crimes against children, they individually and collectively, in organizations such as the False Memory Syndrome Foundation (FMSF), found allies in academia (Loftus & Ketcham, 1994), in the government, and in media outlets across the political spectrum. Flawed investigations, poorly conducted interviews, and some apparent false allegations and convictions formed the raw material for a widespread backlash against CSA awareness, so that doubts about the credibility of alleged child victims and adult survivors and concern about the competence of agencies, advocacy centers, and law enforcement organizations that investigate CSA may continue to this day to affect case investigations and outcomes (Olafson, 2002, 2004).

This pattern of discovery and rediscovery of CSA (often connected with feminist movements), followed by powerful class and gender-based backlashes, is more than a century old (Herman, 1992; Olafson et al., 1993; Sacco, 2009). Herman succinctly summarized the necessity for a feminist analysis of this repeated pattern:

Without a feminist analysis, one is at a loss to explain why the reality of incest was so long suppressed by supposedly responsible professional investigators, why public discussion of the subject awaited the women's liberation movement, or why the recent apologists for incest have been popular men's magazines. (p. 3)

Have the effects of the most recent backlash subsided, or are they still evident in recent statistics documenting a relative decline in substantiated sexual abuse cases in the past 15 years? Has CSA actually declined since the mid-1990s, or are cases left unreported to social services or, if reported, not forwarded for investigation or not substantiated? Although it has been argued that the decrease in CSA reporting represents a true decline (Jones, Finkelhor, & Kopiec, 2001), there is disagreement about whether this represents an actual decline and, if so, what the reasons for it are (Putnam, 2003). The most recent National Incidence Study (Sedlak et al., 2010), which interviewed reporters and found no evidence of reluctance to report, suggests that this may be a true decline, one that parallels declines in other forms of violent crime in recent years (Berliner, 2011).

Was Any Crime Committed?

A separate challenge arises because of the nature of certain kinds of CSA acts. Other forms of criminal child abuse would also be criminal if perpetrated on an adult. Thus, it is abusive and criminal to beat, starve, rape, or deliberately injure either an adult or a child. There are, however, forms of CSA that are criminal only because the victim is a child and therefore incapable of consent (Finkelhor, 1979). After a century of confusion and denial in both popular culture and professional opinion, it was finally firmly established in the last years of the 20th century that sex between adults and children is both criminal and wrong, is never the responsibility of the child, and that children lack the ability to consent (Olafson, 2002; Olafson et al., 1993). “A child's age, dependent status, and inability to consent make nonviolent CSA both criminal and wrong, even when the child does not appear to be physically or emotionally damaged by the encounter” (Olafson, 2004, p. 152).

It is important to stress that nonviolent sexual contact would not be criminal between consenting adults. Adults who are “courted” can consent to sexual contact or they can decline. Children who are “groomed” (and grooming often superficially resembles courting) cannot legally or ethically consent to sexual contact because they are minors (Finkelhor, 1979). In addition, children's dependent status and developmental limitations hamper their ability to decline (Summit, 1983). One hopes that the excellent work done at the end of the last century on this issue has put to its final resting place the “consenting” or “seductive” child of almost 100 years of misguided psychiatric and popular literature in Western Europe and the United States (Finkelhor, 1979; Herman, 1981, 1992; Olafson, 2002, 2004; Olafson et al., 1993). However, it must be noted that some adolescents who have been victimized via personal or Internet connections view these sexual contacts as consensual and do not see themselves as victims. They have been described as “compliant victims,” and they pose special challenges both for intervention and for treatment (Berliner, 2011; Wolak, Ybarra, Mitchell, & Finkelhor, 2007).

Impact of CSA

Perhaps because CSA encompasses many different kinds of acts, the responses and symptoms of victims and adult survivors vary greatly, from essentially asymptomatic to lifelong, disabling psychological, behavioral, and health consequences. Putnam (2003) writes that sexually abused children “constitute a very heterogeneous group with many degrees of abuse about which few generalizations hold” (p. 269). Even when sexually abused children do not experience long-standing psychological symptoms, they are at increased risk for future victimization; impaired adult functioning; and altered attitudes about self, others, and the world (Berliner, 2011; Fargo, 2009).

The following variables affect the severity and duration of victim symptoms and behaviors: prior or concurrent traumas, pre-existing psychological disorders, the nature of the abuse, relationship to the perpetrator, duration of the abuse, level of support by the nonabusive caregiver, and gender. Poorer long-term outcomes are associated with the following abuse characteristics: contact rather than noncontact abuse; penetration (for both child and adult victims, rape is associated with the highest rates of posttraumatic stress disorder [PTSD] of any form of interpersonal violence; Bloom & Reichert, 1998); sexual abuse with aggression, violence, or coercion; sexual abuse that begins early and lasts through more than one developmental stage; and a close relationship (generally familial) with the perpetrator. In addition, sexually abused boys generally experience worse short- and long-term outcomes than do girls (Beitchman et al., 1992; Berliner, 2011; Putnam, 2003).

The presence of a supportive, protective, nonoffending parent who believes the child (generally the mother) following disclosure predicts better recovery from CSA (Cohen & Mannarino, 1998, 2000). Negative parental responses are associated with worse outcomes for children (Berliner, 2011).

All forms of childhood trauma and maltreatment are associated with affective disorders, and the relationship between CSA and both major depression and dysthymia is especially strong. In women, a general history of CSA is associated with rates of depression three to five times those of nonabused women (Putnam, 2003). Rates are higher the more severe the abuse. Thus, in one population study, children who reported CSA with intercourse had an increased odds ratio of 8:1 for major depression when compared to nonabused controls (Fergusson, Horwood, & Lynskey, 1996). Because girls are more likely to be sexually abused than are boys, at least one study has shown that, when a history of CSA is controlled for, the usual 2:1 adult gender differential for depression rates between adult men and women disappears (Whiffen & Clark, 1997).

When CSA is severe and long lasting, symptoms can include disabling PTSD, dissociative disorders, drug and alcohol dependence, anxiety disorders, conduct disorders, vulnerability to revictimization, and high-risk sexual behaviors (Beitchman et al., 1992; Fergusson et al., 1996, Olafson, 2004; Putnam, 2003). Sexually abused children have higher rates of PTSD (more than one third of them) than do children who have experienced other forms of maltreatment or traumas not related to abuse (Berliner, 2011; Dubner & Motta, 1999). In severe early childhood cases, brain development is affected (Cohen, Perel, DeBellis, Friedman, & Putnam, 2002). Sexually abused children have more behavior problems than nonabused children, including oppositional and conduct problems (Pollio, Deblinger, & Runyan, 2011).

Drug Dependence and Alcoholism

Data analysis from the National Longitudinal Study of Adolescent Health, which uses a nationally representative sample of adolescents (14,078), showed that sexual abuse alone increased the odds for adolescent drinking behavior more than histories of neglect or abuse alone and almost identical odds with those who had repeated multiple forms of maltreatment, such as neglect, physical abuse, sexual abuse (Shin, Edwards, Heeren, & Amodeo, 2009). The authors conclude, “It appears that childhood maltreatment has an independent effect on the risk of adolescent alcohol use regardless of parental alcoholism” (p. 231). Women who experienced any type of sexual abuse in childhood (nongenital, genital, and intercourse) were roughly three times (2.93) more likely than nonabused girls to report drug dependence as adults. Indeed, CSA was more strongly associated with drug or alcohol dependence than with any of the psychiatric disorders (Kendler et al., 2000).

Of the traumagenic dynamics of CSA identified by Finkelhor and Browne (1985) a generation ago, two stand out as distinguishing CSA from other forms of childhood trauma and maltreatment. These are Traumatic Sexualization and Stigmatization.

Traumatic Sexualization

Traumatic Sexualization linked to a CSA history increases the risk for children and adolescents to have sexual behavior problems (Andrews, Corry, Slade, Issakids, & Swanston, 2004) and for adolescents and adults (primarily males) to become sexual offenders against children (Jesperson, Lalumiere, & Seto, 2009). Sexually abused adolescent girls are at increased risk for early pregnancies and both boys and girls for engaging in promiscuous and high-risk sexual behaviors that increase likelihood of early pregnancies and/or exposure to sexually transmitted diseases, including HIV (Fargo, 2009; Noll, Shenk, & Putnam, 2009; Putnam, 2003; van Roode, Dickson, Herbison, & Paul, 2009). A previous history of CSA strongly predicts subsequent revictimization by rape, as well as an increased arrest rate in adulthood in both sexes for sexual crimes and prostitution (Widom & Ames, 1994). In the current state of the research, it appears that these associations between a history of CSA and this array of problematic sexual behaviors do not strongly exist for emotional abuse and neglect, although very few studies have examined this relationship (Jesperson et al., 2009; Widom & Ames, 1994). Of 10 studies that report data on child physical abuse, seven found a higher prevalence of physical abuse histories among sex offenders versus non–sex offenders, four at the level of significance, with a weighted odds ratio of 1.60 for all 10 studies or about half the 3.36 odds ratio for the CSA–sexual offending correlation (Jespersen et al., 2009). It is, however, important to stress that the great majority of boys and girls sexually and/or physically abused during childhood do not become adolescent or adult sexual offenders (Jespersen et al., 2009).

The pattern of high-risk sexual behaviors from late adolescence into early adulthood appears to differ for males and females. In a longitudinal study of a New Zealand birth cohort, almost one third of the women and 9.1% of the men reported contact CSA. Sexually abused women reported significantly more partners, unwanted pregnancies, abortions, and sexually transmitted diseases than nonabused women in early adulthood (ages 18–21); these effects decreased over time in reports at ages 26 and 32. By contrast, abused men showed no significantly elevated rates in the 18- to 21-age period but had elevated rates for sexual partners and elevated rates of herpes simplex 2 after age 21. The authors conclude that high-risk sexual behaviors for women with sexual abuse histories peak in late adolescence and lessen with age, whereas high-risk sexual behaviors by men with CSA histories appear somewhat later and carry into adulthood, a gendered developmental finding that has implications for treatment interventions (van Roode et al., 2009).

When CSA is not experienced by a young child as painful or violent, it still leaves a residue as the developing child learns that what came disguised as affectionate attention from a loving caregiver was instead sexual exploitation (Freyd, 1996). The consequences can leave CSA survivors bewildered, avoidant, and profoundly distrustful about adult love and sexuality. A case example of a survivor of sexual abuse by the night attendant in his residential treatment home when he was aged 10 to 13 demonstrates the dilemma. This adult survivor reported in treatment that the gifts, sweet attentions, and affection he received from this man constituted the only love he ever received in childhood, either from his severely addicted parents or from any of the dozen sets of foster parents with whom he had been placed before residential treatment. “If he didn't love me as a child, then there was no one who loved me,” he reported, “and that leaves me heartbroken.” When it became public during a subsequent criminal trial that he had been one of dozens of victims in this offender's preferred age range (males, aged 9–12), this adult survivor now in his 30s was overwhelmed with sadness, rage, and shame at his youthful gullibility and after years of sobriety returned to heavy drinking.

Stigmatization

In cultures around the world and with varying intensity, to be sexually assaulted is to be dishonored, defiled, and permanently stigmatized. Victims rather than perpetrators are likely to experience debilitating shame, victims are often blamed, and perpetrators more often than not face no punishment (Olafson, 2004). In those surviving patriarchal cultures dominated by traditional notions of family honor, such as rural Pakistan or Egypt, brothers or fathers have still been reported to “cleanse” their families by killing daughters or sisters who have been sexually assaulted (Krug, Dahlberg, Mercy, Zwi, & Lozano, 2002).

Although some milder victim blaming occurs with all interpersonal crimes (and to a lesser degree, even with accidents), stigmatization is especially severe for sexual assault victims (Janoff-Bulman, 1992). In ancient Greece, a stigma was a mark of disgrace, reproach, or shame. Like Greek slaves or criminals who were literally branded on their bodies by the stigmata that defined their status, CSA victims often feel marked in the flesh by what has been done to them, so that CSA victimization is often linked to intensely disabling feelings of shame (Feiring & Taska, 2005). Kim, Talbott, and Cicchetti (2009) write, “Shame is a highly aversive, debilitating, affective experience with a profound negative evaluation of the self” and add, “Shame's intensity may be more than the individual can bear” (p. 363), resulting in attempts to flee from the feeling by isolating oneself, shutting down, becoming angry, abusing substances, and even becoming violent. Kim and colleagues examine the relationship between CSA-linked shame in adult women and the level of interpersonal conflict in their personal and family lives and recommend treatment strategies that assess and address the links of sexually abused women's shame to their interpersonal conflicts in adulthood. Well-established treatment programs for sexually abused children and their caregivers use the construction of trauma narratives, psycho-education, and cognitive processing to reduce the shame that many victims experience (Cohen, Mannarino, & Deblinger, 2006; Resick & Schnicke, 1993).

Interventions

Prevention

In all areas of child maltreatment, prevention research has lagged behind research for intervention and treatment, and this is especially so for CSA. In addition, because the risk factors for CSA differ from the overlapping risk factors for partner violence, child physical abuse, and child neglect, CSA prevention strategies must differ. The usual social service prevention efforts target the stresses of poverty in order to reduce family violence and child neglect by means of supportive family and community services. These social and economic programs do not adequately target the risk factors for CSA because this crime does not appear to be so strongly linked to poverty and its stresses. CSA prevalence rates cross class lines in ways that prevalence rates for violence and neglect do not.

School-based educational programs about CSA concepts, self-protection, and directions to tell a trusted adult if approached or abused have been effective in increasing children's knowledge about CSA and about self-protection. The sole long-term study found that female college students who had received such school-based sexual abuse education programs were significantly less likely to have been sexually abused than girls who had not (Gibson & Leitenberg, 2000). However, such programs do not reach the many preschool victims of CSA.

Across social classes, the presence in the home of a protective mother (rather than one who is physically or mentally ill, disabled, or addicted) has been identified as a protective factor from CSA (Berliner, 2011). Home visiting programs and family services that strengthen and support mothers have the potential to reduce all forms of child maltreatment, including CSA (MacLeod & Nelson, 2000). Parental monitoring of Internet website access by children and adolescents has now become an essential part of CSA prevention, and efforts to increase public awareness of this danger should be intensified (Wolak et al., 2007).

A wave of federal, state, and local legislation that requires sex offender registration, residential restrictions, and community notification began in 1990 and continues until the present day. In their recent critical review of such programs, Letourneau and Levenson (2011) write, “In summary, there is little evidence that sex offender registration and notification laws achieve the intended goal of preventing known sex offenders from repeating their crimes. There are unintended consequences of these laws that interfere with an offender's rehabilitation and reintegration in the community …” (p. 313). The authors are especially concerned about and critical of juvenile sex offender registration legislation (the federal Adam Walsh Act) and recommend several changes, including that only high-risk youth should be registered and that there should exist opportunities for removal from registration in the absence of new sex crimes (Letourneau & Levenson, 2011).

Child Forensic Interviewing

CSA most often occurs in private, so that generally only the victim and perpetrator know that it is happening. Unlike with child physical abuse or child neglect, witnesses to CSA are rare, and diagnostic or definitive physical findings occur in fewer than 5% of cases, although nonspecific physical findings are more common (Finkel, 2011; Frasier & Makoroff, 2006). Because there are generally fewer corroborative indicators for CSA than for other forms of child maltreatment, the child's willingness to disclose and the quality of the child forensic interview are crucial to determine what, if anything, has happened or is taking place and to gather evidence necessary for protection and prosecution. After the problematic mass day care cases of the 1980s and early 1990s, aggressive action was taken to establish an evidence base for effective child forensic interviewing. Among the dozens of researchers who engaged in this essential work, the achievements of Michael Lamb and his colleagues at the National Institute of Child Health and Human Development (NICHD) and in field and laboratory research facilities in England, Sweden, Canada, and the United States stand out for their range and rigor (Brown & Lamb, 2009; Kuehnle & Connell, 2009; Lamb, Hershkowitz, Orbach, & Esplin, 2008).

By the mid 1990s, child forensic interviewing experts in the field agreed on the basic principles necessary to invite full and accurate information from children. These strategies included not only the structure and placement of questions and prompts but also evidence-based guidelines about interviewer manner. However, field studies quickly established that teaching these principles to social workers and police officers who conducted most interviews had little effect on their actual behavior during interviews with children. In response, the NICHD experts wrote an interview protocol structured throughout with open questions, a protocol that has been tested in thousands of field interviews in four countries and has by far the strongest evidence base of any child forensic interviewing model (Kuehnle & Connell, 2009; Lamb et al., 2008). Originally developed for use with child physical and sexual abuse allegations, this protocol can also be adapted for use with child witnesses to domestic violence and other crimes and for the interviewing of alleged juvenile offenders (Lamb et al., 2008).

The NICHD protocol does not, however, work equally well in all cases (Olafson & Kenniston, 2008). More work is still needed to establish evidence-based guidelines or protocols for very young children, for children with developmental disabilities, and for nondisclosing or partially disclosing children and adolescents (Brown & Lamb, 2009; Faller, 2007; Lyon & Ahern, 2011; Pipe, Lamb, Orbach, & Cederborg, 2007; Tishelman & Geffner, 2010a, 2010b). There is a need for rigorous comparative field and laboratory work about protocols and guidelines such as the Narrative Interview, the Cognitive Interview, anatomical dolls, and anatomical drawings (Brown & Lamb, 2009). For nondisclosing or reluctant children in otherwise corroborated CSA cases, research in Israel and the United States has now demonstrated that more than a single interview may be essential, but protocols for repeated interviews have yet to be fully developed and tested (La Rooy, Lamb, & Pipe, 2009; Tishelman, Meyer, Haney, & McLeod, 2010). The National Child Advocacy Center (NCAC) has developed and field tested a six-session model for such children, and this NCAC model is currently being updated and researched (Carnes, Nelson-Gardell, Wilson, & Orgassa, 2001; Carnes, Wilson, & Nelson-Gardell, 1999; Faller, personal communication, June 24, 2010; Kuehnle & Connell, 2009).

Treatment

Although less common than other forms of childhood trauma and maltreatment, CSA has been the area in which the most rigorous and extensive research for childhood trauma and maltreatment has been conducted. Indeed, the single treatment with the strongest evidence base for childhood trauma and maltreatment, Trauma-Focused Cognitive Behavioral Therapy (TF-CBT), had its origins in pioneering work of the treatment of CSA (Cohen et al., 2006); TF-CBT has now been expanded to cover other forms of childhood trauma, maltreatment, and traumatic loss (Cohen et al., 2006; Deblinger & Heflin, 1996; see Mannarino & Cohen, this issue).

TF-CBT is component-based, and it offers psycho-education about CSA and other traumas for parent and child; coping and stress reduction strategies; affect expression and modulation; and a central gradual exposure component focused on creation of a trauma narrative by the child, using language, art work, poetry, or dance. Inaccurate and unhelpful trauma-related cognitions by both parent and child are gently challenged through Socratic questioning and other techniques, and sessions end with a child–parent sharing of the child's trauma narrative and safety planning. Parent sessions also focus on teaching and practicing child behavior management, including training in the implementation of a simple and effective set of sexual behavior rules (Cohen et al., 2006).

In a review of 24 child abuse treatments funded by the Office of Victims of Crime, TF-CBT was the only treatment to earn a “well supported and efficacious” rating (Saunders, Berliner, & Hanson, 2003). In a subsequent meta-analysis, Silverman and colleagues (2008) found TF-CBT to be the only one to meet criteria as “well established.” TF-CBT has been researched thoroughly in seven pre–post or quasi-experimental studies and at least seven randomized controlled studies with comparison groups (Pollio et al., 2011). In one of the largest randomized controlled TF-CBT trials, children treated with TF-CBT showed significantly more improvement on measures assessing PTSD, depression, behavior problems, shame, and interpersonal trust than children in the comparison group who were treated with Child-Centered Therapy. Parents showed greater improvement in depression, abuse-specific distress, support for the child, and effective parenting practices (Cohen et al., 2004). A one-year follow-up showed that improvements in both parents and children continued, demonstrating the effectiveness of TF-CBT treatment over time (Deblinger, Mannarino, Cohen, & Steer, 2006).

Information about training for agency providers in TF-CBT is available at the National Child Traumatic Stress Network website (http://www.nctsn.org), and a web-based training program that offers continuing education credits is at the Medical University of South Carolina website (musc.edu/tfcbt). Treatment developers emphasize that ongoing clinical supervision and peer review are essential following training to ensure fidelity to the model and address unique clinical issues as they arise. TF-CBT research continues with applications to children in foster care, group treatments, children in residential treatment programs, and children who have undergone other traumas (Pollio et al., 2011).

Although researchers have recently begun applying and studying TF-CBT with older adolescents (Cohen, Mannarino, Perel, & Staron, 2007), most of the research has focused on children and adolescents through age 15. Beginning with adulthood, there are well-established treatment programs for rape and sexual trauma victims, including evidence-based Cognitive Processing Therapy and trauma treatments by Resick and Schnicke (1993), Briere and Scott (2006), and others. Additional comparative treatment outcome studies for older adolescents are still needed, especially for those with sexual behavior and substance abuse problems linked to their CSA histories and for those in residential and juvenile facilities.

There are a number of promising practices for both group and individual treatment currently being developed and researched for these often very troubled adolescents. Prolonged Exposure Therapy (Foa, Chrestman, & Gilboa-Schechtman, 2009), Dialectical Behavior Therapy for suicidal adolescents (Miller, Rathus, & Linehan, 2007), Integrative Treatment (Briere & Lanktree, 2008), Trauma and Grief Component Therapy for Adolescents (Saltzman, Layne, Pynoos, Steinberg, & Aisenberg, 2001), and other possibly efficacious treatments being delivered and tested through the National Child Traumatic Stress Network and elsewhere promise that, in the next decades, more proven treatments will become available for children, adolescents, and their caregivers who have been affected by CSA. Further information about well-established, possibly efficacious, and promising treatment interventions currently being explored can be accessed at the National Child Traumatic Stress Network website (http://www.nctsn.org).

Future Directions

Prevention, detection, assessment, intervention, and treatment of CSA have developed impressively since the “rediscovery” of CSA in the late 20th century, but the field is still relatively new. Much more needs to be done about prevention, especially for preschool victims. Because it appears that a majority of children either do not disclose or delay disclosing, secondary prevention efforts should include public education of all adult caretakers to be alert to possible signs and symptoms of ongoing victimization. For those children whose cases are investigated and who are ready and able to talk to interviewers, there exists an excellent evidence-based interviewing protocol. Research on additional interview strategies to reach younger children, those with developmental delays, and those in high suspicion cases who deny or recant during formal interviews is underway, and results should be forthcoming in this decade. For children and early adolescents, there exists a well-established treatment, TF-CBT. Additional TF-CBT studies are underway for older adolescents, especially those with polyvictimization histories, and also for a number of similar CBT group and individual treatment protocols. Additional gender-specific interventions that address the unique challenges of male-on-male sexual abuse, female-on-female sexual abuse, and specific interventions to help adolescents and young adults with sexual behavior problems linked to their sexual abuse histories would broaden the scope of evidence-based treatments.