Section III
Evidence-Based Treatments and other Effective Approaches
Chapter 8
Determining Best Practice for Treating Sexually Victimized Children
Benjamin E. Saunders
The race is not always to the swift, nor the battle to the strong, but that's the way to bet.
āDamon Runyon
Determining ābest practiceā in any endeavor, including psychotherapy with child and adolescent victims of sexual assault, is necessarily a moving target for three reasons. First, the phenomenon may change over time. In the past, the term child sexual abuse signified a relatively narrow set of experiences. It typically was taken to mean abuse by a family member (father-child or perhaps sibling sexual assault) or sexual assault of young children by adult strangers with a primary sexual interest in children, that is, pedophiles. Today, child sexual abuse (CSA) may indicate a wide range of sexual victimization experiences encountered by children and adolescents, such as being the subject of pornographic pictures distributed on the Internet, sexual assault by a peer or dating partner, children being made to observe adults engaging in sexual activity or watch pornography, drug- or alcohol-facilitated rape, exploitation through prostitution, or a long list of other sexually related victimization experiences. Our definitional perspective about the breadth of sexual victimization in childhood and adolescence has expanded over time and this change affects ideas about what might be considered best therapy practice with a much wider range of victims.
A second reason is that standards used by the field to determine best clinical practice also change over time. The suitability of various psychological therapies for CSA victims was judged differently years ago than it is today. Specifically, the weight given to scientific evidence for treatment efficacy in making determinations about best practice has increased substantially. Two decades ago, few methodologically sound, clinical treatment research studies that could be used to guide practice had been conducted with child or adolescent victims of sexual abuse (Finkelhor & Berliner, 1995; Kolko, 1987; Saunders & Williams, 1996). Today, simply having an innovative and intriguing idea for a new treatment, a persuasive theoretical argument presented by a well-respected leader in the field, and some anecdotal reports of treatment success with interesting cases are no longer sufficient criteria for an intervention to be judged a best practice. Now, a threshold level of empirical evidence for treatment efficacy is required. Exactly what the threshold is may be debated, but some positive findings from empirical tests of treatment effects are required. So, the rules of the best practice game tend to change over time.
Finally, and most importantly, new knowledge is constantly being acquired through increasingly sophisticated scientific research and from frontline practice experience, sometimes at a dizzying pace. New knowledge frequently means that previously well-accepted ideas about common interventions need to be revised or even abandoned altogether, as do judgments about what therapies constitute best practice. Treatments found to be only minimally effective, ineffective, or even harmful need to be revised or discarded altogether. Interventions that remain untested or inadequately tested increasingly are viewed with skepticism as other treatments are more thoroughly tested and found to be efficacious. And, totally new interventions may need to be developed and tested in response to new knowledge. The quantity and pace of these changes frequently are difficult for the field to digest, and incorporating them into everyday practice rarely goes smoothly. Historically in the mental health field, empirical evidence for efficacy has not been a major consideration in psychotherapeutic treatment selection. However, this perspective is changing and clinicians now are called on to respond to new knowledge by changing their longstanding practice approach, which sometimes is met with resistance. Because of these reasons, any description of specific best practices is necessarily time-bound, tied to the accepted scientific knowledge and practice lore of the day, and likely to change in the future.
Given the dynamic nature of the field, determining current best practice means practitioners need to regularly reevaluate the clinical principles, assumptions, conventional practice wisdom and research findings that guide the way they conduct their practice with CSA victims. New information concerning the definitions and the epidemiology of CSA; the impact of sexual victimization experiences on child functioning; the biological, psychological, and social mechanisms that underlie the development of abuse-related problems by victims; and improved approaches to assessment and treatment of victims and their families will challenge current assumptions and lead to changes in norms for best clinical practice. This chapter examines several questions and issues where new knowledge or new conceptualizations are affecting judgments about best practice treatment of CSA victims. The implications for best practice with CSA victims are discussed. Some of the conclusions made may seem provocative, which is intended.
Treating āSexual Abuseā
It is not uncommon for children who have been sexually victimized to be referred for āsexual abuse treatmentā as if all sexual abuse incidents and all child reactions to sexual assault are homogeneous. This sort of phraseology implies that being a sexual abuse victim is the sole defining characteristic of a child, the cause of all of his or her problems, and a singular approach to treatment is sufficient. It also implies that sexual abuse itself is something to be treated.
Unlike other forms of child maltreatment, children often are automatically referred to mental health treatment after a report of sexual abuse. In a longitudinal study of 201 children living in families reported to authorities for allegations of intrafamilial sexual abuse, physical abuse, or domestic violence, children in families reported for sexual abuse were twice as likely to be referred to mental health treatment and received three times more sessions of counseling compared to children reported for other forms of family violence even after assessed mental health problems were controlled (Saunders, Williams, & Rheingold, 2003; Saunders, Williams, Smith, & Hanson, 2005). These findings suggest that compared to other child victims, sexual abuse victims are much more likely to be referred to treatment because of what happened to them rather than due to any emotional difficulties they may have. Professionals tend to react more swiftly to CSA reports than other types of maltreatment in getting child victims mental health services even though the mental health needs of CSA victims may be similar to those of children suffering other types of victimization. Sexual abuse seems to spark a professional reaction at all levels that is more intense compared to other types of maltreatment.
Sexual abuse is a historical event in the lives of some children that unfortunately cannot be changed. Best practice treatment does not target the event. Rather it seeks to reduce abuse-related emotional and behavioral difficulties and prevent the emergence of new abuse-related problems. As noted above, sexual abuse incidents are highly diverse and the problems that may result from being sexually abused vary substantially in quality and intensity from victim to victim. Some victims may be resilient and suffer few problems, whereas others have extraordinarily complex difficulties. Few, if any, cognitive, emotional, or behavioral problems are exclusive to being sexually abused. Therefore, having a history of sexual abuse with no other assessment information should not dictate a particular treatment approach and the notion of treating sexual abuse is a misnomer the field should discard. Best practice is for treatment to be responsive to the individual psychological needs of each victim and not be based primarily on the nature of the emergent abuse report.
Do All Sexually Abused Children Need Treatment?
A large and robust scientific research literature documents that experiencing sexual assault in childhood increases victimsā risk for the development of an assortment of emotional, behavioral, and social problems, sometimes dramatically (Berliner, 2011; Kendler et al., 2000; Kilpatrick et al., 2003; Nelson et al., 2002; Putnam, 2003; Saunders et al., 1999). There is little doubt that compared to nonvictims, victims of sexual assault in childhood or adolescence are much more likely to develop serious mental health disorders such as posttraumatic stress disorder, major depression, problematic substance use, behavior disorders, delinquency, and many other problems.
However, when interpreting the CSA impact literature for the purposes of best practice treatment planning, it is critical to remember what is meant by increased risk. A risk factor is a characteristic that is associated with a greater likelihood of a particular outcome occurring. Not everyone with a risk characteristic will develop the associated outcome, and not everyone with the outcome will have the risk factor. Also, a risk factor is not necessarily a cause of the outcome. It is simply a probabilistic relationship. The presence of the characteristic only raises the actuarial probability that the outcome will occur, not insure it. The degree the risk factor raises the probability can be small, medium, or large. For example, sexual assault in childhood has been found to be a strong risk factor for some outcomes such as anxiety disorders and substance abuse, increasing the odds of having them by factors ranging from 3 to 8 times depending on the specific disorder (Chaffin, Silovsky, & Vaughn, 2005; Kilpatrick, Saunders, & Smith, 2003; Saunders et al., 1999). However, it is a moderate risk factor for depression, increasing the odds about 2 times (Kendler et al., 2000; Saunders et al., 1999).
More interesting, it turns out that specific victim characteristics (e.g., gender, age) and sexual assault incident characteristics (e.g., sexual penetration, perceived life threat during the assault) have been found to better explain the risk for mental health problems associated with child sexual assault (Acierno et al., 2001; Hanson, Borntrager, Self-Brown, Kilpatrick, Saunders, Resnick, & Amstadter, 2008; Kilpatrick et al., 2003; McCauley et al., 2009). If victims and the sexual assaults they suffer have more of these characteristics, the risk for various mental health problems considerably increases compared to situations where they are not present. Consequently, some CSA cases may carry only a small or moderate increase in risk while others present an extremely high risk for the development of mental health problems. So the well-accepted (and technically accurate) principle that CSA is a risk factor for many problems has become more complicated as increasingly sophisticated research has been conducted. It is not just a history of some form of CSA, but a range of more specific characteristics that are important to understanding the risk for victims. This information affects how to do best practice because it needs to be incorporated into assessment and treatment planning to better tailor intervention to an individual victim.
These findings concerning risk force us to recognize and acknowledge that though the increased risk is sometimes striking, for no disorder or problem is the prevalence rate for the wide range of CSA victims 100%. For example, in a national probability sample of adult women, Saunders et al. (1999) found that nearly one-third of the CSA victims had a lifetime history of PTSD, a prevalence rate 3 times greater than for nonvictims of CSA. But, this finding also means that more than two-thirds of female CSA victims had not met full PTSD diagnostic criteria. The general pattern of these findings is similar to other research studies with community samples. In repeated studies, commonly 40% of victims report no or minor mental health problems (Finkelhor & Berliner, 1995; Putnam, 2003). Therefore, somewhat surprisingly, research has found that some children can experience very serious traumatic events such as sexual abuse, and not necessarily develop serious mental health problems (Fitzgerald, Danielson, Saunders, & Kilpatrick, 2007). Sexual abuse clearly raises the likelihood for developing serious mental health problems, and a large portion of CSA victims will develop serious and long-lasting problems. However, some will not. For whatever reasons, these children are resilient to the potentially severe traumatic impact of sexual abuse.
Recognizing the reality of resilience has implications for conducting best practice and raises the tricky question: Should sexually abused children who exhibit no or subclinical symptoms receive mental health treatment? Such a question clearly challenges common practice where the large majority of CSA victims are routinely referred for treatment because of the nature of the abuse they experienced regardless of their problem profile. For some, this question may seem ridiculous on its face because of this longstanding practice. However, if the answer is, āOf course they need treatment,ā the next question is, for what? What problems will be corrected through treatment? A final question is: If one believes resilient and naturally recovering children should be engaged in mental health treatment, should they receive the same type of treatment as children with very serious abuse-related problems?
Another facet that must be considered when answering these questions is the potential side effects of engaging a child in mental health treatment when it may not be needed. Any treatment, including psychological therapies, can have unintended, harmful consequences. All treatments have an outcome matrix. Some clients get better, some get worse, and some stay the same. Even the best evidence-based therapies are not always h...