The National Task Force on Juvenile Sexual Offending consists of 20 members and 20 advisory members (NAPN, 1993). The Task Force was formed in 1986 after National Adolescent Perpetrator Network (NAPN) members (treatment providers and intervention specialists from more than 800 programs) supported the idea of creating a group to develop standards for the assessment and treatment of juvenile sex offenders. Recognizing that sufficient research did not yet exist to warrant a presentation of intervention standards, the National Task Force (as cited in NAPN) articulated a set of assumptions intended to reflect the current thinking relevant to a comprehensive systems response to juveniles who have sexually offended. These assumptions are summarized below.
Treatment programs for juvenile sex offenders have proliferated during the past decade. According to NAPN (1988), there were only 20 such programs in the United States in 1982. The 1994 Safer Society Program's national survey (Freeman-Longo et al.) identified 684 programs.
To adequately address both the needs of individual juveniles who have committed sex offenses and the needs of the community, a continuum of care is recommended (Bengis, 1997; NAPN, 1993). Offering a range of interventions and placement options makes it possible to provide cost-effective interventions while placing paramount importance on community safety. Such a continuum, as described in the Oregon Report on Juvenile Sex Offenders (Avalon Associates, 1986), may include:
The NAPN (1993) stressed that the primary objective of interventions with juveniles who have sexually offended is community safety. Cellini (1995) described the primary goals of treatment interventions with these juveniles as helping them to gain control over their sexually abusive behaviors and to increase their prosocial interactions with peers and adults. Similarly, Becker and Hunter (1997) described the main treatment objectives as preventing further victimization, halting the development of additional psychosexual problems, and helping the juvenile develop age-appropriate relationships with peers.
To accomplish these goals, highly structured interventions, frequently involving written treatment contracts, are recommended (Morenz and Becker, 1995). Treatment approaches include individual, group, and family interventions. Although group therapies often are described as the treatment of choice (NAPN, 1993), empirical support for this claim is lacking (NAPN, 1993; Weinrott, 1996). Similarly, cotherapy teams, preferably involving a female therapist and a male therapist, also are recommended (NAPN, 1993), but the necessity of such teams has not been demonstrated.
As Marshall and Barbaree (1990) noted in their review of the effectiveness of adult cognitive-behavioral sex offender treatment programs, most cognitive-behavioral programs combine individual treatment approaches with group therapy. Individual treatment typically addresses sexual preference interventions and some aspects of social functioning. Marshall and Barbaree pointed out, however, that individual therapy is expensive and often is not cost effective. Group therapy can be a more efficient means of concurrently presenting the educational components of treatment to a number of offenders. Furthermore, male-female therapist teams can model egalitarian relationships between the sexes for group members, and group members may be able to draw on their own experiences as offenders to provide valuable insights into other offenders' difficulties. Marshall and Barbaree also noted that group processes can facilitate new ways of thinking and social interaction that are unavailable in "traditional individualized treatment." On the other hand, the potential advantages of group therapies must be weighed against the possible disadvantages related to negative peer group associations, as have been identified in the juvenile justice field (e.g., Fagan and Wexler, as cited in Henggeler, Melton, and Smith, 1992).
The first step in treatment typically involves helping the juvenile accept responsibility for his or her behavior (Becker and Hunter, 1997). A number of factors (e.g., legal defense strategies and parental disbelief), however, can make this a difficult task. Minimizing and denying abusive behavior are common responses and are typically viewed as problematic (NAPN, 1993). Barbaree and Cortoni (1993) noted that denial is so often considered such an obstacle to effective treatment that many programs will not accept individuals who are unremitting in their denial. Barbaree and Cortoni also observed, however, that once the juvenile sex offender's denial and minimization are reduced, the offender can begin to empathize with the victim. Barbaree and Cortoni consider the reduction of denial and minimization and the development of empathy with the victim to be the necessary "first step" in facilitating the offender's motivation for treatment and behavior change.
Recommended treatment content areas for juveniles who have sexually offended typically include sex education, correction of cognitive distortions (cognitive restructuring), empathy training, clarification of values concerning abusive versus nonabusive sexual behavior, anger management, strategies to enhance impulse control and facilitate good judgment, social skills training, reduction of deviant arousal, and relapse prevention (Becker and Hunter, 1997; Hunter and Figueredo, 1999; NAPN, 1993). Other relevant interventions include training in vocational and basic living skills, assistance with academics, resolution of personal victimization experiences, assistance with coexisting disorders or difficulties, resolution of family dysfunction and impaired sibling relationships, and development of prosocial relationships with peers, dating skills, and a positive sexual identity (Becker and Hunter, 1997; Hunter and Figueredo, 1999; NAPN, 1993). Research comparing adolescent sex offenders with a group of runaways found that the former were especially deficient in their general knowledge about AIDS and safe sex practices (Rotheram-Borus, Becker, Koopman, and Kaplan, as cited in Becker and Kaplan, 1993). Given this finding, the importance of focusing treatment on sexually transmitted diseases and safe sex is obvious.
Goocher (1994) noted that leaders in the field of juvenile sex offender treatment, such as Judith Becker and John Hunter, have argued that programs designed to focus exclusively on sex-offending behaviors are of limited value and have recommended a more holistic approach. Goocher further pointed out that, in view of the individual needs and developmental histories of these juveniles, "quasi-corrections models" of addressing sex offending are not adequate. Goocher noted that many residential treatment programs for juvenile sex offenders have been based on quasi-corrections models of treatment adapted from work with adult sex offenders. Goocher also observed how, in one program, staff seemed to replicate the juveniles' power and control behaviors and secretive behavior in the staff's own interactions among themselves and in their interactions with the institution's managers, with other units, and with the juveniles. He recommended that the staff in such programs be sensitive to their positions as role models and guides for juveniles who are attempting to move beyond their life experiences and offense histories and that staff receive adequate training to enable them to perform this function.
Miner and Crimmins (1995) identified social isolation from positive interactions with peers and families as a possible factor that may explain why some seemingly prosocial juveniles engage in sexually aggressive acts. They suggested that treatment efforts should break the process of social isolation and noted that most programs do this through group and social-cognitive interventions. They further recommended family interventions and facilitation of positive school attachments and positive emotional attachments in general as treatment goals.
Weinrott (1998a) noted that some treatments that are theoretically sound but have not been empirically related to sexual recidivism may also be appropriate for juvenile sex offenders. For example, Weinrott, noting that truancy is empirically associated with sexual recidivism, recommended that treatment actively target improved school performance. In addition, because appropriate and effective dating skills can increase access to appropriate sexual partners, Weinrott and others (e.g., Bourke and Donohue, 1996) emphasized development of dating skills as a treatment component. Weinrott also encouraged more aggressive interviewing techniques, such as interrogation approaches used by law enforcement, to get through denial quickly so that treatment can proceed in a more timely fashion.
Although psychopharmacological interventions, including sex-drive reducing medications such as medroxyprogesterone, have been found to be effective in reducing sex offending in adult offenders, they can have serious side effects. Such medications, when used with juveniles, can have possible negative effects on normal development and growth. Consequently, ethical concerns related to the use of these medications with juveniles are substantial (Hunter and Lexier, 1998).
Other medications sometimes are used with juveniles as part of a comprehensive treatment approach. For example, Hunter and Lexier (1998) noted reports from the professional literature that describe the utility of selective serotonin reuptake inhibitors (SSRI's). Lane (as cited in Hunter and Lexier, 1998) reported that SSRI's often have sexual dysfunction side effects such as suppressed sexual desire and delayed ejaculation. However, as Hunter and Lexier noted, the role of serotonin in regulating sexual behavior is not fully understood. Many questions concerning psychopharmacological approaches remain. These questions include which juveniles are likely to benefit from such an approach and at what dosages (Hunter and Lexier, 1998).
Addressing Deviant Arousal
Weinrott (1998a) stressed that juvenile sex offenders with deviant sexual arousal should be provided with treatment that effectively addresses this problem. Most programs that address deviant arousal do so through covert sensitization, a treatment approach that teaches juveniles to interrupt thoughts associated with sex offending by thinking of negative consequences associated with abusive behavior (Becker and Kaplan, 1993; Freeman-Longo et al., 1994). Weinrott raised the concern that this technique, as typically used, may not be vivid enough to be effective for adolescents who might not have the language abilities to design effective fantasies to counter deviant thoughts or who may simply find the task too boring. He also stated that behavioral conditioning with noxious stimuli, such as ammonia and, possibly, low-intensity electric shock, may be effective. The National Task Force on Juvenile Sexual Offending, however, advised that use of aversive therapies with juveniles is controversial (NAPN, 1993). It recommended that, when used, aversive stimuli should be self-administered by the juvenile, with appropriate consent from the juvenile, parent, and referring authority. Although the National Task Force advised against electric shock, it did not elaborate as to why adequate safeguards cannot be effectively applied.
Some treatment approaches that have been used with sex offenders, such as masturbatory satiation, are designed to render deviant fantasies or thoughts boring through repetition (Becker and Kaplan, 1993). Masturbatory conditioning, however, has presented practical as well as ethical concerns, because the approach requires asking the juvenile to masturbate, and may include masturbating to deviant stimuli with the goal of ultimately reducing such arousal (Becker and Kaplan, 1993; Bourke and Donohue, 1996; Morenz and Becker, 1995). Furthermore, as Hunter and Lexier (1998) observed, empirical findings concerning the effectiveness of any arousal conditioning approach are confounded by the inclusion of these approaches as part of a comprehensive treatment program. Consequently, Hunter and Lexier concluded that very little is known about the effectiveness of these approaches for reducing deviant arousal or about the types of juveniles for whom they may be most effective.
Vicarious sensitization (VS) is a relatively new treatment technique that may avoid some of the ethical concerns presented by other approaches. VS is a form of aversive conditioning that pairs deviant arousal with negative experiences. It involves exposing the juvenile to audiotaped crime scenarios designed to stimulate arousal and then, immediately afterwards, showing an aversive video that presents the negative social, emotional, physical, and legal consequences of sexually abusive behavior. Weinrott, Riggan, and Frothingham (1997) reported a study comparing a group of juvenile sex offenders who were administered a course of VS with a group who were on a waiting list but who had not yet received VS. Both groups received standard cognitive therapy during the study period. Phallometric assessment and self-report measures at 3 months revealed significantly reduced deviant arousal for the juveniles who had received VS. Furthermore, although the juveniles on the waiting list did not improve during the study period, they evidenced improvement after they received VS treatment. Although noting the limitations of a 3-month followup period, Weinrott and colleagues described VS as a technique that, used in conjunction with specialized cognitive therapy, may be an effective approach for reducing deviant arousal in juveniles who are sexually aroused by prepubescent children. As in all areas of sex offender treatment, additional research is needed to assess the effectiveness of this approach, including its long-term effectiveness.
Rasmussen (1999) argued that adequate family support can help reduce recidivism and that treatment programs that involve families are likely to be more effective than those that do not. As Gray and Pithers (1993) observed, however, families vary in terms of their motivation and ability to effectively facilitate their child's treatment. Gray and Pithers described strategies that can engage the cooperation of family members and reported that parents found the following approaches useful:
(1) written information on relapse prevention, cognitive distortions, and the consequences of sexual abuse; (2) educational videotapes of adolescent abusers discussing their relapse process and the need to be held accountable; (3) literature on the recovery process of sexual abuse victims; (4) referrals to treatment groups for adult survivors of sexual abuse; (5) the opportunity to be included periodically in sessions of the adolescent abuser group; (6) support groups for parents of abusive adolescents; and (7) attention to the concerns of the juvenile's siblings in the treatment process. (p. 314)
Lee and Olender (1992) described a teaching-family model of community-based residential treatment that has been used with juvenile sex offenders and other children. They noted that this specialized foster care approach can be very restrictive and can provide intensive treatment in a more homelike atmosphere, depending on the program components required for a particular child at a particular time. Through this approach, juveniles receive interventions in a more naturalistic setting, enabling them to acquire and practice prosocial life skills in situations similar to everyday life. The approach focuses directly on behaviors and uses a systematic reward program (a token economy) to enhance positive motivation. It also uses cognitive-behavioral approaches to facilitate behaviors such as impulse control, effective problem solving, moral and ethical decisionmaking, and so on. Foster parents trained to be "teaching parents" use techniques that have been researched and found useful for managing intense and emotionally volatile behaviors, and they use a curriculum to facilitate skills necessary for social competence and independent living. Foster parents are provided with support services, and juveniles participate in group counseling interventions. Lee and Olender reported that initial implementation research, conducted as part of the Ohio Youth Services Network's evaluation of sex offender treatment programs throughout the State, found that the program provided "high quality, appropriate care of adolescent offenders and emotionally disturbed youth" (p. 74). Outcome research is under way.
Using a Relapse Prevention Model
Relapse prevention initially was designed to help substance abusers prevent reoccurrence of substance-abusing behavior. Then Pithers, Marques, Gibat, and Marlatt (as cited in Barbaree and Cortoni, 1993) applied relapse prevention to adult sex offenders to reduce sexual reoffending. Gray and Pithers (1993) applied relapse prevention to the treatment and supervision of children and adolescents with sexual behavior problems.
Relapse prevention requires that juveniles learn to identify factors associated with an increased risk of sex offending and use strategies to avoid high-risk situations or effectively manage them when they occur. Gray and Pithers (1993) noted, however:
A high degree of motivation and integrity is required for a client to continually monitor signs of his relapse process and to invoke coping strategies, even when it feels like a sacrifice to do so. Without the dedication derived from the empathy for sexual abuse victims developed in treatment, RP [relapse prevention] risks becoming an intellectual exercise that educates offenders about what they need to do to avoid reoffending but that finds offenders lacking the motivation to use this knowledge. (p. 299)
When relapse prevention is applied to children, greater emphasis is placed on external supervision to prevent further victimization (Gray and Pithers, 1993). The relapse prevention approach is theoretically sound; however, as with other components of treatment for juveniles who have sexually offended, empirical studies investigating the effectiveness of this approach are lacking.
Some of the interventions described above appear appropriate for some juveniles who have committed sex offenses, but others do not. Furthermore, many of the target areas described above are relevant not only for sex offenders but also for juveniles who commit other types of offenses. In view of the many studies identifying general delinquency and antisocial attitudes and behavior among juveniles who exhibit sexual behavior problems, Weinrott (1998a) suggested that relevant empirically based treatment interventions for juvenile delinquents be used with those who commit sex offenses, whenever the interventions are indicated. Similarly, as Rasmussen (1999) stressed, "Treatment programs should be structured to address the factors that contribute to and maintain all criminal behavior, not just sexual offending" (p. 81).
Prentky (1995, 1997) presented potential target areas of treatment for juvenile and adult sex offenders and corresponding modalities for intervention. Although most of the modes of treatment presented by Prentky are discussed in this literature review, a few have not been mentioned, such as childhood victim survivors' group therapy and expressive therapy.
Table 3 provides a guide to treatment intended to reduce offending behaviors. It incorporates Prentky's (1995, 1997) work, treatment and modalities discussed in this literature review, and the authors' clinical understanding of these issues. The table presents clinical interventions but does not cover other strategies such as supporting appropriate academic placements, school attendance, and vocational training. As emphasized throughout this literature review, individualized assessment should guide the development of an appropriate treatment plan for each individual; the information presented in table 3 should not be rigidly applied.
Specialized Treatment for Juveniles Who Have Committed Sex Offenses
In spite of the proliferation of programs specifically designed for juvenile sex offenders, evaluation of these specialized approaches has been limited. For example, as Weinrott (1996) observed, most sex offender treatment programs have learning about the "sexual assault cycle" at their core. The cycle is used to help juveniles conceptualize their offending behaviors, including the associated feelings and distorted thinking that contribute to and follow their abusive acts. Becker (1998) described the cycle concept that was developed by Ryan, Lane, Davis, and Isaac (as cited in Becker, 1998). The concept is based on the premise that offending is preceded by a negative self-image that contributes to negative coping strategies when the juvenile anticipates negative responses from others, perceives such responses, or both. To avoid such negative anticipated or perceived reactions, the juvenile withdraws, becomes socially isolated, and fantasizes to compensate for resulting feelings of powerlessness and a lack of control. This process culminates in the sex offense, which results in more negative experiences, more feelings of rejection, and an increasingly negative self-image; and the cycle continues.
Weinrott (1996) pointed out that in spite of the fact that the sexual assault cycle has been used in sex offender treatment for nearly 20 years, this model has not been empirically validated. Furthermore, as Weinrott noted, although the cycle may fit many juveniles who have committed sex offenses, it does not explain the abusive behavior of all such offenders, including those described as "naive experimenters," those who desist from their abusive behavior, those who perpetrate sex offenses as part of a group, and those whose sexual behavior may be a result of significant psychopathology or deviant sexual arousal.
In their editorial, "Don't Shoot, We're Your Children: Have We Gone Too Far in Our Response to Adolescent Sexual Abusers and Children With Sexual Behavior Problems?" Chaffin and Bonner (1998) cautioned against the "conviction" that those working in the field have found the right approach. They wrote that such "dogma" might include the following beliefs:
That sex offender-specific treatment is the only acceptable and effective approach and that all teens and children who have performed inappropriate sexual behaviors must receive it; that a history of personal victimization is usually present, is a direct cause of abusive sexual behaviors, and must be a focus of treatment; that denial must be broken; that hard, in-your-face confrontation is synonymous with good therapy; that treatment must be long term and involve highly restrictive conditions; that deviant arousal, deviant fantasies, grooming [of victims] and deceit are intrinsic features; that parents and families of offenders are generally dysfunctional; that long-term residential placement is commonly required; that behaviors always involve an offense cycle or pattern that must be identified; that these teenagers and their parents must face the fact that they have a compulsive, incurable, life-long disorder; and that these youngsters are such dangerous predatory criminals that neighborhoods must be notified of their presence. Despite their wide acceptance, it is our opinion that clear, empirical scientific support for each and every one of these conventional wisdoms is either minimal or nonexistent. (p. 314)
Chaffin and Bonner (1998) reported that they knew juveniles who felt required to "confess" to sex offenses they did not commit and to deviant fantasies they did not have, because they thought they would be discharged from the treatment program if they did not comply. The authors also expressed concern that "overly broad applications" of fantasy journals, addiction/compulsion programs, shaming approaches, and programs that aggressively encourage empathy with victims could negatively affect these juveniles. Chaffin and Bonner further pointed out that although rates of detected sexual reoffenses appear relatively low (around 5 to 15 percent), the lack of untreated comparison groups prevents us from knowing whether treatment has been effective. In fact, they stressed, "Empirically, we cannot say whether treatment helps, hurts, or makes no difference" (p. 316). Chaffin and Bonner's views are consistent with Weinrott's (1996), who stated:
The prevailing view is that early clinical intervention is needed to break the cycle of sexual deviance, and that intervention should take the form of lengthy, offense-specific, peer-group therapy. There is not a shred of scientific evidence to support this stance. (p. 85)
Chaffin and Bonner (1998) and Weinrott (1996) have observed that at this point, it is not possible to say whether one type of treatment is better than another, with the possible exception of delinquency-focused multisystemic treatment, which appears to be more effective than individual counseling with juveniles who have committed sex offenses. Furthermore, as Weinrott noted, there also is no evidence to support a "heavy handed" correctional or justice response.
A study that appears to raise questions about the efficacy of specialized treatment for juveniles who have committed sex offenses was conducted by Lab, Shields, and Schondel (1993). The researchers compared the recidivism rates for juveniles treated in a specialized sex offender treatment program with rates for juveniles referred to community-based treatment programs generally lacking specialized programs for sex offenders. The study found that recidivism rates for both groups were low and that the outcome for juveniles treated in the sex-offense-specific program was no better than that for those treated in non-offense-specific programs. Like Chaffin and Bonner (1998), Lab, Shields, and Schondel concluded, "These results suggest that the growth of interventions has proceeded without adequate knowledge of how to identify at-risk youth, the causes of the behavior, and the most appropriate treatment for juvenile sex offending" (p. 543). Methodological problems, however, may have compromised the utility of this study (Weinrott, 1996).
In contrast to the Lab, Shields, and Schondel (1993) findings are results from a study by Kimball and Guarino-Ghezzi (1996), who compared 75 juvenile sex offenders treated in sex-offense-specific programs with sex offenders treated in non-offense-specific programs. Although placement in treatment programs was not randomized, juveniles did not vary significantly on their prior record, previous sexual deviance, or exposure to sexual, physical, and substance abuse. Findings revealed that juveniles placed in sex offender treatment demonstrated more positive attitudes and greater skill acquisition than those in nonspecific treatment. They were more likely to accept full responsibility for their offenses, to express remorse related to victim impact, and to articulate practical relapse prevention concepts and strategies. They also were significantly more successful in completing their first aftercare placements (70.6 percent, versus 41.2 percent for nonspecific treatment placements). At the time of this report, followup results were limited to 6 months. Findings suggested that participation in sex offender treatment contributed to lower rates of reoffending.
Although the Kimball and Guarino-Ghezzi (1996) treatment outcome results appear encouraging, these findings are tempered by other findings indicating that participants in the sex offender programs received more treatment than those in nonspecific programs. Those placed in sex-offense-specific treatment programs received significantly more therapy sessions, including group sessions that focused specifically on offending behavior. They also received significantly more family therapy (51.8 percent, versus 30.8 percent for those in nonspecific treatment). In addition, they received more treatment for nonsexual factors contributing to their sex offending; such treatment included family therapy, interpersonal skills training, stress and anxiety management, and relapse prevention. They also remained in treatment for significantly longer periods than those who received nonspecific programming (an average of 15.7 months, versus 7.1 months in nonspecific treatment). Thus, it is unclear whether a non-offense-specific treatment program comparable to a sex offender treatment program in terms of intensity and breadth of services would yield outcome results comparable to those of the Kimball and Guarino-Ghezzi study, especially for offenders who do not evidence patterns of deviant arousal.
In another study, Becker (as cited in Weinrott, 1996) described the effectiveness of cognitive-behavioral treatment used with a sample of juveniles who abused children younger than themselves. In addition to psychoeducational and cognitive approaches, this treatment used interpersonal skills training and behavioral interventions to reduce deviant arousal. Results indicated a 10-percent recidivism rate for sex offending. This finding, however, was based only on juveniles who completed the program. Furthermore, the followup period was relatively short (1 year), and no control group was used.
Weinrott (1998a) noted that in spite of the limited treatment research, empirically based approaches should be emphasized in the treatment of juvenile sex offenders. For example, he encouraged practitioners to provide juvenile sex offenders who engage in various types of delinquent behaviors with empirically based treatment approaches that have been designed specifically for delinquent populations.
Treatment for Juveniles Who Are Delinquent
The following studies describe research that has assessed the effectiveness of interventions with juveniles who commit various types of offenses, not just sex offenses. As previously noted, because general delinquency and antisocial attitudes and behavior are frequently found in juveniles who have committed sex offenses, these treatment approaches may be relevant and effective with these juveniles.
Izzo and Ross (1990) conducted a meta-analysis of rehabilitation programs designed for all juvenile delinquents, not just those who have committed sex offenses. Their findings suggested that programs based on cognitive therapy were twice as effective as those using other approaches. They defined cognitive therapy as approaches that employed one or more of six intervention modalities: problem solving, negotiation skills training, interpersonal skills training, rational-emotive therapy, role playing and modeling, and cognitive behavior modification.
More recently, Lipsey and Wilson (1998) conducted a meta-analysis of 200 experimental or quasi-experimental studies to assess the effectiveness of treatment interventions used with juvenile offenders. Because of variability between treatment approaches and sample characteristics, the findings from this meta-analysis require further study and should in the meantime be considered to be suggestive only. The findings are, however, consistent with reviews of the literature (Tolan and Guerra, 1994) and previous meta-analytic results (Lipsey, as cited in Lipsey and Wilson, 1998).
In sum, Lipsey and Wilson (1998) found that among noninstitutionalized juveniles, treatments that focused on interpersonal skills (e.g., social skills training, anger management, and moral education) and used behavioral programs consistently yielded positive effects. Contrary to findings from previous studies (Lipsey, as cited in Lipsey and Wilson, 1998; Tolan and Guerra, 1994), individual counseling also showed consistent positive effects. This surprising finding may be a result of the types of interventions that this study considered as "individual counseling." One approach involved a 12-week reality therapy program that emphasized client accountability and responsibility, behavior assessments, and action plans.
The other treatment that was included in the "individual counseling" category was multisystemic therapy (MST) (Lipsey and Wilson, 1998). MST is an empirically based intervention that has been validated with chronic juvenile delinquents and substance-abusing juveniles (Henggeler et al., 1998). It is also the only approach that has been empirically validated as effective with juvenile sex offenders (Borduin et al., 1990), although the sample size in the validation study was small and the comparison treatment did not involve current treatment approaches. MST confronts antisocial behavior in juveniles by targeting their "social-ecological context" (i.e., their family, neighborhood, school, and community) (Henggeler et al., 1998). Thus, although interventions may (or may not) involve individual interventions with the juveniles, this approach certainly cannot be considered individual counseling in the usual sense. MST individual interventions, for example, may involve parent figures, with or without the juvenile present. The importance of MST for juvenile sex offenders has increasingly been noted (Bourke and Donohue, 1996; Cellini, 1995; Swenson et al., 1998).
Results of the meta-analysis (Lipsey and Wilson, 1998) further indicated that other interventions with noninstitutionalized juvenile offenders have shown positive but less consistent evidence of effectiveness. These interventions include programs that provide multiple services (e.g., vocational training, skills-oriented education, job placement, community supervision) and those that require restitution or supervision through probation and parole. Mixed but generally positive effects were found in some studies for the following interventions: employment-related services, academic programming, advocacy and casework approaches, and family and group counseling. In contrast, weak or no effects were consistently found for early release, deterrence, vocational, and wilderness/challenge programs.
Findings regarding institutionalized juvenile offenders indicated consistent, positive effects for programs that focused on interpersonal skills (Lipsey and Wilson, 1998). One treatment approach, the teaching-family home model mentioned previously (Lee and Olender, 1992; Lipsey and Wilson, 1998), involves juveniles who frequently are referred from detention facilities or placements more restrictive than foster care. The teaching-family home model uses "teaching parents" to help juveniles develop necessary life skills and enhance social competence.
Positive but less consistent results were found for behavioral programs, community residential approaches, and programs that provided multiple services. Inconsistent evidence of mixed but generally positive effects was found for individual counseling, guided group interventions, and group counseling. In contrast, weak or no effects were consistently found for milieu therapy approaches.
Lipsey and Wilson (1998) noted that the impact of the most effective treatments on recidivism was substantial:
The most effective treatment types had an impact on recidivism that was equivalent to reducing a .50 control group baseline to around .30. In other words, we estimated that without treatment the recidivism would have been 50%. If they received the most effective of the treatments reviewed in this meta-analysis, their recidivism would have dropped to about 30%. (p. 333)
Chamberlain and Reid (1998) contrasted traditional community group placements with multidimensional treatment foster care (MTFC) to investigate an alternative to MST for juveniles whose parents were unable, for various reasons, to provide the "corrective or therapeutic parenting" the juveniles needed. Like MST, MTFC involves multiple treatment modes and targets, including individual therapy, family therapy, and interventions at home, at school, and among peer groups. In the Chamberlain and Reid study, chronic juvenile delinquents, including some juvenile sex offenders, were randomly assigned to either MTFC or traditional community-based group care settings. Results indicated that juveniles in MTFC had significantly fewer justice system referrals and returned home to relatives more often than those in community-based group care settings. Multiple regression analysis showed that assignment to the MTFC treatment condition was a better predictor of reduced offense rates (official and self-reported) than other well-known predictors.
Attrition From Sex-Offense-Specific Treatment
Studies of treatment programs for juveniles who have sexually offended have demonstrated high rates of treatment dropouts. For example, Becker (1990) found that only 27.3 percent of her sample attended 70 to 100 percent of scheduled therapy sessions and only 45.4 percent completed at least half of the sessions. Kraemer, Salisbury, and Spielman (1998) reported that completion rates for residential juvenile sex offender programs in Minnesota appeared to range from 30 to 50 percent. Their study suggested that older age and impulsivity were associated with treatment dropout. Rasmussen (1999) found that only half of the subjects in her sample completed the initial stage of their treatment and one-third failed to complete the full course of treatment once they began. (The remaining subjects either were not referred for treatment or did not follow through on the referral.) Schram, Milloy, and Rowe (1991) found that most offenders terminated treatment as soon as their sentence or court order ended. Only 39 percent of their sample completed treatment.
Similarly, Hunter and Figueredo (1999) reported that more than 50 percent of the subjects in their sample terminated or were terminated from treatment during the first year. Although 20 percent of these juveniles ended treatment for reasons unrelated to their behavior or attitudes (e.g., family relocation), 33 percent were expelled from the program as treatment failures. Of the "treatment failures," more than 75 percent were terminated because they were noncompliant with attendance and therapeutic directives. Only 11.4 percent of the "treatment failures" were terminated because of recidivism (4.9 percent for sex offenses, 6.6 percent for other types of offenses).
In another study, O'Brien (as cited in Weinrott, 1996) found that only 6 percent of 200 juvenile sex offenders who completed a treatment program committed another sex offense after being referred to the program. Although the study did not provide information about the number of juveniles who dropped out of treatment prematurely, the researchers did note that half of the juveniles who reoffended did so before they completed the treatment program.
High rates of treatment attrition are extremely important. Studies with juvenile sex offenders (Hunter and Figueredo, 1999) and adult sex offenders (Becker and Hunter, as cited in Rasmussen, 1999; Hanson and Buissièbuire, 1998; Marques, Day, Nelson, and West, as cited in Hunter and Figueredo, 1999; Marshall et al., as cited in Rasmussen, 1999) suggest that failing to complete treatment is associated with higher rates of recidivism for both sex offenses and other types of offenses.
Segregating Versus Integrating Juveniles Who Have Committed Sex Offenses
Historically, treating juveniles who have committed sex offenses in a setting specifically designed for sex offenders has been considered "optimal" (Morenz and Becker, 1995). The literature, however, indicates that the effectiveness of this approach has not been proven. In fact, as some of the studies reviewed above suggest, other approaches (e.g., MST and MTFC) may be more beneficial. Noting the absence of significant differences between groups of juvenile sex offenders and other juvenile offenders in the research that they and others have conducted, Jacobs, Kennedy, and Meyer (1997) concluded, "The similarities are indicative of commensurate therapeutic needs for both types of offenders" (p. 201).
Milloy (1994) asked the question, "But what is specialized sex offender treatment?" She pointed out that "specialized" treatment for sex offenders typically includes components such as sex education, social skills, anger management, acceptance of responsibility for one's offenses, and empathy for victims. Yet these components may be appropriate for juvenile offenders in general.
As Milloy (1994) pointed out, one of the arguments in favor of specialized and segregated sex offender treatment programs is that these offenders frequently intimidate staff and other residents through their manipulative or aggressive behaviors. The results of Milloy's study, however, suggested that juveniles who committed sex offenses were not more likely to be exploitative, manipulative, or aggressive than juveniles who committed other types of offenses. The frequency of verbal and physical threats did not differ between the groups, and the sex offenders did not present increased management risks or security risks within the institution.
In conclusion, Milloy (1994) indicated that no controlled studies have been published investigating the effect of segregating juvenile sex offenders from the general delinquent population. She stated, "This fact, coupled with the findings from this study, suggest that the segregation of juvenile sex offenders is a costly approach whose worth is unproven" (p. 10).
Whether juveniles who have been sexually abusive should be grouped with juveniles who have committed nonsexual offenses or with juveniles who have other behavioral problems is a complex issue. Clearly, other factors must be considered when designing appropriate treatments and treatment settings. Among these factors is the safety of all juveniles involved, since the juveniles who have committed sex offenses might become targets themselves or might target others.
Another factor cited as supportive of segregated treatment units is the reduction of staff training needs that results when intensive training in sex-offending issues is provided only to those whose jobs involve this specialized treatment (Bengis, 1997). Other arguments in favor of segregated units include the possibility that such units may form stronger and more effective treatment cultures (Bengis). On the other hand, research has suggested that delinquent peer group association may increase risk (Ageton, as cited in Prentky et al., 2000). Controlled studies using random assignment to comparison groups are necessary to help resolve the issue of whether juveniles who have committed sex offenses should be segregated from other juveniles in residential care.
In the meantime, the importance of individualized assessment and treatment planning cannot be overemphasized. As Kavoussi, Kaplan, and Becker (1988) point out, the heterogeneity of juvenile sex offenders "suggests that no single treatment regime will be effective in all cases" (p. 243). Furthermore, a one-size-fits-all approach can be costly and may be harmful to the juveniles and their families (Becker, 1998). As Chaffin and Bonner (1998) point out,
[P]erhaps it is time to emphasize some flexibility and compassion in which treatments we choose and to which individual youngsters we apply them and to realize that individual need, not dogma, should dictate what must be accomplished (p. 316).
Facilitating Safety in Residential Treatment Settings
The issue of community safety exists regardless of whether a juvenile sex offender remains in the community or is placed in a segregated or unsegregated residential facility. NAPN (1993) provided specific recommendations to facilitate safety in residential treatment facilities. These recommendations suggest that such programs should ensure the following:
Other recommendations from the National Task Force on Juvenile Sexual Offending (as cited in NAPN, 1993) include having clear rules about personal space boundaries and touching. Recommendations also include having night staff who remain awake and monitor residents both randomly and at frequent, planned intervals throughout the night. Ross and Villier (1993) provided more detailed recommendations related to screening program applicants, selecting staff, and designing and supervising living units in a way that maximizes the safety of residents and staff.
Treatment of Young and Preadolescent Children With Sexual Behavior Problems
Gray (as cited in Araji, 1997) proposed that treatment goals that balance community safety and the promotion of developmentally appropriate competencies are most effective in treating children with sexual behavior problems. More specifically, Gray and Pithers (1993) suggested that abusive behaviors might be most effectively addressed by targeting risk factors that predispose a child to sexual behavior problems or that precipitate or perpetuate the problems. Gray and Pithers proposed the following approaches:
Another component of treatment for children with sexual behavior problems is the "prevention team" (e.g., selected family members, care providers, and community advocates) (Gray and Pithers, 1993). The prevention team is of primary importance when intervening with young children who do not have the developmental capacity for self-monitoring.
According to Johnson (1991), interventions with children who are sexually abusive and aggressive should involve reporting the sexual behaviors to appropriate agencies, such as protective services and the police; working with appropriate agencies to ensure the safety of the victim, potential victims, and the abusing child; and working with the courts responsible for juveniles. Johnson observed that interventions should begin with an assessment of the child's treatment needs, to facilitate appropriate placement and treatment.
In her book, Araji (1997) described 10 treatment programs and practices for children with sexual behavior problems, including Johnson's (as cited in Araji). Araji identified these programs by reviewing the professional literature and attending workshops and through personal correspondence. She stated that the programs described simply represent current treatment efforts and trends, noting that the effectiveness of most of the programs has not been demonstrated. The federally funded work of Pithers et al. (1998a, 1998b), as described below, and the ongoing work of Bonner, Walker, and Berliner (as described in Araji, 1997) are important exceptions.
Most of the programs reviewed by Araji use theories from the sex abuse literature. Some appear to emphasize personal histories of sexual abuse as a factor contributing to sexual behavior problems in children, although the literature suggests that this issue may be overemphasized in the context of multiple risk factors.
The programs reviewed by Araji also used child development literature and designed interventions that were appropriate to different ages and cognitive and developmental levels. Programs typically targeted prevention of perpetration. Techniques frequently involved modifications of approaches used with adults or adolescents who commit sex offenses, such as the relapse prevention and assault cycle approaches discussed previously.
All of the programs reviewed used a cognitive-behavioral approach, although some also used other orientations, such as those based on psychodynamic and attachment theories. Cognitive-behavioral interventions included skill development to promote prosocial coping and problem solving, age-appropriate interpersonal relationships and sexual behaviors, and abuse prevention strategies. In her review, Araji noted that because no treatment approach has been demonstrated to be superior to others, treatment that combines theories and methods might better meet the needs of these children and their families.
Treatment modalities in the programs reviewed by Araji include individual, group, pair, and family therapy. Most providers appeared to prefer group therapies. Araji's views appear consistent with those of Johnson (1991), who stated: "The group format allows the therapists to use the group members to help each other understand and work on the 'touching' problems. The aim is to help the children interact without being sexually or behaviorally inappropriate" (p. 11). Araji also noted that groups can help reduce children's social isolation and are efficient in terms of cost and time. Others (Friedrich and Gil, as cited in Araji, 1997) consider pair therapy (two children treated together) more beneficial. Advocates of the pair therapy approach believe that it may minimize anxieties, avoid rejections, and enhance controlled peer interactions.
Developmental issues. Other factors considered of great importance when intervening with children who have been sexually abusive are developmental issues. As Friedrich (as cited in Araji, 1997) noted, substantial differences may exist between a 6-year-old child who has been sexually aggressive and a 10-year-old child who has been sexually aggressive. Even if the acts appear similar, differences may include the meaning the child attributes to the act, differences in peer relationships, and other factors (including, for a child who has been the victim of sexual abuse, the length of time between the victimization and the child's abusive behavior).
Friedrich (as cited in Araji, 1997) also argued that sexual aggression in children reflects difficulties with a child's ability to modulate emotions and behavior. Sexual aggression is considered to be similar to other behavioral and psychological problems or disorders, such as fire setting, stealing, and posttraumatic stress disorders. Interventions found effective with these other forms of behavioral and emotional dysregulationsuch as increasing parental supervision and positive interactions with parentscan be valuable for children who have been sexually abusive and may be sufficient for eliminating such behaviors in some children. Friedrich also argued that when children have suffered traumas, the underlying issues that may have resulted require intervention if positive, lasting changes are to be achieved.
Family involvement. Although the programs reviewed by Araji (1997) varied in terms of the range of interventions they provided to parents or other caregivers, all of them involved parents or other caregivers, either in group interventions or through other approaches. Treatment goals with caregivers typically included improving parental supervision and parenting skills and increasing parental knowledge about sex abuse; in some programs, treatment goals also included providing specific training to help parents help their children succeed at relapse prevention. In view of the high levels of stress, personal and interpersonal difficulties, and impaired parent-child attachments found in their study (as described previously, in the section on "Young Children Who Have Committed Sex Offenses: Family Characteristics"), Pithers et al. (1998a) noted the need for group treatment for parents of children with sexual behavior problems. Pithers et al. suggested that such groups address issues of parental attachment, parental training, social-relational skills, trauma resolution, and, when indicated, the opportunity to grieve the loss of an idealized child and family.
Specialized therapeutic foster homes have been developed in some areas to provide interventions for children who are sexually abusive and require out-of-home placement but not residential care. One small study (Ray et al., 1995) involved 15 children who came from chaotic, violent, and abusive homes and were placed in therapeutic foster homes. These youngsters typically were under 13 years old, but occasionally older children with cognitive difficulties were accepted into the program. Researchers found that the children evidenced improvements in behavior, emotional adjustment, social functioning, family relationships, and overall adjustment. Improvements in life skills were not statistically significant but appeared to be moving in the expected direction. Although four of the children displayed inappropriate sexual behavior early in treatment, none of the children continued to do so at the completion of treatment. Followup interviews indicated that the children continued to have serious emotional and behavioral problems, but with the exception of two of the children, their sexually abusive behavior appeared to have subsided. This study is limited by its small sample size, lack of a comparison group, and other problems. In spite of these limitations, however, the advantages of foster care approaches in helping to stabilize a child and provide appropriate interventions warrant further study.
As Araji (1997) noted in her book, "Sexually abusing behavior by children is a complex phenomenon presented by multiproblem youth and, frequently, multiproblem families. . . . The programs, agencies, and practices reviewed all recognize the importance of developing individualized treatment plans" (p. 184). In addition, Araji noted that although a variety of interventions may be required, ranging from community-based approaches to residential care, "helping families to create safe, predictable, and growth promoting relationships among family members is key to helping the sexually reactive and sexually aggressive child" (p. 187).
A comparative study. As noted previously (in the section on "Young Children Who Have Committed Sex Offenses: Types and Classifications"), Pithers et al. (1998b) identified five subtypes of children with sexual behavior problems: sexually aggressive, nonsymptomatic, highly traumatized, abusive reactive, and rule breaker. Their investigations also revealed some differences in how children in various subtype classifications responded to different types of treatment.
At intake, the children and their families were randomly assigned to one of two 32-week treatment conditions. One treatment involved expressive therapy, reportedly recommended by some national experts as the treatment of choice for children with behavioral problems. The other treatment was a substantially modified form of relapse prevention. Both approaches involved parents in parallel group interventions. The Child Sexual Behavior Inventory-3 (CSBI-3) was used to measure progress.
Results indicated that children in most of the subtypes evidenced similar degrees of change regardless of treatment modality. The highly traumatized children, however, benefited significantly more from modified relapse prevention than from expressive therapy. In fact, highly traumatized children who were in expressive therapy actually evidenced a slight increase in sexualized behavior. The number of children classified as sexually aggressive evidencing a reduction in sexual behavior problems was slightly larger in expressive therapy than in modified relapse prevention therapy, but this finding was tempered by the fact that a similar number of children in expressive therapy who were classified as sexually aggressive had an increase in sexual behavior problems.
Results further indicated that children in some subtypes responded well to treatment, whereas those in other subtypes did not. For example, more than half of the highly traumatized children evidenced significant reductions in problematic sexual behavior after the first 16 weeks of treatment. In contrast, only 7 percent of the sexually aggressive children demonstrated significant decreases in their sexual behavior problems.
Treatment of Juveniles With Cognitive or Developmental Disabilities
Special interventions may be necessary for juveniles with intellectual and cognitive impairments. Furthermore, these juveniles may be difficult to engage in standard treatment approaches. Langevin, Marentette, and Rosati (1996) proposed that learning difficulties may affect therapy in at least two ways. First, during therapy sessions, a person with learning difficulties may not be able to process the same information that a person of average intellectual abilities could. Second, individuals with learning difficulties may have developed negative attitudes toward learning situations and, "in particular, avoid classroom type experiences where they may have met failure and derision from other students" (p. 145). As a result, these individuals may prefer to avoid therapeutic situations that resemble their negative experiences, such as psychoeducational programs and other cognitive-behavioral approaches.
Langevin, Marentette, and Rosati (1996) found some support for these theories in their study of adult sex offenders. Although they did not find that the subjects' attitudes toward therapy were significantly related to education or level of intelligence, they did find a negative correlation between attitude and Halstead Reitan Impairment Index scores. In other words, individuals who evidenced significant neuropsychological impairment on the Halstead Reitan Index evidenced more negative attitudes toward therapy.
A review of the literature (Stermac and Sheridan, 1993) regarding treatment of "developmentally disabled" adults and adolescents revealed a "dearth of work in this area" (p. 237). Most studies have focused on adult offenders and have stressed behaviorally oriented interventions. Pharmacological approaches also have been used with developmentally disabled sex offenders. As noted previously (in the section on "Treatment Approaches: Overview"), sex-drive reducing medications such as medroxyprogesterone can be effective in reducing sex offending, but because of potentially serious side effects and ethical concerns, the use of these medications for juveniles requires appropriate informed consent from guardians; additionally, the appropriateness of these medications for juveniles who have committed sex offenses has been questioned (Hunter and Lexier, 1998).
Most interventions involving adolescents with developmental disabilities who have committed sex offenses have used approaches modified from adult sex offender treatment programs (Stermac and Sheridan, 1993). Strategies to enhance learning and generalizing skills and coping strategies are recommended. Modified relapse prevention strategies have been found to be effective with some cognitively impaired sex offenders. Yet, as Stermac and Sheridan (1993) pointed out, relapse prevention emphasizes self-management and therefore may not be appropriate for all intellectually or cognitively impaired sex offenders.
Langevin, Marentette, and Rosati (1996) urged treatment professionals to reach out to these juveniles. They suggested the following steps:
Research concerning intellectual, cognitive, and neurological impairments in juvenile sex offenders (previously discussed in the section on "Characteristics: Academic and Cognitive Functioning") also points to the necessity of developing individualized interventions that are tailored to the special needs of these juveniles. Although a more indepth discussion of specialized interventions with juveniles who have intellectual, cognitive, and neurological problems is beyond the scope of this Report, Ferrara and McDonald (1996) provide a detailed discussion of treatment strategies and techniques that may be useful. These authors draw on work from other related fields, such as the treatment of persons with brain injuries, and apply this knowledge to interventions designed for juveniles who are sexually aggressive. For example, Ferrara and McDonald describe techniques designed specifically to facilitate learning, promote attention and concentration, and improve recall. Treatment approaches described are multimodal, applied in multiple settings, and tailored to the juvenile's individual needs.
Individuals providing treatment for juveniles with sexual behavior problems must be personally and professionally qualified (Association for the Treatment of Sexual Abusers, 1997a; NAPN, 1993). Personal qualifications include being emotionally healthy, having respect for oneself and others, using good listening skills, and having the ability to empathize. Professional qualifications include relevant education, training, and experience. Treatment providers should receive training before they begin their interventions. Training should then take place on a continuing basis, so providers can stay current with this evolving field.
More specifically, Goocher (1994) stressed the importance of "adequate training in normal adolescent development, the etiology and behavior manifestations of psychiatric disorders, and how to reinforce initial efforts of young people to learn new patterns of behavior and to come to terms with their own personal histories" (p. 249). Goocher also recommended additional training in how to help juveniles develop adequate verbal and personal skills and problem-solving abilities.
To be effective, Friedrich (as cited in Araji, 1997) suggested that therapists who work with sexually aggressive children should receive good training in issues pertaining to victimization and the development of violence and aggression. Araji (1997) noted that therapists also must be well aware of normative childhood sexual behaviors. Furthermore, the importance of developmental issues regarding attachment and the capacity for moral reasoning, empathy, and autonomy cannot be ignored (Pithers, Kashima, Cummings, Beal, and Buell, as cited in Araji, 1997). These suggestions clearly are important for those who treat adolescents and those who treat younger children.
Working with juveniles who have sexual behavior problems is a challenging job. In addition to concerns about protecting community safety, providing sound treatment, and dealing with significant human suffering, individuals who work with these juveniles are exposed to a great deal of distorted thinking and deviant sexual behavior. As NAPN (1993) observed, "Systems must be aware of potential emotional/psychological impacts on providers and take steps to protect against or counter negative effects" (p. 46).
| Juveniles Who Have Sexually Offended