Chapter 5: Effectiveness of Treatment for Juveniles Who Sexually Offend

Introduction

Sex offenders have received considerable attention in recent years from both policymakers and the public. This is due at least in part to the profound impact that sex crimes have on victims and the larger community. While most perpetrators of sex crimes are adults, a significant percentage of sexual offenders are under age 18.

Given the prevalence of sexual offending by juveniles, and the potential links between sexually abusive behavior during adolescence or childhood and sexual offending later in life, therapeutic interventions for juveniles have become a staple of sex offender management practice in jurisdictions across the country. Indeed, the number of treatment programs for juveniles who commit a sexual offense has increased significantly over the past 30 years. Worling and Curwen (2000), for example, reported that only one specialized treatment program for adolescent sexual offenders was operating in the United States in 1975. By 1995, the number of programs serving juveniles had increased to more than 600. In 2008, more than one-half (699) of the approximately 1,300 sex-offender-specific treatment programs operating in the United States provided treatment services to juveniles. While most (494) of the juvenile programs in 2008 provided treatment services to adolescents, about 30 percent (205) provided services to children 11 years old and younger. Overall, adolescents accounted for about 23 percent and children 11 years old and younger accounted for about 3 percent of all clients (adult and juvenile) treated in sex-offender-specific treatment programs in the United States in 2008 (McGrath et al., 2010). 

Treatment approaches for juveniles who commit sexual offenses also have changed since the 1970s. For many years, treatment for juveniles was largely based on models used with adult sexual offenders. However, as knowledge about the developmental, motivational and behavioral differences between juvenile and adult sexual offenders has increased, therapeutic interventions for juveniles have become more responsive to the diversity of sexually abusive behaviors and the specific offending-related factors found among adolescents and children. 

"Juveniles who commit sexual offenses are diverse in terms of their offending behaviors and future public safety risk."

Juveniles who commit sexual offenses are clearly quite diverse in terms of their offending behaviors and future risk to public safety. In fact, they appear to have far more in common with other juvenile delinquents than they do with adult sexual offenders. This is a common theme in the literature, and the diversity found in the offending behavior and risk levels of juveniles who commit sexual offenses, as well as the dissimilarity that exists between juveniles who commit sexual offenses and their adult counterparts, were both acknowledged by the experts at the SOMAPI forum. Juveniles are generally more impulsive and less aware of the consequences of their behavior than adults. And while a few sexually abusive behaviors in youth are compulsive and reflective of a recurrent pattern of social deviance, others may be more isolated and not indicative of a long-term behavior pattern. Therapeutic interventions for juveniles are increasingly taking this diversity into account, along with family, peer, and other social correlates that are related to sexually abusive behavior in youth. Still, it appears that far more change is needed. As Letourneau and Borduin (2008, pp. 290–291) have pointed out:

Although the research literature reviewed earlier strongly indicates that sexually offending youths are influenced by multiple ecological systems, most current treatments focus heavily on presumed psychosocial deficits in the individual youth .... Another problem with the predominant approaches to treatment is the fact that many sexually offending youths desist from future offending (even in the absence of intervention).

While there is strong scientific evidence that therapeutic interventions work for criminal offenders overall, the effectiveness of treatment for sexual offenders–whether juveniles or adults–has been subject to considerable debate. Some people argue that treatment can be at least modestly effective. Others are uncertain or outright skeptical that sex offender treatment works. While inconsistent research findings and measurement shortcomings no doubt have contributed to the ongoing controversy, a body of scientific evidence has emerged in recent years suggesting that therapeutic interventions for juveniles who commit sexual offenses can and do work.

"Many sexually offending youth desist from future offending, even in the absence of intervention."

This chapter reviews the scientific evidence on the effectiveness of treatment for juveniles who commit sexual offenses. It was developed to support informed policy and program development at the federal, state, and local levels. The chapter summarizes what is scientifically known about the impact of treatment on the recidivism of juveniles who sexually offend. (For more information on "Recidivism of Juveniles Who Commit Sexual Offenses ,"see chapter 3 in the Juvenile section.) It presents key, up-to-date research findings from single studies of treatment effectiveness as well as from research that synthesizes information from multiple treatment effectiveness studies.

Issues To Consider

While there is growing interest in crime control strategies that are based on scientific evidence, determining what works is not an easy task. It is not uncommon for studies of the same phenomena to produce ambiguous or even conflicting results, and there are many examples of empirical evidence misleading crime control policy and practice because shortcomings in the quality of the research were overlooked (see, for example, Sherman, 2003, and McCord, 2003). The importance of basing conclusions about what works on highly trustworthy and credible evidence cannot be overstated, and both the quality and consistency of the research evidence has to be considered.

Single Studies

In the field of criminology, there is general agreement that certain types of single studies–namely, well-designed and executed experiments, or randomized controlled trials (RCTs)–provide the most trustworthy evidence about an intervention's effectiveness (Sherman et al., 1998; MacKenzie, 2006; Farrington & Welsh, 2007).¹

While RCTs are an important method for determining the effectiveness of an intervention, they can be difficult to implement in real-life settings. RCTs are expensive and require a level of organizational (and at times, community) cooperation that can be difficult to obtain.² In practice, various constraints can preclude an evaluator from using an RCT, and relatively few of these studies have been used in the assessment of sex offender treatment.

When RCTs cannot be used, researchers examining the effectiveness of an intervention typically employ the next best approach, a quasi-experiment. Many quasi-experiments are similar to RCTs, but they do not employ random assignment. These studies typically involve a comparison of outcomes–such as recidivism–observed for treatment participants and a comparison group of individuals who did not participate in treatment. In this approach, researchers try to ensure that the treated and comparison subjects are similar in all ways but one: participation in the treatment program.³ When treatment and comparison subjects are closely matched, the study can be capable of producing highly trustworthy findings. 

Synthesis Research: Systematic Reviews and Meta-Analysis

There also is agreement in the scientific community that single studies are rarely definitive (Lipsey, 2002; Petrosino & Lavenberg, 2007; Beech et al., 2007). Individual studies with seminal findings certainly do exist, but single studies–even RCTs–can produce misleading results (Lipsey, 2002). Hence, single-study findings must be replicated before meaningful conclusions can be made, and the effectiveness of an intervention can best be understood by examining findings from many different studies (Petticrew, 2007). Researchers typically accomplish this through synthesis research, such as a systematic review. A systematic review adheres to a preestablished protocol to locate, appraise, and synthesize information from all relevant scientific studies on a particular topic (Petrosino & Lavenberg, 2007). Methodological quality considerations are a standard feature of most systematic reviews today, and studies that fail to reach a specified standard of scientific rigor are typically excluded from the analysis.⁴

Systematic reviews are increasingly incorporating a statistical procedure called meta-analysis. In practice, meta-analysis combines the results of many evaluations into one large study with many subjects. This is important because single studies based on a small number of subjects can produce distorted findings about a program's effectiveness (Lipsey, 2002). By pooling the subjects from the original studies, meta-analysis counteracts a common methodological problem in evaluation research–small sample size–thereby helping the analyst draw more accurate and generalizable conclusions.⁵ In addition, meta-analysis focuses on the magnitude of effects found across studies rather than their statistical significance. Determining effect sizes is important because, as Lipsey (2002) points out, an outcome evaluation of an individual program "can easily fail to attain statistical significance for what are, nonetheless, meaningful program effects." Hence, effect size statistics provide the researcher with a more representative estimate of the intervention's effectiveness than estimates derived from any single study or from multistudy synthesis techniques that simply calculate the proportion of observed effects that are statistically significant. When systematic reviews and meta-analyses are done well, they provide the most trustworthy and credible evidence about an intervention's effectiveness.

Summary of Research Findings

Findings From Single Studies

Several single studies examining the effectiveness of treatment programs for juveniles who sexually offend have been undertaken in recent years, and these studies have consistently found at least modest treatment effects on both sexual and nonsexual recidivism. Worling and Curwen (2000), for example, used a quasi-experimental design to examine the effectiveness of a specialized community-based treatment program that provided therapeutic services to adolescents and children with sexual behavior problems and their families. While treatment plans were individually tailored for each offender and his or her family, cognitive-behavioral and relapse prevention strategies were used, and offenders typically were involved in concurrent group, individual, and family therapy. Recidivism rates were calculated using survival analysis for a treatment group consisting of 58 adolescents (53 males and 5 females) and a comparison group consisting of 90 adolescents (86 males and 4 females). Comparison group subjects consisted of three subgroups: juveniles who refused treatment, juveniles who received an assessment in the program only, and juveniles who dropped out of the community-based program prior to completing 12 months of treatment.⁶ To determine potential effects of group differences, the researchers also examined whether the treatment and comparison group subjects differed in any meaningful way on various factors related to recidivism (e.g., prior criminal history, offender demographics, victim characteristics); no significant differences between the treatment and comparison group subjects were found. 

Based on a 10-year followup period, Worling and Curwen (2000) found that the juveniles in the treatment group had significantly better outcomes than comparison group members on several measures of recidivism (see table 1).⁷ For example, the sexual recidivism rate was 5 percent for the treatment group compared to 18 percent for the combined comparison group. The recidivism rates for any offense were 35 percent for the treatment group and 54 percent for the combined comparison group. In fact, for every measure of recidivism employed in the study, the treatment group had lower recidivism rates than comparison group members who either refused treatment, received an assessment only, or dropped out of the program prior to completing 12 months of treatment.

In 2010, Worling, Litteljohn, and Bookalam reported findings from a followup analysis that extended the followup period for the original sample of study subjects to 20 years. Study subjects were, on average, 31.5 years old at the end of the 20-year followup period. The analysis demonstrated that the positive treatment effects originally observed by Worling and Curwen (2000) using a 10-year followup period had persisted over a longer period of time.

Table 1. Recidivism Rates for Treatment vs. Comparison Groups

Recidivism Measure	10-Year Recidivism Rate (%)		20-Year Recidivism Rate (%)
	Treatment Group (n=58)	Comparison Group (n=90)	Treatment Group (n=58)	Comparison Group (n=90)
Sexual charge	5*	18	9*	21
Nonsexual violent charge	19*	32	22*	39
Any charge	35**	54	38*	57
* p < .05. ** p < .01. Sources: Worling & Curwen, 2000; Worling, Litteljohn, & Bookalam, 2010.

The 2010 analysis by Worling, Littlejohn, and Bookalam mirrored Worling and Curwen's (2000) original investigation in the following ways. First, recidivism was examined using charges for sexual, nonsexual violent, nonviolent, and any new offense. Second, comparison group subjects consisted of three subgroups: juveniles who refused treatment, juveniles who received an assessment in the program only, and juveniles who dropped out of the community-based program prior to completing 12 months of treatment. Third, the researchers examined whether the treatment and comparison group subjects differed in any meaningful way on various factors related to recidivism, and no significant differences were found. Treatment and comparison group subjects were not significantly different in terms of personal characteristics, offense characteristics, or any of the assessment test scores examined (Worling, Littlejohn, & Bookalam, 2010). (For more information on the "Assessment of Risk for Sexual Reoffense in Juveniles Who Commit Sexual Offenses," see chapter 4 in the Juvenile section.)

Based on the 20-year followup period, Worling and his colleagues (2010) found that adolescents who participated in specialized treatment were significantly less likely than comparison group members to receive subsequent charges for sexual, nonsexual violent, nonviolent, or any crime (see table 1). Interestingly, the 20-year recidivism rates reflect only small increases over the 10-year recidivism rates reported by Worling and Curwen (2000). In discussing their findings, Worling and his colleagues (2010, p. 56) concluded:

The results of this investigation suggest that specialized treatment for adolescents who offend sexually leads to significant reductions in both sexual and nonsexual reoffending–even up to 20 years following the initial assessment .... The results of this investigation also support the finding that only a minority of adolescents who offend sexually are likely to be charged for sexual crimes by their late 20s or early 30s.

Another study that found positive treatment effects was conducted by Waite and colleagues (2005). The researchers examined treatment effectiveness using a sample of juveniles who had been incarcerated for sexual offenses. The study compared the recidivism outcomes of two groups. One consisted of juveniles who participated in an intensive sex offender treatment program in a specialized, self-contained living unit of the correctional facility. The other consisted of juveniles who received less intensive treatment and remained housed within the general population of the correctional facility. Several recidivism outcomes were examined using a 10-year followup period. While the study did not employ random assignment or an equivalent "no-treatment" comparison group, it is one of the few studies to examine treatment effectiveness for incarcerated juveniles who have committed sexual offenses. The researchers found that study subjects who participated in the more intensive, self-contained treatment program had lower recidivism rates for any crime (47 percent compared to 71 percent) and for nonsexual violent crime (31 percent compared to 47 percent) than the incarcerated juveniles who received less intensive treatment and who remained housed in the facility's general population. The sexual recidivism rates for the two groups, however, were not significantly different (about 5 percent for both the treatment and comparison groups).

Finally, Seabloom and colleagues (2003) examined the effects of a community-based treatment program for juveniles who sexually offend. Treatment was based on principles of sexual health and it involved individual, group, and family therapy. Based on an average followup period of about 18 years, the researchers found that treated juveniles had a lower sexual recidivism rate than untreated juveniles. Positive treatment effects also were reported by Wolk (2005). Based on a 3-year followup period, treated juveniles had a recidivism rate of 26 percent for any offense compared to a rate of 60 percent for untreated juveniles.

Although none of the evaluations referenced above randomly assigned study subjects to treatment and control conditions, a series of studies focusing on the use of multisystemic therapy (MST) with juveniles who sexually offend have employed an experimental–or RCT–design. MST is a community-based intervention that has been used with serious and chronic juvenile offenders in jurisdictions across the country. It was developed in the late 1970s based on the premise that individual, family, and environmental factors all play a role in shaping antisocial behavior. MST works within multiple systems (i.e., individual, family, school) to address the various causes of a child's delinquency (Henggeler, 1997), and it has been adapted to the special needs of juveniles who sexually offend (Letourneau et al., 2009).

While the effectiveness of MST with juvenile offenders in general has been documented both in individual studies and systematic reviews, research on its effectiveness with juveniles who commit a sexual offense is still emerging. The first study to examine the impact of MST on the recidivism of juveniles who sexually offend was conducted more than 20 years ago by Borduin and colleagues (1990). While the study employed random assignment, the sample size was very small. Only 16 adolescents (and their families) were randomly assigned to either home-based MST services or outpatient therapy. Based on a 3-year followup period, Borduin and his colleagues reported that the adolescents who received MST treatment had significantly lower sexual and nonsexual recidivism rates than their comparison group counterparts. MST-treated adolescents in the study had a sexual rearrest rate of 12.5 percent compared to a sexual rearrest rate of 75 percent for the comparison group subjects. The rearrest rates for nonsexual crimes were 25 percent for MST-treated adolescents and 50 percent for comparison group subjects.

More recently, Borduin, Schaeffer, and Heiblum (2009) examined the efficacy of MST with juveniles who sexually offend using a somewhat larger sample of 48 adolescents.⁸ Based on a followup period of 8.9 years,⁹ the researchers found significantly lower recidivism rates for juveniles who received MST treatment. The sexual recidivism rate was 8 percent for MST-treated subjects compared to 46 percent for the comparison group subjects. The nonsexual recidivism rate was 29 percent for MST-treated adolescents compared to 58 percent for comparison group subjects. MST-treated juveniles also spent 80 percent fewer days in detention facilities compared to their control group counterparts.

The most recent evaluation of MST's effectiveness with juveniles who sexually offend also employed an experimental design (Letourneau et al., 2009). As part of the study, Letourneau and her colleagues randomly assigned juveniles who sexually offend to MST treatment (n=67) or treatment as usual (n=60) conditions. Based on initial analyses using 1-year and 2-year followup periods, the researchers found that MST-treated youth had significantly lower rates of self-reported sexual behavior problems and delinquency and reduced risk of out-of-home placements compared to study subjects receiving treatment as usual (Letourneau et al., 2009; Swenson & Letourneau, 2011).

In summary, several single studies designed to evaluate the effectiveness of treatment for juveniles who commit a sexual offense have been conducted in recent years. While only a handful of these studies have employed an experimental design, a matched comparison group, or statistical control of factors that are linked to treatment effects, the weight of the available evidence–although it is far from definitive–suggests that treatment for juveniles who sexually offend can be effective. Studies employing an RCT design have demonstrated the efficacy of MST in reducing the recidivism of juveniles who commit sexual offenses. It should be noted, however, that these studies have been conducted by program developers and are based on samples that are relatively small in size. Independent evaluations that employ larger sample sizes should be undertaken to further establish the effectiveness and transportability of MST with juveniles who sexually offend. Nevertheless, MST was identified as an effective program in the 2011 National Criminal Justice Association (NCJA) survey.

"Rigorous studies have found that MST is effective in reducing the recidivism of juveniles who commit sexual offenses."

Recent research on other treatment approaches has also produced positive results. While it is difficult to isolate treatment effects and identify the specific treatment approaches that are most effective, interventions that address multiple spheres of juveniles' lives and that incorporate cognitive-behavioral techniques along with group therapy and family therapy appear to be most promising. However, there is a clear need for more high-quality research that can better demonstrate the effectiveness of various treatment approaches delivered in the community as well as in secure settings. Studies that employ random assignment or equivalent treatment and comparison group conditions–achieved through matching or statistical controls–are greatly needed.

Findings From Synthesis Research

One of the most frequently cited studies of the effectiveness of juvenile treatment was conducted by Reitzel and Carbonell (2006). Their meta-analysis included 9 studies and a combined sample of 2,986 juvenile subjects, making it one of the largest studies of treatment effectiveness for juveniles who sexually offend undertaken to date. Two of the studies in the analysis employed random assignment. The treatment approaches most often were based on cognitive-behavioral and relapse-prevention techniques, although other approaches such as sexual trauma therapy and psychosocial education were also represented in the analysis.

Based on an average followup period of nearly 5 years, the researchers found an average sexual recidivism rate of 7.37 percent for treated juveniles. By comparison, the average sexual recidivism rate for comparison group members was 18.93 percent. Further, the researchers reported that every study in the analysis yielded a positive treatment effect. Overall, an average weighted effect size of 0.43 was found, indicating "that for every 43 sexual offenders receiving the primary treatment who recidivated, 100 of the sexual offenders in the comparison group (i.e., those receiving comparison treatment or no treatment) recidivated" (Reitzel & Carbonell, 2006; p. 409).

Interestingly, two of the four strongest treatment effects found in the meta-analysis were from studies of MST treatment. In addition, Reitzel and Carbonell did not find that studies of cognitive-behavioral treatment had stronger treatment effects than studies of noncognitive-behavioral approaches. However, the researchers speculated that a number of confounding factors may have influenced this finding, including difficulties associated with categorizing studies based on their treatment approach. In discussing the overall findings from their analysis, Reitzel and Carbonell (2006, p. 417) stated:

It is encouraging that results supported previous findings ... and suggested the effectiveness of JSO treatment in the reduction of sexual recidivism, although methodological issues and reporting practices in the individual studies comprising this meta-analysis warrant caution in the interpretation of results.

Another meta-analysis that found positive treatment effects was conducted by Winokur and colleagues (2006). The analysis is important because it employed a protocol that assessed the methodological quality of potentially relevant research and excluded studies that did not reach a sufficient standard of scientific rigor. Overall, seven rigorous recidivism studies were included in the meta-analysis—one RCT and six studies that matched treatment and comparison subjects on relevant demographic and criminal history characteristics. Of the seven studies in the analysis, three examined treatment delivered in a community-based outpatient setting, three examined treatment delivered in a residential setting, and one examined treatment delivered in a correctional setting. In all seven studies, treatment involved some type of cognitive-behavioral approach. The average followup time across the seven studies was 6 years.

The researchers found that adolescents who completed sexual offender treatment had significantly lower recidivism rates than untreated adolescents. Positive treatment effects were found for sexual recidivism,¹⁰ nonsexual violent recidivism,¹¹ nonsexual nonviolent recidivism,¹² and any recidivism.¹³ Treated juveniles had sexual recidivism rates ranging from 0 to 5 percent across the seven studies. By comparison, sexual recidivism rates for untreated comparison group subjects ranged from 5 to 18 percent. Nonsexual recidivism rates ranged from 10 to 36 percent for treated subjects compared to 10 to 75 percent for untreated subjects. Based on their findings, Winokur and his colleagues (2010, pp. 23–24) concluded:

According to the results, there is a small to moderate positive effect of treatment on the recidivism rates of JSO. Specifically, juveniles who complete a cognitive-behavioral treatment program are less likely to commit a sexual or nonsexual re-offense than are juveniles who do not receive treatment, receive an alternative treatment, or do not complete treatment .... The sparse results from the subgroup analyses indicate that cognitive-behavioral treatment is effective in both community and residential settings.

Other recent meta-analyses have also found positive treatment effects. Walker and colleagues (2004), for example, conducted a meta-analysis of 10 studies involving a combined sample of 644 study subjects. The researchers found that treatments for male adolescent sexual offenders, particularly cognitive-behavioral approaches, were effective. Walker and his colleagues reported a treatment effect size of 0.37, meaning that only 37 treated study subjects recidivated for every 100 untreated study subjects who recidivated. More recently, St. Amand, Bard, and Silovsky (2008) reviewed 11 studies that examined the outcomes of treatments provided to children ages 3–12 with sexual behavior problems. The researchers found that both sexual-behavior-focused and trauma-focused interventions were effective at reducing sexual behavior problems among this population. In terms of important practice elements, St. Amand and her colleagues found that parenting management skills were particularly important in reducing sexual behavior problems in children. 

Finally, Drake, Aos, and Miller (2009) conducted a meta-analysis of five rigorous studies of sex offender treatment programs for youth as part of a larger study on evidence-based public policy options to reduce crime and criminal justice system costs. The researchers found that sex offender treatment programs for juveniles reduced recidivism, on average, by 9.7 percent. In addition, the treatment programs produced a net return on investment of more than $23,000 per program participant, or about $1.70 in benefits per participant for every $1 spent.

In summary, a handful of systematic reviews employing meta-analysis have examined the effectiveness of treatment for juveniles who commit sexual offenses in recent years. While there is widespread agreement among researchers that the evidence is far from definitive, these studies have consistently found that sex offender treatment works, particularly MST and cognitive-behavioral treatment approaches. Cost-benefit analysis also demonstrates that sex offender treatment programs for youth can provide a positive return on taxpayer investment.

Summary

"Therapeutic interventions for juveniles who sexually offend can and do work. While MST has been shown to be effective, single studies and meta-analyses on other treatment approaches have also produced positive results."

Given the prevalence of sexual offending by juveniles, therapeutic interventions for juveniles who sexually offend have become a staple of sex offender management practice in jurisdictions across the country. Indeed, the number of treatment programs for juveniles who commit sexual offenses has increased over the past 30 years, and the nature of treatment itself has changed as the developmental and behavioral differences between juvenile and adult sexual offenders have become better understood. Yet, despite the growth and widespread use of treatment with juveniles who sexually offend, uncertainty about the effectiveness of treatment in reducing recidivism is not uncommon. While inconsistent research findings and the fact that few high-quality studies of treatment effectiveness have been undertaken to date have contributed to the uncertainty, both the pattern of research findings and quality of the evidence have been changing in recent years.

This review examined the recent evidence on the effectiveness of treatment for juveniles who commit sexual offenses. While there is widespread agreement among researchers that the knowledge base is far from complete, the weight of the evidence from both individual studies and synthesis research conducted during the past 10 years suggests that therapeutic interventions for juveniles who sexually offend can and do work.

"The Stetson School's specialized program for treating children and youth with sexual behavior problems was identified as an effective program in the National Criminal Justice Association survey. The program is located in Barre, Massachusetts, and it provides individualized, trauma-sensitive treatment services for preteens as well as adolescents."

Rigorous studies have demonstrated the efficacy of MST in reducing the recidivism of juveniles who commit sexual offenses. Recent research–both single studies and meta-analyses–on other treatment approaches has also produced positive results. For example, Worling, Littlejohn, and Bookalam (2010) found that the juveniles who participated in a community-based treatment program had significantly better outcomes than comparison group members on several measures of recidivism. Based on a 20-year followup period, adolescents who participated in specialized treatment were significantly less likely than comparison group subjects to receive subsequent charges for sexual (9 percent compared to 21 percent), violent nonsexual (22 percent compared to 39 percent), or any (38 percent compared to 57 percent) new offense. The researchers also found that only a minority (11.49 percent) of the adolescent study subjects were charged with a sexual crime as an adult. Waite and colleagues (2005) found that incarcerated juveniles who received intensive treatment in a self-contained housing unit of the correctional facility had better recidivism outcomes than incarcerated juveniles who received less intensive treatment and who remained in the facility's general population. Also, meta-analyses conducted by Reitzel and Carbonell (2006), Winokur and colleagues (2006), and Drake, Aos, and Miller (2009) all found positive treatment effects. Winokur and his colleagues (2006) reported that cognitive/behavioral treatment is effective in both community and residential settings. 

Juveniles who sexually offend are clearly quite diverse in terms of their offending behaviors and future public safety risk. In fact, they appear to have far more in common with other juvenile delinquents than they do with adult sexual offenders. Research is demonstrating that there are important developmental, motivational, and behavioral differences between juvenile and adult sexual offenders and also that juveniles who commit sexual offenses are influenced by multiple ecological systems (Letourneau & Borduin, 2008). Hence, therapeutic interventions that are designed specifically for adolescents and children with sexual behavior problems are clearly needed. Moreover, treatment approaches that are developmentally appropriate; that take motivational and behavioral diversity into account; and that focus on family, peer, and other contextual correlates of sexually abusive behavior in youth, rather than focusing on individual psychological deficits alone, are likely to be most effective. The need for tailored rather than uniform treatment approaches was acknowledged by the experts at the SOMAPI forum. In addition, there is an emerging body of evidence suggesting that the delivery of therapeutic services in natural environments enhances treatment effectiveness (Letourneau & Borduin, 2008) and that the enhancement of behavior management skills in parents may be far more important in the treatment of sexually abusive behaviors in children than traditional clinical approaches (St. Amand, Bard, & Silovsky, 2008).

While the knowledge base regarding the effectiveness of treatment for juveniles who sexually offend is both expanding and improving, significant knowledge gaps remain. The need for more high-quality studies on treatment effectiveness has long been a theme in the literature, and both RCTs and well-designed quasi-experiments that examine treatment effects using equivalent treatment and comparison groups are greatly needed. Sound RCTs can provide the most trustworthy evidence about treatment effectiveness, but as Cook (2006) points out, they "are only sufficient for unbiased causal knowledge when" a correct random assignment procedure is chosen and properly implemented, "there is not differential attrition from the study across the groups being compared," and "there is minimal contamination of the intervention details from one group to another." Propensity score matching and other advanced techniques for controlling bias and achieving equivalence between treatment and comparison subjects can help enhance the credibility of evidence produced through quasi-experiments. Following their study of treatment effectiveness for adults in California–one of the few treatment studies to employ a randomized design–Marques and colleagues (2005) emphasized the importance of including appropriate comparison groups in future treatment outcome studies, and they urged researchers who assess the effects of treatment "to control for prior risk by using an appropriate actuarial measure for both treatment and comparison groups." Synthesis studies that are based on prudent exclusionary criteria and that employ the most rigorous analytical methods available are also needed. Systematic reviews and meta-analyses that are based on the most rigorous studies, incorporate statistical tests to discover potential bias, and explore how methodological and contextual variations impact treatment effects are well-equipped to provide policymakers and practitioners with highly trustworthy evidence about what works. Future research should also attempt to build a stronger evidence base on the types of treatments that work. Empirical evidence that specifies which types of treatment work or do not work, for whom, and in which situations, is important for both policy and practice. The need for high-quality studies that help identify offender- and situation-specific treatment approaches that work was acknowledged by the national experts who participated in the 2012 SOMAPI forum. Trustworthy evidence on the treatment modalities and elements that are effective with juveniles who have committed sexual offenses was also identified as a pressing need.

Notes

1RCTs are considered superior for discovering treatment effects and inferring causality because of their capacity to create valid counterfactuals and reduce bias. Modeled on laboratory experiments, RCTs have several key features, most notably the use of random assignment. In random assignment, the researcher randomly decides which study subjects participate in treatment and which do not. The random assignment of subjects creates the optimal study conditions for comparing treated and untreated subjects and making causal inferences about the impact of the intervention. 2 In addition, there may be resistance to the use of random assignment on the grounds that withholding potentially beneficial treatment from some study subjects for the sake of research is unethical. 3 This is often accomplished by matching the treatment and comparison group members on factors that are related to the outcome of interest. Sometimes statistical techniques are employed retrospectively to create equivalence between the treated and comparison subjects. 4 Methodological quality considerations typically include an assessment of the following: the study's ability to control outside factors and eliminate major rival explanations for an intervention's effects; the study's ability to detect program effects; and other considerations, such as attrition and the use of appropriate statistical tests. Based on the assessment, studies of substandard quality are typically excluded from the analysis. In addition, studies that are included in the analysis may be weighted based on their relative scientific rigor.  5 Meta-analysis also generates a summary statistic called the average effect size, which helps the analyst determine not only if the intervention is effective, but also how effective it is. There are several methods used to calculate effect sizes, as described in Lipsey and Wilson (2001). The mean-difference effect size is common when outcomes are continuously measured; the odds-ratio effect size is common when outcomes are measured dichotomously. 6 Of the 46 juveniles who received an assessment in the program, only 30 received some form of treatment outside the program being studied. 7 The researchers also found that sexual interest in children was a predictor of sexual recidivism, and that factors commonly related to delinquency overall–such as prior criminal offending and an antisocial personality–were predictive of nonsexual recidivism. 8 The research also examined whether MST treatment improved important family, peer, and academic correlates of juvenile sexual offending. 9 Study subjects were, on average, 22.9 years old at the end of the followup period. 10 p < .01. 11 Ibid. 12 p < .001. 13 Ibid.