Adresse e-mail : [email protected] (H. Latrille)
Introduction. - A number of studies have shown that the treatment of adolescents who have engaged in sexually abusive behavior (AESABs) reduces their risk of sexual recidivism. However, few studies have examined the internal processes leading to these results.
Objective. - The purpose of our research was to determine whether the treatment of AESABs (N = 43) influences the following factors related to sexual offending: deficient social skills, social isolation, ineffective coping strategies, and cognitive distortions. The impact of motivation for change and trauma symptoms on therapeutic progress was also examined.
Method. - Pre-post treatment analyses were completed.
Results. - The results of our pre-post treatment analyses indicate significant changes in social isolation, coping strategies and cognitive distortions, and a significant reduction of trauma symptoms. Only social skills remained deficient. The data on motivation were not sufficient to establish whether motivation was linked to the progress of the subjects.
Discussion. - Our study highlights the importance of interventions targeting the acquisition of social skills in AESABs. Trauma is also an important factor to consider in the assessment of therapeutic progress, despite not being directly related to sexually abusive behaviour. The limits of the study highlight the importance of establishing an a priori program evaluation framework.
Keywords: adolescents, sexual aggression, therapeutic change, motivation, trauma.
Therapeutic change in adolescents who have engaged in sexually abusive behavior
Adolescents who have engaged in sexually abusive behavior (AESABs) have recently become a research focus, in the wake of research on sexual aggression in adults (Viens et al., 2012). A high proportion of adult sexual offenders report having committed their first sexual offense while an adolescent (Oxnam & Vess, 2008; Pullman & Seto, 2012; Veneziano & Veneziano, 2002). Barbaree and Marshall (2006) suggest that AESABs are responsible for 20% of all sexual offenses and 30–50% of sexual offenses against children. Canadian police data for 2014 indicates 26% of all suspected perpetrators of sexual offenses involving victims younger than 18 years were adolescents (12–17 years old) (Allen & Superle, 2016). In light of these findings, programs specifically targeting the management and treatment of AESABs have been developed.
Evaluation of therapeutic progress
Several studies have demonstrated that treatment—especially cognitive-behavioural treatment (Andrews & Bonta, 2015; Fanniff & Becker, 2006; Letourneau et al., 2009; Veneziano & Veneziano, 2002; Walker et al., 2004)—can reduce recidivism by AESABs (e.g., Kettrey & Lipsey, 2018; Reitzel & Carbonell, 2006). However, there has been limited research on the treatment processes responsible for these results (Beggs, 2010; Tougas & Tourigny, 2013). To our knowledge, the only study to have evaluated therapeutic progress among AESABs is Viljoen et al.’s (2015) study of 163 AESABs enrolled in a cognitive-behavioural treatment program. That study compared participants’ pre-treatment and post-treatment scores on two instruments used to evaluate the risk of recidivism by adolescents: the Juvenile Sex Offender Assessment Protocol–II (J-SOAP-II; Prentky & Righthand, 2003), which evaluates the risk of sexual and nonsexual recidivism, and the Structured Assessment of Violence Risk in Youth (SAVRY; Borum et al., 2006), which evaluates the risk of violent recidivism. Both instruments are based on dynamic risk factors. Significant positive changes were observed for both instruments, with a strong change observed on the J-SOAP-II and a moderate one on the SAVRY. However, this post-treatment improvement in dynamic risk factors was not associated with lower rates of recidivism. Viljoen et al. (2015) hypothesize that this result indicates that the J-SOAP-II and SAVRY may not capture the full range of treatment-related changes. They therefore conclude that research on the evolution of specific factors throughout the course of treatment would be especially useful.
Factors associated with sexual aggression
Research on acting out by, and/or the persistence of aggression in, AESABs has revealed the existence of several factors that are important determinants of the sexually aggressive behaviours of these individuals (e.g., Veneziano & Veneziano, 2002). These include deficient social skills, social isolation, inappropriate coping strategies, and cognitive distortions.
Deficient social skills and social isolation
Research has demonstrated that AESABs have deficient social and relational skills, and that these manifest as serious difficulties establishing and maintaining social ties, which in turn often result in poor and limited peer relationships (Seto & Lalumière, 2010). These social problems create isolation that reinforces poor self-perceived social efficacy and exacerbates difficulties building relationships (Miner et al., 2016; Rabaux, 2007; Seto & Lalumière, 2010; Tardif, 2015). Both deficient social skills and social isolation are therefore thought to influence the development of sexual offending in adolescence, with sexual aggression arising in part from poor relational skills (Miner et al., 2016) secondary to problematic emotional regulation (Seto & Lalumière, 2010). Furthermore, social isolation is a risk factor for sexual recidivism (Worling & Langström, 2006). It should be noted that these difficulties are truer for AESABs who have victimized children, who appear to be more reserved and isolated (Glowacz & Born, 2013; Veneziano & Veneziano, 2002), and particularly anxious in their relationships (Gunby & Woodhams, 2010; Hendriks & Bijleveld, 2004; Hunter et al., 2003; Hunter et al., 2004; Stevens et al., 2013). On the other hand, AESABs who have victimized peers or adults more closely resemble adolescent nonsexual offenders, especially with regard to antisociality (Glowacz & Born, 2013).
Inappropriate coping strategies
Research on AESABs indicates that these individuals have coping deficits (Prentky et al., 2000). Pagé et al. (2010), noting the scarcity of studies of the association between sexual offending in adolescence and coping, conducted a study to determine whether the coping strategies of AESABs differ from those of their non-offending peers. Analysis of scores on the Coping Inventory for Stressful Situations (CISS; Endler & Parker, 1990) suggests that AESABs are more likely than their non-offending peers to attempt to cope with stressful situations by resorting to emotion-oriented coping strategies rather than task-oriented or avoidance coping strategies. In addition, AESABs tended to focus on their emotional states, which increased stress levels and put them at higher risk of becoming sexual aggressive (Pagé et al., 2010). More recently, Margari et al. (2015) compared the coping strategies of 31 AESABs, 31 nonsexual adolescent offenders, and 31 non-offending peers, using the CISS. No significant difference was observed between AESABs and nonsexual offenders, with both groups favouring avoidant strategies; however, non-offending peers were more likely to resort to emotion-oriented strategies. Task-oriented strategies were rare in all three groups. The authors conclude that avoidant strategies are a problem common to all offenders.
It is widely accepted that AESABs, like their adult counterparts, have cognitive distortions related to their sexual behaviour and their victims’ intentions (Hunter, 2012; Seto & Lalumière, 2010). In adults, these cognitive distortions are associated with sexual recidivism (Helmus et al., 2013). It is thus probable that cognitive distortions also increase the risk of committing a sexual assault in adolescents, especially by minimizing the perceived effects of the assault on the victim or the guilt associated with the act. Accordingly, cognitive distortions are thought to facilitate acting out by AESABs (Calley, 2007; Hunter, 2012), although this has not been proven in this group.
This study therefore analyzed the influence of deficient social skills, social isolation, inappropriate coping strategies and cognitive distortions—all associated with sexual aggression in adolescents—on therapeutic progress. Therapeutic progress may also be influenced by motivation for change, an important determinant of receptivity to treatment (Cortoni & Lafortune, 2009), and trauma symptoms, since violence experienced in childhood may be associated with an increased risk of committing sexual assault in adolescence (Burton, 2003; Hunter et al., 2004).
Factors associated with therapeutic progress
Motivation for change
Motivation is central to therapeutic interventions (Drieschner et al., 2004), and is demonstrated to be an important determinant of receptivity to treatment: low motivation is associated with an increased likelihood of treatment dropout, in turn associated with a higher likelihood of recidivism (Cortoni & Lafortune, 2009). Low motivation is a common problem in adolescents, who rarely enter treatment voluntarily but, rather, typically attend because they are judicially required to do so. Proulx et al. (2012) analyzed predictors of treatment dropout in a sample of 163 AESABs enrolled in group therapy who completed (among other instruments) the University of Rhode Island Change Assessment Scale (URICA; McConnaughy et al., 1983), a questionnaire on the stages of motivation for change. The results indicate that AESABs who dropped out of treatment were in the first, pre-contemplative, stage of motivation, which is characterized by the failure to accept responsibility for one’s actions and the absence of a desire for change. In contrast, AESABs, who completed treatment acknowledged that they had a problem, and had reached the action stage of motivation, characterized by commitment to resolving their problems through treatment (Proulx et al. 2012). Thus, motivation for change is particularly relevant to analyses of treatment effectiveness, as it influences commitment and, by extension, therapeutic progress.
AESABs report a wide range of childhood-victimization experiences, especially sexual and physical abuse (Seto & Lalumière, 2010). In particular, the majority of adolescents who have sexually victimized children were themselves sexually victimized in childhood (Hart-Kerkhoffs et al., 2009; Seto & Lalumière, 2010; Stevens et al., 2013). On the other hand, adolescents who have sexually victimized peers or adults are more likely to have been physically victimized (Gunby & Woodhams, 2010). In their study of 82 AESABs, Oxnam and Vess (2008) observed that adolescents who had been sexually victimized in childhood were more likely to exhibit psychopathological traits and interpersonal problems compared to those offenders without a history of sexual victimisation. These traumatic experiences may thus have led AESABs to develop psychosocial deficits related to risk factors for sexual assault. Viens et al. (2012) studied the effectiveness of group therapy in 29 AESABs and a control group of 22 non-offending adolescents. They reported that adolescents who completed the treatment program exhibited reduced trauma symptoms, especially depression, anger, dissociation, and sexual preoccupation. The association between this reduction and risk factors for sexual aggression was not, however, explored.
The first objective of this study was to determine whether treatment of AESABs leads to significant clinical changes. To this end, four factors related to sexual aggression were analyzed: deficient social skills, social isolation, inappropriate coping strategies, and cognitive distortions. The second objective was to determine whether AESABs’ motivation was associated with observed therapeutic progress. Finally, the third, more exploratory, objective was to determine whether treatment led to changes in trauma symptoms, and whether any such changes were associated with other forms of therapeutic progress.
The data for this study was taken from the database used in Proulx et al.’s (2012) study of predictors of treatment dropouts in a sample of 163 male AESABs who were between 13 and 17 years old and who had participated in a group-therapy program between June 2002 and March 2005. The database contains information on the AESABs’ victims (sex and relationship to the AESAB—intrafamilial or extrafamilial) and on treatment, including the results of psychometric tests administered before, during, and after treatment (Proulx et al., 2012). No other data which would have been useful for analysis of treatment was available.
For this study, the sample was composed of only those AESABs for whom complete data was available on all the tests used to investigate therapeutic progress. Because data on these tests was incomplete in many cases, the final sample was composed of 43 male AESABs who were 13 to 17 years old upon entering treatment. The mean age of the participants was 15 years (SD = 1.08). Of the 43 participants, 42 had victimized only children younger than 12 years, and one had also victimized a female adolescent of 14 years. With the exception of the last victim, all victims were at least 3 years younger than their aggressor
Deficient social skills were evaluated using Matson’s Evaluation of Social Skills in Youngsters (MESSY; Matson et al., 1983) (Appendix A), which evaluates the relational skills of young people aged 4–18 years. The instrument consists of 62 items divided into five subscales: appropriate social skills, inappropriate assertiveness, impulsive and recalcitrant traits, overconfidence, and jealousy. An additional dimension, “miscellaneous items” encompasses items related to social skills, both positive and negative (Matson, 1994). The total score on the test varies between 64 and 308, with higher scores indicating less appropriate social skills. For the purposes of calculation of the total score, the direction of the 23 items on the “appropriate social skills” and the 4 items on the “miscellaneous items” scales are reversed (Matson et al., 2010). Cronbach’s alpha for the MESSY varies between 0.84 and 0.93 (Matson et al., 2010).
Social isolation was evaluated using the Social Support Questionnaire (SSQ; Sarason et al., 1983) (Appendix B), a 27-item questionnaire on social support that identifies the number of individuals in the respondent’s entourage they believe are available for support. This study used the SSQ6 (Sarason et al., 1987), an abridged, 6-item, version of the original instrument. The SSQ6 measures two elements: perceived availability of social support (“N score”, 0–54), and satisfaction experienced (“S score”, 6–36). The higher the N score, the greater the respondent’s belief that individuals in their entourage can provide support. The higher the S score, the more satisfied the respondent is with the support received. Cronbach’s alpha for the SSQ is very high: 0.90–0.93 (Sarason et al., 1987).
Coping strategies were evaluated with the American version for adolescents (13–18 years old) of the Coping Inventory for Stressful Situations (CISS; Endler & Parker, 1990) (Appendix C). The CISS is a self-report questionnaire that evaluates the way individuals react, adapt, or cope with situations that are difficult, anxiety provoking, or disturbing. The questionnaire comprises subscales for three types of strategies: task-oriented, emotion-oriented, and avoidance. The avoidance subscale comprises two subscales: distraction and social diversion. The higher the score on a subscale, the more likely the respondent is to resort to that type of coping strategy. Only the total score of each subscale was analyzed; this varied from 16 to 80 for the three main subscales, from 8 to 40 for the distraction subscale and from 5 to 25 for the social diversion subscale. Cronbach’s alpha for the subscales is 0.76–0.92 (Endler & Parker, 1999).
Cognitive distortions were evaluated using the Questionnaire des Cognitions pour Adolescents [questionnaire on cognitions in adolescents] (QCA; Hunter et al., 1991) (Appendix D), which determines the presence of cognitive distortions through 32 dichotomous items (true/false, scored as 1/0). The total score for the questionnaire is the sum of the item scores and varies from 0 to 32—the higher the score, the more cognitive distortions are present. Hunter at al. (1991) tested the internal consistency of the instrument in a sample of 37 AESABs and reported a Cronbach’s alpha of 0.45 for the initial test and 0.71 for the post-test.
Motivation for change was evaluated using the University of Rhode Island Change Assessment Scale (URICA; McConnaughy et al., 1983) (Appendix E), a self-report questionnaire measuring the stages of motivation in Prochaska and DiClemente’s transtheoretical model of change (1983). The questionnaire comprises 32 questions, divided into four subscales, which correspond to the four stages of change in the model: precontemplation, contemplation, action, and maintenance. The score on each subscale varies from 8 to 40—the higher the score, the more the respondent’s behaviour is congruent with the stage’s definition. Cronbach’s alpha for the subscales is 0.77–0.88 (Greenstein et al., 1999).
The analytical method for the URICA responses consists of combining the subscales, using the formula: contemplation + action + maintenance - precontemplation (Tison et al., 2009). This method yields a continuous measure of motivation for change, which ranges from 16 to 112. Tison et al. (2009), in their examination of the motivation for change of adults with alcohol dependency, considered a normal, standard, profile to correspond to a mean score of 89. To our knowledge, no study of AESABs has used this instrument, which invites some caution when interpreting our results. In this study, a mean score of 80–90 was considered indicative of acceptable motivation for change.
Trauma symptoms were evaluated using the Trauma Symptom Checklist for Children (TSC-C; Brière, 1996) (Appendix F), an instrument designed to ascertain the presence of trauma-related psychological symptoms in young people (8–17 years old). The TSC-C comprises 54 items divided into six subscales: anxiety, depression, post-traumatic stress, sexual preoccupation, dissociation, and anger. The higher the score on a subscale, the more severe the symptom. The scores of the subscales are summed to obtain a total score for the instrument—the higher the total score the more the respondent exhibits trauma symptoms. Cronbach’s alpha for the subscales was satisfactory, varying between 0.77 and 0.89 (Brière, 1996).
Data was collected from June 2002 to March 2005 (Proulx et al. 2012). All data was collected during the various stages of case management. The tests analyzed in this study were administered to adolescents prior to treatment (pre-treatment) and after treatment had been completed (post-treatment); comparison of these results allowed evaluation of the participant’s progress once they had been taken into care. Motivation for change was evaluated during the treatment period—more specifically, two months after treatment had commenced. Treatment sessions were either group or individual. Each participant responded to the tests individually. All participants consented to the use of the information concerning them for research purposes. The data for the present study was extracted in 2015, following ethics approval by the Université de Montréal.
To evaluate therapeutic progress, the MESSY, SSQ, CISS, and QCA results were evaluated pre- and post-treatment. The first analytical procedure was the performance of paired t-tests. The second analytical procedure was the calculation of effect size, using Cohen’s d. As recommended by Cohen (1992), coefficients below 0.2 were considered weak, those around 0.5 were considered moderate, and those around 0.8 were considered strong. The third analytical procedure was the calculation of the percent change in the four factors of interest. This was expressed as the percent of the maximum possible score (POMP) (Cohen et al., 1999), which is the size of the pre-post difference expressed as a percentage. In order to verify the premiss of the study’s second objective, that motivation for change is associated with therapeutic progress, URICA scores were correlated to the POMP scores of the MESSY, SSQ, CISS and QCA. In keeping with Cohen’s (1988) recommendations, correlation coefficients of .10–.23 were considered weak, coefficients of .24–.36 were considered moderate, and coefficients greater than .37 were considered strong. Finally, to determine whether treatment had an impact on trauma symptoms, and whether any such change was associated with changes in the other factors, two further analyses were conducted. The first consisted of a pre-post analysis of TSC-C scores, using the same methodology used with the other instruments. The second consisted of analysis of the correlation between the POMP of the TSC-C, on the one hand, and the POMPs of the MESSY, SSQ, CISS and QCA, on the other.
The results of the pre-post analyses—mean pre- and post-treatment scores for each subscale, results of paired t tests, effect sizes, and POMPs— are presented in Table 1.
Deficient social skills (MESSY)
A significant post-treatment improvement (p = .03) in impulsive and recalcitrant traits was observed; the effect size was weak (d = 0.34) and the mean pre-post change was 6.74%. On the other hand, there was a significant reduction (p = .01) in appropriate social skills; the effect size was weak to moderate (d = 0.40) and the mean pre-post change was 5.55%. No pre-post difference was observed for inappropriate assertiveness (p = .08), overconfidence (p = .12), jealousy (p = .65), miscellaneous items (p = .47), or total score (p = .90).
Social Isolation (SSQ)
A significant increase in the number of people in the participant’s entourage considered available for support was observed post-treatment (p = .00); the effect size was moderate (d = 0.59) and the mean pre-post change was 11.77%. No pre-post difference was observed for satisfaction with support network (p = .07).
Inappropriate coping strategies (CISS)
A significant post-treatment increase (p = .01) in task-oriented coping strategies and decrease in emotion-oriented coping strategies (p = .01) were observed. The effect of treatment was moderate in both cases (Cohen’s d = 0.51 and 0.41, respectively) and the mean pre-post differences were 6.36% and 8.58%. No significant pre-post difference was observed for the avoidant (p = .85), social diversion (p = .65), and distraction (p = .47) subscales.
Cognitive distortions (QCA)
A significant post-treatment improvement in cognitive distortions (p = .01) was observed; the effect size was moderate (d = 0.47), and the mean change was 2.74%.
Impact of motivation on treatment
The mean scores and standard deviations of the URICA and its subscales (corresponding to different stages of change in treatment) are presented in Table II. The highest mean scores were obtained for the contemplation and action subscales. In addition, the composite score for motivation for change (contemplation + action + maintenance - precontemplation) was 82.40, indicating that participants were, on average, conscious of their problem and, for the most part, in a stage of observable change.
To determine whether the level of motivation was associated with observed changes in deficient social skills, social isolation, ineffective coping strategies, and cognitive distortions, URICA total and subscale scores were correlated to the POMPs of the other tests. No significant correlation was observed between motivation for treatment and changes in these factors, which are associated with sexual aggression.
Trauma symptoms: Changes and impact on treatment
The pre-and post-treatment scores on the TSC-C and its subscales are presented in Table III. A significant post-treatment improvement was observed in symptoms related to anxiety (p = .02; d = 0.29), depression (p = .01; d = 0.35), post-traumatic stress (p = .00; d = 0.46), and anger (p = .04; d = 0.34), and in total score (p = .01; d = 0.35). The mean pre-post change was 4.91–8.45%, with the greatest differences observed for post-traumatic stress (8.45%), depression (6.37%), and anger (6.20%). No pre-post difference was observed for sexual preoccupation (p = .262) and dissociation (p = .170).
Correlation analyses were performed to determine the existence of associations between changes in trauma symptoms and changes in social skills, social isolation, coping strategies, and cognitive distortions. The results of the analyses correlating the POMPs of the TSC-C and its subscales to those of the MESSY and SSQ are presented in Table IV. The results of the analyses correlating the POMPs of the TSC-C and its subscales to those of the CISS and QCA are presented in Table V.
Several significant correlations were observed between the POMPs of the TSC-C subscales and those of the other instruments. Thus, the post-treatment decrease in anger-related symptoms (on the TSC-C) was strongly and positively correlated (r = .49; p < .01) to the decrease in inappropriate assertiveness on the MESSY and moderately and positively correlated (r = .32; p < .05) to the decrease in MESSY total score, which measures negative social skills. In addition, the post-treatment decrease in dissociation-related symptoms (on the TSC-C) was moderately and positively correlated (r = .35; p < .05) to the decrease in jealousy on the MESSY.
For the SSQ, there was a strong positive correlation (r = .37; p < .05) between the number of people in the respondent’s entourage considered available for support and the post-treatment decrease in anxiety-related symptoms on the TSC-C. For the CISS, there was a strong positive correlation (r = .31; p < .05) between emotion-based coping strategies and the decrease in anger-related symptoms on the TSC-C. Finally, there was no significant correlation between changes in the QCA and TSC-C results.
The objective of this study was to evaluate the therapeutic progress of AESABs in terms of four factors associated with sexual aggression: inadequate social skills, social isolation, ineffective coping strategies, and cognitive distortions. Motivation for change and the presence of trauma symptoms were also taken into account, as they are associated with therapeutic progress. Significant positive changes were observed for each of the four factors. On the other hand, motivation was not associated with therapeutic progress. Finally, the results indicate a post-treatment reduction of trauma symptoms and the existence of several associations between improvements with regard to trauma symptoms and positive changes in the factors related to sexual aggression.
Evaluation of therapeutic progress
Deficient social skills
The results indicate a 6.74% post-treatment reduction in impulsive and recalcitrant traits. Despite this, participants’ overall social skills remained deficient, and in fact decreased slightly. It should be noted that AESABs who victimize children have been reported to have more deficient social skills and more problematic peer relations than AESABs who victimize peers or adults (Gunby & Woodhams, 2010; Hendriks & Bijleveld, 2004; Hunter et al., 2003; Hunter et al., 2004, Stevens et al., 2013). According to Sion and Blondeau (2012), commitment to a treatment program provides a feeling of affiliation and favours the emergence of individual and collective responsibility. It is therefore possible that participants entering treatment believed they possessed appropriate social skills, and subsequently improved their self-assessment skills. If this were indeed the case, their score on this scale would fall as they progressed through treatment, despite the fact that no change in their actual social skills had actually occurred. It should be recalled that the observed post-treatment reduction in impulsivity reflects better emotional regulation and self-control in social contexts. Nevertheless, participants continued to exhibit some interpersonal deficits, as evidenced by the absence of significant change in inappropriate assertiveness, overconfidence, and jealousy. More clinical work targeting the development of adequate interpersonal skills is therefore indicated.
The results also indicate a reduction in social isolation following treatment. The number of individuals in the respondent’s entourage they believe are available for support significantly increased, and the percent increase (11.77%) was in fact the largest seen. Work on this factor is particularly important, as social isolation is a demonstrated risk factor for sexual recidivism in adolescents (Worling & Langström, 2006) decreasing social isolation would therefore help AESABs avoid reoffending. On the other hand, the results also indicate no change in participants’ satisfaction with their support network. There are two possible explanations for this. First, the Sarason Social Support Questionnaire (SSQ) is a self-report instrument, and thus relies on the respondent’s own assessment of their satisfaction with the support they receive from their entourage. The type of support a participant needs in order to continue to progress and change after treatment is probably different from what they needed previously, and this could lead them to call into question some of their social ties. Second, while treatment allowed AESABs to broaden their support network to include health professionals, they were not necessarily comfortable with these new relationships. It is possible that satisfaction with the support network is a longer-term proposition. Further research is needed on this question.
Inappropriate coping strategies
One goal of treatment is to help AESABs develop new problem-solving skills. In both of the two studies reviewed above, AESABs resorted to inappropriate coping strategies when faced with stressful situations, although the specific strategies differed: Pagé et al. (2010) observed emotion-oriented strategies, while Margari et al. (2015) observed avoidance-oriented ones. In the present study, the mean pre-treatment CISS scores indicate that the AESABs had resorted to all three types of coping strategies, namely, task-oriented, emotion-oriented, and avoidance-oriented. The post-treatment results indicate no change in recourse to avoidance strategies but a strong reduction in recourse to emotion-oriented strategies (6.36%) and a strong increase (8.58%) in recourse to task-oriented strategies revolving around problem identification and positive engagement with the stressor. Post-treatment, AESABs reported that they were managing their stress levels more productively and proactively, and this was one of the largest changes observed. Future research should verify whether this change is persistent, for example by adding an additional follow-up period.
Finally, the analysis of therapeutic progress revealed a positive change in cognitive distortions that excuse or minimize sexual offending. According to Kim et al. (2015), adolescents, who are still developing, are more receptive to cognitive restructuring than are adults. In the present study, there was a significant reduction in cognitive distortions, although this was the lowest percent change observed (2.47%). Treatment therefore appears to have led AESABs to become more aware of the gravity and consequences of their actions; this is a particularly important stage of the managed-care process (Calley, 2007). This increased awareness may be related to the observed changes in coping strategies: an improved ability to identify and modify thoughts related to behaviours may also lead to an improvement in problem-solving skills. The relationship between these two elements should be the subject of future research.
Impact of motivation on treatment
In this study, the level of motivation (composite score: 82.4) two months after the beginning of treatment indicates that most of the AESABs were conscious of their problem and were in a stage of observable change (Tison et al., 2009). However, there was no association between moderate or high motivation, on the one hand, and changes in social skills, social isolation, coping strategies, or cognitive distortions, on the other. It thus appears that motivation for change had no influence on therapeutic progress. However, motivation was only measured once, after two months of treatment, and it is possible that motivation did subsequently improve. Nevertheless, our results do not indicate an association between therapeutic progress and pre-post changes in the four factors related to sexual aggression. Future studies should include collecting data on motivation at multiple points both before and after treatment.
Trauma symptoms: Changes and impact on treatment
Our analysis of pre-post changes in trauma symptoms indicates clear post-treatment improvements. Specifically, there was significant reductions in anxiety, depression, and anger. Further, treatment was associated with a significant reduction in the trauma symptoms commonly observed in AESABs, with percent changes varying from 4.91% to 8.45%. These findings echo those of Viens et al. (2012) who studied the effectiveness of group therapy for AESABs, although these authors also observed a decrease in sexual preoccupation, a result not observed in the present study. Given that most AESABs who victimize children were themselves sexually victimized as children (Hart-Kerkhoffs et al., 2009; Seto & Lalumière, 2010), it is possible that symptoms related to sexual preoccupation and sexual distress require some additional, specialized, treatment, as is the case for symptoms related to dissociation.
According to Andrews and Bonta (2015), emotional factors such as anxiety and psychological discomfort are factors of resistance to change that may hinder treatment. In this study, reductions in symptoms related to anxiety and anger were associated with an improvement in social skills and a reduction in social isolation. This suggests that AESABs who are less anxious are more apt to interact with others. Our analyses also demonstrate that reduced anger was associated with reduced recourse to emotion-oriented coping strategies. Thus, better emotional regulation appears to facilitate more productive and proactive coping with stressful situations. This effect of effective emotional management on coping strategies has also been observed in other populations, such as university students (Arslan, 2010) and adults in the general population (Blanchard-Fields, 2007). Finally, no significant association was observed between a reduction in trauma symptoms and a reduction of thoughts supporting sexually aggressive behaviours. This suggests that cognitive distortions in these AESABs were not driven by psychological symptoms related to ill treatment in childhood. Future research on the development and maintenance of cognitive distortions should therefore focus on factors other than trauma.
Limitations and suggestions for future research
The first limitation of the study was the limited number of participants (N = 43), which prevents generalization of the results. The second limitation is related to the data used. As the data had been collected previously, the measures used were not necessarily the most appropriate for analysis of the risk factors studied. This underscores the importance of measuring the most appropriate factors during treatment. In addition, many variables were measured using self-report instruments, and these instruments are subject to social-desirability biases. Individual interviews or more objective tests could be used in the future. The third limitation is the existence of factors—for example, the nature of the offenses committed— which may have influenced therapeutic progress but were not controlled for. A recent meta-analysis (Kettrey & Lipsey, 2018) emphasizes the importance of distinguishing between specialized adolescent offenders, who commit only sexual crimes, and generalist ones, who commit both sexual and nonsexual offenses, since the two groups do not have the same treatment needs. According to those authors, few studies of the effectiveness of treatment of AESABs have made this distinction. Future research on therapeutic progress should endeavour to do so.
The last limitation of this study is the absence of information on the intensity of treatment. According to Andrews and Bonta’s (2015) principle of risk, the duration and intensity of treatment must parallel the individual’s risk of recidivism. This principle also holds for adolescent offenders (Reitzel & Carbonell, 2006). In this study, no data was available on the risk of recidivism. Future research should examine this element. Finally, it would be useful for research on pre-post treatment changes to include a third observation period, situated several months post-treatment. This would shed light on whether the adolescents’ therapeutic progress is persistent.
The importance of the conceptual framework
The limitations of this study demonstrate that evaluating treatment effectiveness requires more than the simple extraction of data—the data must also be organized to satisfy specific evaluation objectives. In this connection, Cortoni (Cortoni, 2003; Cortoni & Lafortune, 2009) suggests that applied research on program evaluation be developed in tandem with the programs themselves and be subsequently refined as the programs progress. Researchers have suggested using a model which links initial objectives (taking into account sample and services) to short-, medium-, and long-term results, and controls for other variables (Cortoni, 2003; McGuire, 2001).
A conceptual framework for program evaluation that reflects the preoccupations of case workers must clearly define the evaluation objectives (Gaes, 2001) as well as the nature of interventions, the social and/or psychological mechanisms involved, the nature of the expected results, and, in particular, the program participants. Evaluation may be formative or summative. Formative evaluation evaluates therapeutic progress and problems related to interventions, while summative evaluation evaluates global effectiveness—that is, post-treatment changes (Cortoni & Lafortune, 2009). The program-evaluation strategy should also include mechanisms for measuring motivation. In addition, the risk of recidivism is a good indicator of program effectiveness, as reductions in this risk are the primary goal of any treatment program for offenders. Finally, treatment integrity must be evaluated, and must take into account two factors: compliance with the program’s model, and the quality of program delivery (through evaluation of the work of case workers) (McGuire, 2001). It should be noted that the delivery model should be congruent with Andrews and Bonta’s (2015) principle of receptivity, which stipulates that treatment modalities should be effective for the target population and adapted to participants’ individual characteristics.
The steps outlined above can be part of a Logic Model (Frechtling, 2015), which is a visual illustration of a program's resources, activities and expected outcomes. The Model establishes the relationships between various components of the treatment and can be used during treatment program development, implementation, and evaluation. The establishment of this type of evaluation framework at the planning stage of the development of the treatment program leads to a solid evaluation plan that permits sound conclusions about the treatment’s effectiveness. Within is context, the use of multiple qualitative and quantitative research methods enhances the reliability of information.
This study has demonstrated that it is possible to measure therapeutic progress in AESABs receiving treatment. This is an innovative finding, since, to our knowledge only one other study (Viljoen et al., 2015) has investigated this question, and that study used risk-assessment tools that are not necessarily suitable for this purpose. Our results demonstrate that special attention should be given to clinical work that improves the relational skills of AESABs. Moreover, trauma is an important factor to be taken into account in the evaluation of the therapeutic progress of AESABs, although it is not directly related to sexual aggression. Finally, future research should examine the association between therapeutic progress and reductions in the risk of recidivism, while taking care to comply with the requirements of the conceptual framework of program evaluation.
The authors declare that there is no conflict of interest.
Allen, M.K., & Superle, T. (2016). Youth crime in Canada, 2014. Juristat, 36(1).
Andrews, D. A., & Bonta, J. (2015). Le comportement délinquant : analyse et modalités d’intervention (5th ed.) [The Psychology of Criminal Conduct]. Agen : Les Presses de l’Enap.
Arslan, C. (2010). An Investigation of Anger and Anger Expression in Terms of Coping with Stress and Interpersonal Problem-Solving. Educational Sciences: Theory & Practice, 10(1), 25-43.
Barbaree, H. E., & Marshall, W. L. (2006). An introduction to the juvenile sex offender: Terms, concepts, and definitions. In H. E. Barbaree & W. L. Marshall (Eds.), The Juvenile Sex Offender (2nd ed. p.1-18). The Guilford Press.
Beggs, S. (2010). Within-treatment outcome among sexual offenders: A review. Aggression and Violent Behavior, 15(5), 369-379. https://dx.doi.org/10.1016/j.avb.2010.06.005
Blanchard-Fields, F. (2007). Everyday Problem Solving and Emotion: An Adult Developmental Perspective. Current Directions in Psychological Science, 16(1), 26–31. https://dx.doi.org/10.1111/j.1467-8721.2007.00469.x
Borum, R., Bartel, P., & Forth, A. (2006). Manual for the Structured Assessment for Violence Risk in Youth (SAVRY). Psychological Assessment Resources.
Brière, J. (1996). The trauma symptom checklist for children (TSCC): Professional manual. Odessa, FL: Psychological Assessment Resources.
Burton, D. L. (2003). Male adolescents: sexual victimization and subsequent sexual abuse. Child and Adolescent Social Work Journal, 20(4), 277-296.
Calley, N. G. (2007). Integrating theory and research: The development of a research‐based treatment program for juvenile male sex offenders. Journal of Counseling & Development, 85(2), 131-142. https://dx.doi.org/10.1002/j.15566678.2007.tb00455.x
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Erlbaum.
Cohen, J. (1992). A power primer. Psychological bulletin, 112(1), 155-159. https://dx.doi.org/10.1037/0033-2909.112.1.155
Cohen, P., Cohen, J., Aiken, L. S., &West, S. G. (1999). The problem of units and the circumstance for POMP. Multivariate behavioral research, 34(3), 315-346. https://dx.doi.org/10.1207/S15327906 MBR3403_2
Cortoni, F. (2003). La recherche appliquée à l’appui de l’efficacité des programmes [Applied research in support of program effectiveness]. Forum sur la recherche correctionnelle, 2. p.36-38.
Cortoni, F., & Lafortune, D. (2009). Le traitement correctionnel fondé sur des données probantes : une recension [Research based correction treatment: a literature review]. Criminologie, 42(1). 61-89. https://dx.doi.org/10.7202/029808ar
Drieschner, K. H., Lammers, S. M., & van der Staak, C. P. (2004). Treatment motivation: An attempt for clarification of an ambiguous concept. Clinical Psychology Review, 23(8), 1115-1137. https://dx.doi.org/10.1016/j.cpr.2003.09.003
Endler, N. S., & Parker, J. D. A. (1990). CISS : Coping Inventory for Stressful Situations. Manual. Toronto : Multi-health Systems Inc.
Endler, N. S., & Parker, J. D. A. (1999). CISS : Coping Inventory for Stressful Situations. Manual (2nd ed.). Multi-health Systems Inc.
Fanniff, A. M., & Becker, J. V. (2006). Specialized assessment and treatment of adolescent sex offenders. Aggression and Violent Behavior, 11(3), 265-282. https://dx.doi.org/10.1016/j.avb.2005.08.003
Frechtling JA (2015). "Logic Models". International Encyclopedia of the Social & Behavioral Sciences (pp. 299-305). Elsevier. doi:10.1016/b978-0-08-097086-8.10549-5. ISBN 978-0-08-097087-5.
Gaes, G. (2001). Évaluation d’un programme : Lignes directrices pour poser les bonnes questions [Program Evaluation: Guidelines for Asking the Right Questions]. In L. Motiuk & R. Serin (Eds.), Compendium 2000 des programmes correctionnels efficaces (vol. 1, p. 209-213). Ministère des Approvisionnements et Services Canada.
Glowacz, F., & Born, M. (2013). Do adolescent child abusers, peer abusers, and non-sex offenders have different personality profiles?. European Child & Adolescent Psychiatry, 22(2),117-125. https://dx.doi.org/10.1007/s00787-012-0333-2
Greenstein, D. K., Franklin, M. E., & McGuffin, P. (1999). Measuring motivation to change: An examination of the University of Rhode Island Change Assessment Questionnaire (URICA) in an adolescent sample. Psychotherapy, 36(1), 47-55. https://dx.doi.org/10.1037/h0087813
Gunby, C., & Woodhams, J. (2010). Sexually deviant juveniles: comparisons between the offender and offence characteristics of ‘child abusers’ and ‘peer abusers’. Psychology, Crime & Law, 16(1-2), 47-64. https://dx.doi.org/10.1080/10683160802621966
Hart-Kerkhoffs, L. A., Doreleijers, T. A., Jansen, L. M., van Wijk, A.P., & Bullens R. A. (2009). Offense related characteristics and psychosexual development of juvenile sex offenders. Child and Adolescent Psychiatry and Mental Health, 3(1), 1-10. https://dx.doi.org/10.1186/1753-2000-3-19
Helmus, L., Hanson, R. K., Babchishin, K. M., & Mann, R. E. (2013). Attitudes Supportive of Sexual Offending Predict Recidivism A Meta-Analysis. Trauma, Violence, & Abuse, 14(1), 34-53. https://dx.doi.org/10.1177/1524838012462244
Hendriks, J., & Bijleveld, C.C.J.H. (2004). Juvenile sexual delinquents: Contrasting child abusers with peer abusers. Criminal Behaviour and Mental Health, 14(4). 238-250. https://dx.doi.org/10.1002/cbm.591
Hunter, J. A. (2012). Patterns of sexual offending in juveniles risk factors. In E. P. Ryan, J. A. Hunter & D. C. Murrie (Eds.). Juvenile sex offenders : A guide to Evaluation and treatment for mental health professionals (p. 34-53). Oxford University Press.
Hunter, J. A., Becker, J. V., Kaplan, M., & Goodwin, D. W. (1991). Reliability and discriminative utility of the Adolescent Cognitions Scale for juvenile sexual offenders. Sexual Abuse: A Journal of Research and Treatment, 4(4), 281-286. https://dx.doi.org/10.1177/107906329100400403
Hunter, J. A., Figueredo, A. J., Malamuth, N. M., & Becker, J. V. (2003). Juvenile sex offenders: Toward the development of a typology. Sexual abuse : A Journal of Research and Treatment, 15(1), 27-48. https://dx.doi.org/10.1177/107906320301500103
Hunter, J. A., Figueredo, A. J., Malamuth, N. M., & Becker, J. V. (2004). Developmental pathways in youth sexual aggression and delinquency: Risk factors and mediators. Journal of Family Violence, 19(4), 233-242. https://dx.doi.org/10.1023/B:JOFV.0000032633.37269.1d
Kettrey, H.H. & Lipsey, M.W. (2018). The effects of specialized treatment on the recidivism of juvenile sex offenders: a systematic review and meta-analysis. Journal of Experimental Criminology, 14(3), 361-387. https://dx.doi.org/10.1007/s11292-018-9329-3
Kim, B., Benekos, P. J., & Merlo, A. V. (2015). Sex offender recidivism revisited : Review of recent meta-analyses on the effects of sex offender treatment. Trauma, Violence, & Abuse, 1-13. https://dx.doi.org/10.1177/1524838014566719
Letourneau, E. J., Henggeler, S. W., Borduin, C. M., Schewe, P. A., McCart, M. R., Chapman, J. E., & Saldana, L. (2009). Multisystemic Therapy for Juvenile Sexual Offenders: 1-Year Results From a Randomized Effectiveness Trial. Journal of Family Psychology, 23(1), 89-102. https://dx.doi.org/10.1037/a0014352
Margari, F., Lecce, P. A., Craig, F., Lafortezza, E., Lisi, A., Pinto, F., Stallone, V., Pierri, G., Pisani, R., Zagaria, G., Margari, L., & Grattagliano, I. (2015). Juvenile sex offenders: Personality profile, coping styles and parental care. Psychiatry Research, 229(1), 82-88. https://dx.doi.org/10.1016/j.psychres.2015.07.066
Matson, J. L. (1994). Matson evaluation of social skills with youngsters: Manual (2nd ed.). IDS Publishing.
Matson, J. L., Rotatori, A. F., & Helsel, W. J. (1983). Development of a rating scale to measure social skills in children: The Matson evaluation of social skills with youngsters (MESSY). Behaviour Research Therapy, 21(4), 335-340.
Matson, J. L., Neal, D., Fodstad, J. C., Hess, J. A., Mahan, S., & Rivet, T. T. (2010). Reliability and validity of the Matson evaluation of social skills with youngsters. Behavior Modification, 34(6), 539-558. https://dx.doi.org/10.1177/0145445510384844
McConnaughy, E. A., Prochaska, J. O., & Velicer, W. F. (1983). Stages of change in psychotherapy: measurement and sample profiles. Psychotherapy: Theory, Research and Practice, 20, 368-375. https://dx.doi.org/10.1037/h0090198
McGuire, J. (2001). Élaboration d’un modèle logique de programme à l’appui de l’évaluation [Development of a Program Logic Model to Assist Evaluation]. In L. Motiuk & R. Serin (Eds.), Compendium 2000 des programmes correctionnels efficaces (vol. 1, p. 235-248). Ministère des Approvisionnements et Services Canada.
Miner, M. H., Romine, R. S., Robinson, B. E., Berg, D., & Knight, R. A. (2016). Anxious attachment, social isolation, and indicators of sex Drive and compulsivity predictors of child sexual abuse perpetration in adolescent males?. Sexual abuse: A Journal of Research and Treatment, 28(2), 132-153.
Oxnam, P., & Vess, J. (2008). A typology of adolescent sexual offenders: Millon Adolescent Clinical Inventory profiles, developmental factors and offence characteristics. The Journal of Forensic Psychiatry & Psychology, 19(2), 228-242.
Pagé, C. A., Tourigny, M., & Renaud, P. (2010). Analyse comparée d’un groupe de jeunes délinquants sexuels et de jeunes non délinquants sexuels : y a-t-il une différence au niveau de leurs stratégies d’adaptation ? [A comparative analysis of youth sex offenders and non-offender peers: Is there a difference in their coping strategies?]. Sexologies, 19(2), 109-118. https://dx.doi.org/10.1016/j.sexol.2009.11.001
Prentky, R., Harris, B., Frizzell, K., & Righthand S. (2000). An actuarial procedure for assessing risk with juvenile sex offenders. Sexual Abuse: A Journal of Research and Treatment. 12(2), 71-93. https://dx.doi.org/10.1177/107906320001200201
Prentky, R., & Righthand, S. (2003). Juvenile Sex Offender Assessment Protocol II (J-SOAP-II) manual. Washington, DC: Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention, U.S. Department of Justice.
Proulx, J., Tourigny, M., & Lafortune, D. (2012). Les prédicteurs de la fin prématurée du traitement chez les adolescents délinquants sexuels : une étude multisite [Predictors of premature withdrawal from treatment among adolescent sexual offenders : a multisite study]. In M. Tardif, M. Jacob, R. Quenneville & J. Proulx (Eds.), La délinquance sexuelle des mineurs. Approches cliniques (vol. 1, p. 567-596). Les Presses de l'Université de Montréal.
Pullman, L., & Seto, M. (2012). Assessment and treatment of adolescent sexual offenders: Implications of recent research on generalist versus specialist explanations. Child Abuse and Neglect, 36, 203-209. https://dx.doi.org/10.1016/j.chiabu.2011.11.003
Rabaux, J. (2007). Les mineurs délinquants sexuels [Minors who sexually offend]. Journal du droit des jeunes, 265(5), 15-21. https://dx.doi.org/10.3917/jdj.265.0015
Reitzel, L. R, & Carbonell, J. L. (2006). The effectiveness of sexual offender treatment for juveniles as measured by recidivism: A meta-analysis. Sexual Abuse: A Journal of Research and Treatment, 18(4), 401-421. https://dx.doi.org/10.1177/107906320601800407
Sarason, I.G., Levine, H.M., Basham, R.B., & Sarason, B.R. (1983). Assessing social support : the social support questionnaire. Journal of Personality and Social Psychology, 44(1), 127-139. https://dx.doi.org/10.1037/0022-35188.8.131.52
Sarason, I. G., Sarason, B. R., Shearin, E. N., & Pierce, G. R. (1987). A brief measure of social support: Practical and theoretical implications. Journal of Social and Personal Relationships, 4(4), 497-510. https://dx.doi.org/10.1177/0265407587044007
Seto, M. C., & Lalumière, M. L. (2010). What is so special about male adolescent sexual offending? A review and test of explanations through meta-analysis. Psychological Bulletin, 136(4), 526-575. https://dx.doi.org/10.1037/a0019700
Sion, V. & Blondeau, S. (2012). La thérapie de groupe [Group therapy]. In M. Tardif, M. Jacob, R. Quenneville & J. Proulx (Eds.), La délinquance sexuelle des mineurs : approches cliniques (vol. 1, p. 295-301). Les Presses de l'Université de Montréal. https://dx.doi.org/10.4000/books.pum.6116
Stevens, P., Hutchin, K., French, L., & Craissati, J. (2013). Developmental and offence-related characteristics of different types of adolescent sex offender: A community sample. Journal of Sexual Aggression, 19(2), 138-157. https://dx.doi.org/10.108 0/13552600.2011.645889
Tardif, M. (2015). Processus de socialisation et délinquance sexuelle [Socialisation and sexual offending]. In M. Tardif (Eds.), La délinquance sexuelle des mineurs : théories et recherches (vol. 2, p. 187-307). Montréal : Les Presses de l’Université de Montréal. https://dx.doi.org/10.4000/books.pum.2880
Tison, P., Hautekeete, M., Taquet, P., & Thouvenin, M. (2009). Les alcoolodépendants présentent des différences dans les croyances addictives selon leur motivation au changement [Alcohol addicts present different addiction related beliefs depending on their motivation to change]. Journal de thérapie comportementale et cognitive, 19(1), 35-40. https://dx.doi.org/10.1016/j.jtcc.2009.04.002
Tougas, A. M., & Tourigny, M. (2013). L'étude des mécanismes de changement, une avenue de recherche prometteuse pour optimiser les programmes de traitement destinés aux jeunes en difficulté: enjeux conceptuels et méthodologiques [The study of mechanisms of change as a promising avenue for optimizing treatment programs for troubled youth : conceptual and methodological challenges]. The Canadian Journal of Program Evaluation, 27(2), 61-86.
Veneziano, C., & Veneziano, L. (2002). Adolescents sex offenders : A review of the literature. Trauma, Violence & Abuse, 3(4), 247-260. https://dx.doi.org/10.1177/1524838002237329
Viens, S., Tourigny, M., Lagueux, F., & Étienne, L. (2012). Évaluation des effets d'une thérapie de groupe pour adolescents auteurs d'agressions sexuelles [Assessing the effects of group therapy among juvenile sex offenders]. Criminologie, 45(1), 243-270.
Viljoen, J. L., Gray, A. L., Shaffer, C., Latzman, N. E., Scalora, M. J., & Ullman, D. (2015). Changes in J-SOAP-II and SAVRY scores over the course of residential, cognitive-behavioral treatment for adolescent sexual offending. Sexual abuse : A Journal of Research and Treatment, 1-33. https://dx.doi.org/10.1177/1079063215595404
Walker, D. F., McGovern, S. K., Poey, E. L., & Otis, K. E. (2004). Treatment effectiveness for male adolescent sexual offenders: A meta-analysis and review. Journal of Child Sexual Abuse, 13(3-4), 281-293. https://dx.doi.org/10.1300/J070v13n03_14
Worling, J. R., & Langström, N. (2006). Risk of sexual recidivism in adolescents who offend sexually : Correlates and assessment. In H. E. Barbaree & W. L. Marshall (Eds.) The Juvenile Sex Offender (2e ed., p. 219-247). The Guilford Press.
Tables and Figures
Table I : Pre-post treatment analysis on sexual offending related factors and magnitude of change
Average (standard deviation)
Magnitude of change
Note. * p < 0,05
Table II : Average URICA scores in treatment
Table III : Analysis results pre and post treatment for trauma symptoms and magnitude of change
Average (standard deviation)
Magnitude of change
Note. * p < 0,05
Table IV : Correlations between percentages of change on the TSCC and percentages of change on the MESSY and the SSQ (n = 41) a
Note. TSCC = Trauma Symptom Checklist for Children ; MESSY = Matson’s Evaluation of Social Skills in Yougsters ; SSQ = Social Support Questionnaire ; PT Stress = Post-traumatic stress
a Incomplete observations are excluded n = 41
* p < 0,05 ; ** p < 0,01
Table V : Correlations between percentages of change on the TSCC and percentage of change on the CISS and the QCA (n = 41) a
Note. TSCC = Trauma Symptom Checklist for Children ; CISS = Coping Inventory for Stressful Situations ; QCA = Questionnaire des Cognitions pour Adolescents ;PT Stress = Post-traumatic stress
a Incomplete observations are excluded n = 41
* p < 0,05