Elsevier

Child Abuse & Neglect

Volume 24, Issue 7, July 2000, Pages 965-982
Child Abuse & Neglect

Adolescent sexual offender recidivism: success of specialized treatment and implications for risk prediction1,

https://doi.org/10.1016/S0145-2134(00)00147-2Get rights and content

Abstract

Objective: To evaluate the success of specialized community-based treatment for reducing adolescent sexual reoffending and explore the predictive utility of variables assessed regarding sexual and nonsexual recidivism.

Method: Recidivism data (criminal charges) were collected for 58 offenders participating in at least 12 months of specialized treatment at the SAFE-T Program. Data were also collected for a comparison group of 90 adolescents who received only an assessment (n = 46), refused treatment (n = 17), or dropped out before 12 months (n = 27). Follow-up interval ranged from 2 to 10 years (M = 6.23, SD = 2.02). Offenders completed a battery of psychological tests to provide standardized data regarding social, sexual, and family functioning.

Results: Recidivism rates for sexual, violent nonsexual, and nonviolent offenses for treated adolescents were 5.17%, 18.9%, and 20.7%, respectively. The Comparison group had significantly higher rates of sexual (17.8%), violent nonsexual (32.2%), and nonviolent (50%) recidivism. Sexual recidivism was predicted by sexual interest in children. Nonsexual recidivism was related to factors commonly predictive of general delinquency such as history of previous offenses, low self-esteem, and antisocial personality.

Conclusions: Results support the efficacy of treatment for adolescent sexual offenders and are consistent with the notion that sexual recidivism is predicted by unique factors unrelated to general (nonsexual) reoffending.

Resumen

Objetivo: Evaluar el éxito de un programa de tratamiento especializado comunitario para reducir el recidivismo en adolescentes ofensores sexuales, y para explorar la utilidad predictiva de una serie de variables en relación con el recidivismo sexual y no sexual.

Método: Se recogieron una serie de datos sobre el recidivismo (cargos criminales) de 58 ofensores que participaron en un programa de tratamiento especializado (Programa SAFE-T) durante al menos doce meses. También se recogieron datos de un grupo comparación formado por 90 adolescentes que únicamente recibieron una evaluación (n = 46), que rechazaron el tratamiento (n = 17) o que lo abandonaron antes de los doce meses (n = 27). Se relaizó un seguimiento de los adolescentes con un rango entre dos y diez años (M = 6.23, SD = 2.02). Los ofensores cumplimentaron una baterı́a de tests psicológicos que permitió obtener datos normalizados sobre su funcionamiento social, sexual y familiar.

Resultados: Las tasas de recidivismo de agresiones sexuales, agresiones violentas de carácter no sexual, y agresiones no violentas, para los adolescentes que recibieron tratamiento fue del 5.17%, 18.9% y 20.7% respectivamente. El grupo comparación tuvo tasas significativamente más altas de recidivismo para agresiones sexuales (17.8%), agresiones violentas de carácter no sexual (32.2%) y agresiones no violentas (50%). El recidivismo de agresiones sexuales fue predicho por el intrés sexual en niños/as. El recidivismo no sexual estuvo relacionado con factores habitualmente predictores de la delincuencia general, tales como una historia de agresiones previas, baja autoestima, y personalidad antisocial.

Conclusiones: Los resultados apoyan la eficacia del tratamiento para adolescentes ofensores sexuales y son consistentes con la noción de que el recidivismo sexual puede ser predicho por factores únicos no relacionados con el recidivismo general (de carácter no sexual).

Introduction

MANY ADULT SEXUAL offenders begin offending as adolescents and are not detected for many years—if at all Abel et al 1985, Briggs and Hawkins 1996, Elliott et al 1995, Groth et al 1982, Knight and Prentky 1993. Of course, some adolescent sexual offenders are apprehended, and approximately 20% of all people charged with a sexual offense in North America are juveniles Federal Bureau of Investigation 1993, Statistics 1997. In response to the obvious need to reduce the risk of adolescent sexual reoffending, the number of treatment programs has increased significantly in North America during the past 20 years. In the US, for example, although there was only one specialized treatment program for adolescent sexual offenders in 1975 (Knopp, 1985), there were over 600 by 1995 (Freeman-Longo, Bird, Stevenson, & Fiske, 1995).

There is notable consensus regarding the issues to be addressed in the specialized treatment of adolescent sexual offenders Association for the Treatment of Sexual Abusers 1997, National Adolescent Perpetrator Network 1993. Treatment goals include increasing offender accountability; assisting offenders to understand and interrupt the thoughts, feelings, and behaviors that maintain sexual offending; reducing deviant sexual arousal, if present; improving family relationships; enhancing victim empathy; improving social skills; developing healthy attitudes towards sex and relationships; and reducing the offenders’ personal trauma, if present. Despite the agreement regarding treatment goals, and the recent proliferation of treatment programs, however, little is known about the success of specialized treatment.

Since the development of the first comprehensive treatment program for adolescent sexual offenders in 1975, there have been only 10 published reports of criminal recidivism following specialized treatment (see Table 1). Although sexual recidivism rates for treated offenders are below 15% across studies, there are a number of factors that make it difficult to conclude that treatment is effective. First, 8 of the 10 investigations did not include a comparison group; therefore, it is impossible to ascertain the relative impact of treatment on recidivism. Second, no investigation used a mean follow-up period beyond 4 years, and many studies had a mean follow-up period of less than 3 years. Given that sexual offenders may present a risk of reoffending for many years after treatment Furby et al 1989, Hanson et al 1993, such brief follow-up periods will necessarily result in low recidivism rates. Third, most investigators have used criminal convictions or offender self-report as the measure of recidivism. Recidivism data based on official reports are inevitably conservative estimates of sexual reoffending given that many victims never report their abuse (e.g., Statistics Canada, 1993); however, the use of criminal convictions or offender self-report is likely to produce significantly lower estimates of reoffending than the use of criminal charges (Hanson, 1997). Indeed, the sexual assault recidivism rate for treated offenders from studies in Table 1 using criminal charges is 14% (17 out of 120) whereas the recidivism rate from studies using convictions or self-report is significantly lower at 8% (52 out of 635), χ2 (1, N = 755) = 4.33, p < .05 (data from Kahn & Lafond, 1988, were excluded as the recidivism measure was unspecified in their study). Finally, although it is likely that some adolescent offenders moved to another state following treatment (all published investigations have been conducted in the United States), investigators have used only local and/or state records rather than a national registry.

In contrast to the relative dearth of follow-up studies with adolescents, there have been many published investigations of the success of specialized treatment for adult sexual offenders. Although some reviewers have noted that there is no empirical evidence to support the efficacy of specialized sexual offender treatment (e.g., Furby et al 1989, Quinsey et al 1993), others have suggested that there is at least minimal empirical support (e.g., Hall 1995, Marshall and Pithers 1994). Miner (1997) observed that methodological shortcomings make it difficult to draw conclusions regarding treatment outcome. In particular, he noted that many investigators studying the efficacy of adult sexual offender treatment have employed a single-group, follow-up design. When untreated offenders have been included, Miner noted that it is common for researchers to use nonequivalent comparison groups such as treatment dropouts or refusers. Hanson (1997) argued that if random assignment to treatment is not possible, it is important for researchers who choose to examine such “convenience” comparison groups to assess and control for pre-treatment differences between the groups on factors known to be related to recidivism. In a recent meta-analysis, Hanson and Bussière (1998) found that risk of sexual recidivism among primarily adult male sexual offenders was associated with factors related to sexual offending such as victim age (children), victim gender (male), victim relationship to offender (extrafamilial), prior sexual and nonsexual offending, and sexual preference for children. Conversely, the authors found that nonsexual reoffenses were associated with factors commonly ascribed to nonsexual recidivism (e.g., Andrews and Bonta 1994, Gendreau et al 1996). Specifically, those sexual offenders with nonsexual reoffenses tended to be younger, single, and had a history of antisocial behavior.

Although there are no comparable meta-analyses regarding sexual recidivism for adolescent sexual offenders, there is an extensive literature related to general (i.e., nonsexual) recidivism with adolescents (e.g., Farrington 1989, Loeber 1990, Moffitt 1993). Common risk factors for general recidivism among adolescents include antisocial personality, previous criminal involvement, negative self-image, economic disadvantage, parental rejection, negative parent-child relationships, interpersonal aggression, and poor social relationships.

The purpose of the present study was to examine the success of specialized adolescent sexual offender treatment by comparing subsequent recidivism rates between treated offenders and a comparison group. Although it was not possible to use random assignment at the inception of this study (in 1987), we ensured that a comprehensive battery of psychological tests was used in our clinical assessments. In addition to the clinical utility of these results, we planned to use test scores to control for any significant pre-treatment differences between the treatment and comparison groups on the variables thought to be related to recidivism.

A secondary goal of this study was to examine the predictive utility of the variables assessed with respect to both sexual and nonsexual recidivism. Given the recent focus in the literature regarding the prediction of sexual recidivism with adult sexual offenders (e.g., Hanson and Bussiere 1998, Proulx et al 1997, Quinsey et al 1995), and the absence of empirical data regarding risk prediction for adolescent sexual offenders Bonner et al 1998, Ryan 1998, we wanted to explore the relationships between a number of predictor variables and both sexual and nonsexual recidivism. We anticipated that nonsexual recidivism would be related to factors predictive of general (nonsexual) juvenile delinquency, listed above, such as history of prior offenses, antisocial personality, low self-esteem, and family relationship difficulties, for example. Based on previous research with adult sexual offenders, we expected that sexual recidivism would be related specifically to sexual-assault variables such as deviant sexual interest/behavior and victim-selection factors (i.e., victim age, gender, and relationship).

Section snippets

Treatment program

The Sexual Abuse, Family Education and Treatment (SAFE-T) Program is a specialized, community-based program that provides sexual abuse specific assessment, treatment, consultation, and long-term support to (1) child victims of incest and their families—including adult incest offenders, (2) children with sexual behavior problems and their families, and (3) adolescent sexual offenders and their families. Following comprehensive clinical and psychometric assessments, treatment plans are

Results

Before comparing recidivism rates, the Treatment group and the three Comparison groups were contrasted on a number of potentially confounding variables to ensure that the groups were not significantly different. The data related to these comparisons are presented in Table 2 (non-dichotomous variables) and Table 3 (dichotomous variables). Remarkably, there were no significant group differences on any of the factors that have been linked to the risk of sexual or nonsexual recidivism; therefore,

Discussion

Results support the efficacy of specialized community-based treatment at the SAFE-T Program for reducing the risk of adolescent sexual recidivism. Relative to the Comparison group, there was a 72% reduction in sexual recidivism for adolescents completing at least 12 months of assessment and treatment. Furthermore, although previous research has found that many treated sexual offenders are likely to be charged with subsequent nonsexual offenses, participation in specialized treatment was

Acknowledgements

This study was possible only as a result of the dedication of the staff at the SAFE-T Program and through the courage of the adolescent sexual offenders and their families. We are also grateful to the operational support of the Thistletown Regional Centre, and we thank Adrienne Perry and Sabrina Ramdeholl for their valuable comments on a previous draft.

References (81)

  • J.R Worling

    Adolescent sibling-incest offendersDifferences in family and individual functioning when compared to nonsibling sex offenders

    Child Abuse & Neglect

    (1995)
  • G.G Abel et al.

    Sexual offendersResults of assessment and recommendations for treatment

  • T.M Achenbach

    Manual for the Youth Self-Report and 1991 Profile

    (1991)
  • D.A Andrews et al.

    The psychology of criminal conduct

    (1994)
  • Association for the Treatment of Sexual Abusers. (1997). Ethical standards and principles for the management of sexual...
  • J.D Atcheson et al.

    A study of juvenile sex offenders

    American Journal of Psychiatry

    (1954)
  • A.T Beck et al.

    Beck Depression Inventory manual

    (1987)
  • J.V Becker

    Treating adolescent sexual offenders

    Professional Psychology: Research and Practice

    (1990)
  • B.R Blishen et al.

    Sex differences in a socioeconomic index for occupations in Canada

    Canadian Review of Sociology and Anthropology

    (1978)
  • B.R Blishen et al.

    A revised socioeconomic index for occupations in Canada

    Canadian Review of Sociology and Anthropology

    (1976)
  • B.L Bonner et al.

    Assessment of adolescent sexual offenders

    Child Maltreatment

    (1998)
  • C.M Borduin et al.

    Multisystemic treatment of adolescent sexual offenders

    International Journal of Offender Therapy and Comparative Criminology

    (1990)
  • M.L Bourke et al.

    Assessment and treatment of juvenile sex offendersAn empirical review

    Journal of Child Sexual Abuse

    (1966)
  • J.M Brannon et al.

    Adolescent sex offendersInvestigating adult commitment-rates four years later

    International Journal of Offender Therapy and Comparative Criminology

    (1995)
  • J.F Bremer

    Serious juvenile sex offendersTreatment and long-term follow-up

    Psychiatric Annals

    (1992)
  • D.L Burton et al.

    The 1996 Safer Society Foundation Treatment Provider and Program Survey

    (1998)
  • B.J Bushman et al.

    Meta-analysis of factor analysesAn illustration using the Buss-Durkee Hostility Inventory

    Personality and Social Psychology Bulletin

    (1991)
  • A.H Buss et al.

    An inventory for assessing different kinds of hostility

    Journal of Clinical Psychology

    (1957)
  • D.P Farrington

    Early predictors of adolescent aggression and adult violence

    Violence and Victims

    (1989)
  • Uniform crime reports for the United States

    (1993)
  • R.E Freeman-Longo et al.

    1994 Nationwide survey of treatment programs and models

    (1995)
  • L Furby et al.

    Sex offender recidivismA review

    Psychological Bulletin

    (1989)
  • P Gendreau et al.

    A meta-analysis of the predictors of adult offender recidivismWhat works!

    Criminology

    (1996)
  • H.G Gough

    Californian Psychological InventoryAdministrator’s guide

    (1987)
  • J.B Greenhouse et al.

    An introduction to survival analysisStatistical methods for analysis of clinical trial data

    Journal of Consulting and Clinical Psychology

    (1989)
  • A.N Groth et al.

    Undetected recidivism among rapists and child molesters

    Crime and Delinquency

    (1982)
  • M.P Hagan et al.

    A comparison of treatment outcomes between adolescent rapists and child sexual offenders

    International Journal of Offender Therapy and Comparative Criminology

    (1996)
  • G.C.N Hall

    Self-reported hostility as a function of offense characteristics and response style in a sexual offender population

    Journal of Consulting and Clinical Psychology

    (1989)
  • G.C.N Hall

    Sexual offender recidivism revisitedA meta-analysis of recent treatment studies

    Journal of Consulting and Clinical Psychology

    (1995)
  • R.K Hanson

    How to know what works with sexual offenders

    Sexual Abuse: A Journal of Research and Treatment

    (1997)
  • Cited by (233)

    • Comparing the onset of child sexual abuse perpetration from adolescence into adulthood: Are there unique risks, and what does this mean for prevention?

      2020, Child Abuse and Neglect
      Citation Excerpt :

      Current estimated lifetime prevalence rates of child sexual abuse (CSA) range between 18–20 % for girls and 7–8 % for boys, and as high as 38 % in some African countries (Pereda, Guilera, Forns, & Gómez-Benito, 2009; Stoltenborgh, van IJzendoorn, Euser, & Bakermans-Kranenburg, 2011; World Health Organisation, 2014).

    • Safeguarding child rights and enhancing caregiver responsibilities among Canadian parents of youth who sexually offend

      2018, Child Abuse and Neglect
      Citation Excerpt :

      However, Hackett et al. (2014) noted that some caregivers’ responses shifted over time with professional guidance so that they became more committed to supporting the needs of their child. Although the responsibility of caregivers whose child has engaged in sexual offending behaviour is an emerging area of study, existing findings echo other literature that has emphasized the importance of involving caregivers in the treatment of sexual offending behaviour as well as the nature of this involvement (Schmidt, 2014; Worling & Curwen, 2000; Worley et al., 2011; Yoder, Hansen, Lobanov-Rostovsky, & Ruch, 2015; Zankman & Bonomo, 2004). Moreover, the Youth Criminal Justice Act in Canada specifies that significant others (e.g., parents, extended family, community agencies) must be included in the youth’s rehabilitation (Government of Canada, 2002).

    • Best practice in sexual offender rehabilitation and reintegration programs

      2023, Journal of Criminological Research, Policy and Practice
    View all citing articles on Scopus

    This research was supported by a grant from the Toronto Area Office of Probation and Community Services, Ontario Ministry of Community and Social Services.

    1

    The views expressed in this article are the authors’ and do not necessarily represent the views of the Ontario Ministry of Community and Social Services.

    View full text