Sex offender risk assessment in the Netherlands: Towards a risk need responsivity oriented approach Smid, W.

Transcription

1 UvA-DARE (Digital Academic Repository) Sex offender risk assessment in the Netherlands: Towards a risk need responsivity oriented approach Smid, W. Link to publication Citation for published version (APA): Smid, W. J. (2014). Sex offender risk assessment in the Netherlands: Towards a risk need responsivity oriented approach General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible. UvA-DARE is a service provided by the library of the University of Amsterdam ( Download date: 07 Jan 2019

2 Chapter 7 DISCUSSion

3 CHAPTER 7 Chapter 7 Discussion 136

4 DISCUSSION The Risk Need Responsivity (RNR; Andrews, Bonta, & Hoge, 1990; Bonta & Andrews, 2007) principles have proven crucial to the success of correctional interventions (Andrews & Bonta, 2006; French & Gendreau, 2006; Hanson, Bourgon, Helmus, Hodgson, 2009; Landenberger & Lipsey, 2005; Lowenkamp, Latessa, & Holsinger, 2006; Marlowe, Festinger, Lee, Dugosh, & Benasutti, 2006; Wilson, Bouffard, & Mackenzie, 2005). The aim of the present thesis was to assess the effect of sex offender policies in the Netherlands in terms of the Risk principle. To what extent is current practice successful in identifying offenders with the highest recidivism risk; to what extent do they receive treatments of matching intensity; and to what extent do they benefit from their treatment? This final chapter summarizes the main findings of the five studies in this thesis, provides an appraisal of methodological strengths and limitations, and discusses several specific implications for policy and practice, along with suggestions for future research. Main findings The first study (Chapter 2) of this thesis investigated whether the Dutch method of treatment referral resulted in an adequate match between sex offenders retrospectively assessed actuarial risk levels and their treatment referrals. Our findings showed that clinically derived treatment referral was, at best, only moderately in line with the risk principle of the RNR model, as sex offenders of all risk levels were referred to all types of treatment. This led most notably to under-treatment of high-risk rapists and over-treatment of low-risk child molesters. Over a quarter of all convicted rapists returned to society with untreated moderately-high to high risk levels and over a quarter of all convicted child molesters received far more intensive treatments than warranted by their risk levels. The second study (Chapter 3) examined how unstructured clinical treatment selection for sexual offenders affects the group composition in terms of risk levels of the comprising members. Results showed that not only were sex offenders of all risk levels referred to all intensities of treatment, clinical compilation of treatment groups also led to a mixing of all risk levels within the same treatment group. Moreover, treatment duration in the open format group of study was not related to sex offenders risk level. These findings imply that either the low-risk sex offenders were overtreated or the high-risk sex offenders were undertreated. Furthermore, the mixing of risk levels within a treatment group carries the possibility of affiliation between high-risk and low-risk offenders. Several authors have noted the risk of this mixing of low and high-risk offenders as a possible detrimental factor for the recidivism risk in low-risk offenders: Such mismatched treatment intensity may inadvertently promote recidivism risk in 137

5 CHAPTER 7 (originally) low risk offenders (Bonta, Wallace-Capretta, & Rooney, 2000; Lowenkamp & Latessa, 2002). The third study (Chapter 4) assessed which instruments are feasible for the purpose of risk assessment in the Netherlands. The nine internationally most commonly used instruments were administered to a large and representative sample of convicted Dutch sex offenders with more than 10 years follow up time to assess reoffending. Inspection of AUC values and correlations showed that most of the included sex offender risk assessment instruments contributed significantly to the prediction of both sexualand violent (including sexual) recidivism. Several significant differences were observed between the selected instruments regarding estimates of aspects of their validity and reliability. Most notably, the RRASOR (Hanson, 1997) and SVR-20 (Boer, Hart, Kropp, & Webster, 1997) performed least well, whereas Static-2002R (Hanson & Thornton, 2003; Helmus, Thornton, Hanson, & Babchishin, 2012) and Static-99R (Hanson & Thornton, 2000; Helmus et al., 2012) consistently showed the strongest results. The fourth study (Chapter 5) evaluated the effect of inpatient treatment (TBS, Ter Beschikking van de Staat, TBS; at the disposition of the state, i.e., state mandated, involuntary treatment) in terms of sexual and/or violent recidivism. This study, which followed the guidelines of the Collaborative Outcome Data Committee (CODC; 2007a, 2007b), showed that, after controlling for actuarial risk level, sex offenders who were referred to this high intensity treatment recidivated significantly less upon discharge than sex offenders who were discharged from prison. More specifically, moderate-high and high-risk sex offenders who underwent TBS treatment recidivated significantly less upon discharge than untreated moderate-high and high-risk sex offenders discharged from prison. No differences were found between treated and untreated lower risk sex offenders. These results indicate that, in line with The Risk principle, the high intensity TBS treatment was most profitable for high risk sex offenders. The fifth study (Chapter 6) of this thesis took an empirical approach to the prioritization of child pornography notifications. The relatively new group of child pornography consumers is very large, but also known to predominantly consist of individuals who limit their offenses to consumption of child pornography (Seto & Hanson, 2011). However, a subgroup is also involved in the direct victimization of children in the form of contact offending, online offending, or the production of child pornography. These dual offenders are also at higher risk to recidivate (Eke, Seto, & Williams, 2011). The study aimed to statistically identify distinguishing characteristics of these individuals within the larger group of child pornography consumer notifications. Several distinguishing risk factors were observed, suggesting a small but deeply invested subgroup with higher levels of sexual deviance. Of note, targeting dual offenders is a relatively new subject of investigation, and cross validation of our findings is clearly indicated. 138

6 DISCUSSION Overall Conclusion Taking into account all the results in this thesis, the overall conclusion becomes inevitable that the Netherlands is not (yet) doing very well regarding sex offender risk assessment and treatment referral. On a more positive note, however, there is ample room for improvement. There are principal opportunities in the Netherlands to adjust sex offender policies in a way that will increase adherence to the risk principle and create the opportunity for a need oriented treatment tailored to the patients responsivity to lower risk and decrease the number of victims. Appraisal of Methodological Strengths and Limitations An important methodological strength of the studies described in chapter 2, 4 and 5 of this thesis, is that they include a relatively large sample, representative of all convicted and discharged sex offenders from various jurisdictions in the Netherlands. This sampling strategy makes it possible to generalize the present results to a broad group of offenders. Additionally, the relatively long follow up periods over which recidivism was assessed (chapters 4 and 5) compensates for the usual underreporting of sexual offenses. Moreover, the study described in chapter 4 applied nine risk assessment instruments to all offenders within the same sample, thus enabling direct comparison between the instruments. An important strength of the treatment evaluation described in chapter 5 is that, while not a randomized controlled trial, it was conducted consistent with the guidelines of the Collaborative Outcome Data Committee (CODC; 2007a, 2007b). Chapter 6, exploring the feasibility of prioritization of CP notifications was breaking new ground. Furthermore, the sample in this study included all notifications that constitute the police workload and not just the eventually convicted CP offenders. Hence forthcoming results can be directly related back to police practice. The included studies also showed important methodological limitations. First and foremost, incomplete and inconsistent judicial administration affected the quality of the available data for all the included studies. This pertains to missing information required for scoring items of the selected risk assessment instruments, incomplete information regarding recidivism for a number of deceased offenders, the absence of a national database for police arrests, and inadequate administration of treatment progress and reasons for treatment termination by treatment providers. Future Dutch sex offender research will be greatly helped by more structured, standardized and accurate administration of judicial files and treatment files. For without adequate administration it is difficult to perform adequate research and without adequate research it is impossible to design and implement optimal, evidence based practice. Another important limitation is the relatively small size of some of the subsamples. While the total 139

7 CHAPTER 7 samples of the included studies were sufficient and even relatively large, stratification into subgroups of offenders of certain risk levels sometimes resulted in relatively small subgroups. This was especially true for the subgroups of high-risk offenders and the subgroup of dual offenders among the CP-only suspects. Replication of the findings in larger samples is therefore recommended. Further limitations include the limited scope of the research samples in the studies described in chapter 3 and 6; chapter 6 included CP notifications in a single jurisdiction and chapter 3 included the composition of a single treatment group, albeit followed over a long time period. Finally, the treatment evaluation study described in chapter 5 did not include random assignment to the treatment condition and therefore carries the possibility that treatment and control conditions differed in ways not captured by our stratification. Balancing the strengths and limitations of the included studies however, the overall quality of the studies is sufficient to justify the articulation a number of specific implications. Implications of the findings To reduce the number of victims most effectively, it is imperative that sex offender policy is dictated by empirical evidence. At this moment, the empirical evidence strongly suggests that standard use of an actuarial instrument like the Static-2002R or the Static-99R for risk assessment and treatment referral is the best possible choice. The current findings also indicate that it is advisable to use risk assessment instruments in the composition of specific treatment groups to ensure reasonable homogeneity of risk levels within a group. Of course this point becomes less relevant if the adequacy of the original risk assessment at the time of treatment referral has improved. The choice for specific instruments is subject to change based on ongoing research, e.g. if other instruments or methodologies are developed that achieve greater validity. However, at this moment there is an abundance of empirical evidence, including the studies in this thesis, to conclude that actuarial instruments provide a better estimate of risk levels than clinical judgment alone, or the combination of risk assessment instruments and clinical judgment (Hanson & Morton-Bourgon, 2009; Michel et al., 2013; Storey, Watt, Jackson & Hart, 2012; Wormith, Hogg & Guzzo, 2012). This implies that the instrument scores and only the instrument scores should guide risk assessment and treatment referral. Any exceptional occurrence of clinical adjustment to these scores and/ or the resultant conclusions should be extremely well substantiated, with requisite references to empirical support. In terms of implementation, risk level based treatment referrals can be readily achieved by using a structured actuarial risk assessment instrument at every sexu- 140

8 DISCUSSION al conviction to assess offenders risk levels. The Static-2002R and Static-99R are pre-eminently suited for standard use in the Netherlands, as they combine favorable psychometric properties with relative ease of use. The coding of these instruments requires a one-day training and, after reading the case file, will take less than an hour to complete. Therefore, the introduction of standard sex offender risk assessment would incur low cost. Risk and Accountability The introduction of standard structured actuarial risk assessment for sex offenders in the Netherlands will require a significant change in conceptualizing forensic treatment referral. More specifically, it requires a paradigm shift from a disorder oriented- to a to risk-oriented approach. Ever since the installment of the so called psychopath law (psychopatenwet) in 1880, treatment referral has primarily been based on the presence of a diagnosable (offense related) disorder and the associated level of accountability of the offender. The presence of an offense related disorder results in diminished accountability, leading to a (partial) conversion of punishment into treatment. If no disorder can be diagnosed, full accountability will be presumed, in turn leading to full punishment without treatment. The prevailing rationale has been that disordered offenders need treatment instead of punishment, and that curing the disorder related to their offenses will reduce their risk of recidivism. Arguably, it can be justified to not (fully) punish offenders for crimes for which they cannot be held (fully) accountable due to a mental illness. The assumed link between disorder and risk however, cannot be justified. Accumulating evidence has indicated that disorder and recidivism risk are two separate entities, and the most severely disordered offenders are not necessarily the most dangerous offenders in terms of recidivism risk (e.g. Andrews, Bonta, & Wormith, 2006; Bonta, Law, & Hanson, 1998; Hanson & Morton-Bourgon, 1998; Harris, Rice, & Quinsey, 1993). Moreover, prison sentences for fully or largely accountable offenders do not always eliminate their recidivism risk upon release. Altogether, disorders and levels of accountability may be used as factors in the determination of the extent of the punishment, but should be independent of the determination of the level of intervention, as the latter is best determined by risk level. An important advantage of risk assessment-based treatment referral is the possibility to assess risk levels without significant cooperation of the offender, whereas the assessment of a disorder is often problematic without cooperation. Defendants, following the counsel of their solicitors, frequently refuse cooperation with psychological assessment in order to avoid being referred to TBS treatment. The stigma associated 141

9 CHAPTER 7 with TBS, its indefinite treatment duration, and the increasingly stricter conditions for discharge have caused defendants to prefer even a long prison sentence (Heidanus, 2012). Currently, the refusal to cooperate with psychological assessment will result in the absence of a diagnosis and consequentially full accountability, and no referral for treatment. This process causes sex offenders of sometimes high risk levels to eventually return to society untreated. In case of the suggested risk oriented approach, refusal to cooperate with psychological evaluation will offer no advantages to the offenders. Disorder and accountability may guide punishment while, on the other hand, risk will be related to the selection of the appropriate intervention. Refusing psychological evaluation may only result in the maximum penalty because no mitigating circumstances in the form of diminished accountability can be assessed. However, refusal will not interfere with treatment referral because actuarial risk assessment instruments can still be applied, and the appropriate level of intervention can be derived from the resulting risk levels. Treatment The results of the treatment evaluation (Chapter 6) add to the accumulating evidence that treatment levels should be tailored to risk levels. High intensity treatment like TBS should best be reserved for sex offenders of the highest risk levels, as they benefit most from it. Standard use of structured risk assessment will result in more accurate identification of these high-risk sex offenders and treatment referral will be more in line with the RNR principles. Another benefit of the standard use of sex offender risk assessment will be the reduction in cost associated with reducing treatment levels for the low risk offenders. Reserving resources for those offenders most in need of (intensive) interventions seems highly opportune. If, for whatever special circumstances (e.g. because policymakers feel forced to act under the pressure of public opinion), treatment of low-risk offenders appears inevitable, these offenders should at least be treated separately from the offenders with higher risk levels and be provided with treatments of minimal duration and intensity in order to not inadvertently increase their risk, and to save resources for where they are needed the most. By using resources in this more efficient and effective manner, a decrease in the total number of victims can be expected. It is essential that the overall effect of these measures will be assessed taking all sex offenses into account. For when more offenders of higher risk levels are referred to TBS treatment, the absolute number of TBS-incidents might well increase, which may lead to the misperception that the measure is counter- 142

10 DISCUSSION productive. However, recidivism among offenders discharged from prison is expected to decrease much more steeply thus yielding an overall reduction in sex offenses. Future Directions Risk assessment, even state-of-the-art standard actuarial risk assessment, will inevitably remain a fallible enterprise, i.e. it will not eliminate all victimization. Nevertheless, a number of research directions aimed at further improvement may be suggested. First, conceivably, instruments specifically tailored to the Dutch offender population may show superior predictive validity over internationally translated instruments. Another direction would be to develop instruments specifically designed for particular offender groups, such as rapists, child molesters, or CP-only offenders. Although ample research targeted the identification of predictors of recidivism, it remains possible that other risk factors will prove predictive when better operationalized. More specifically, direct assessment of internal characteristics of the offender may add predictive validity over the currently included behavioral correlates of these characteristics. Preferably, the assessment has to be operationalized in ways not dependent on selfreport, to achieve the required reliability. For example, numerous studies have focused on the implicit assessment of deviant sexual interests by means of psychophysiological or neuropsychological testing, such as penile plethysmography (e.g. Marshall, 2014), implicit association task (IAT; e.g. Gray, Brown, MacCulloch, Smith, & Snowden, 2005), Viewing Time measures (VT; e.g. Gress, 2005), or neuroimaging procedures (fmri; e.g. Ponseti et al., 2009) (for a comprehensive overview see Akerman & Beech, 2012). It would be of interest to further develop these measures as well as to develop reliable and valid measures of other internal factors such as offense supportive attitudes/ cognitive distortions or even victim empathy. However, the inclusion of new risk factors or the commissioning of new instruments should always be guided by empirical evidence in support of their validity. Furthermore, the RNR model contains two other crucial elements of successful correctional intervention. The research in this thesis was focused on the risk principle and the assessment of basic risk levels. Once a risk-based selection has been made, and offenders of various risk levels have been referred to treatments of matching intensity, the following step consists of determining treatment focus. Ideally, treatment is focused on the dynamic risk factors, i.e., the offenders criminogenic needs that are changeable, open to the influence of treatment (Bonta & Andrews, 2007). There are several instruments available that assess empirically derived dynamic risk factors, but much more research is needed into the exact nature of the dynamic risk factors, the best way to assess these, how to target these factors in treatment, and the covariation 143

11 CHAPTER 7 between a reduction in dynamic risk factors and actual reductions of recidivism (Serin, Lloyd, Helmus, Derkzen & Luong, 2013). Better knowledge of dynamic risk factors can improve treatment focus and thereby not only enhance treatment effectiveness, but also reduce treatment duration. This seems particularly relevant for the TBS treatment, which, with an average duration nearing 10 years, is unparalleled in the rest of the world. In sum, comprehensive study of the assessment and treatment of dynamic risk factors is the obvious next step in sex offender research. Finally, in addition to the assessment of static or dynamic risk factors for reoffending, there is a need for evidence based etiological theory of sexual offending. Of note, most sex offenses are first time offenses, and prevention of recidivism thus has a ceiling effect. Primary prevention is essential to reduce the number of victims of sex offenses. To become more effective at primary prevention, the field needs a better understanding of the etiology of sexual offending behavior: What motivates individuals to commit a sexual offense in the first place? The accumulation of knowledge and understanding about sexual offending behavior through empirical research will enable us to develop policies that help decrease sexual victimization and promote a safer society. 144

12 DISCUSSION References Akerman, G., & Beech, A. R. (2012). A systematic review of measures of deviant sexual interest and arousal. Psychiatry, Psychology and Law, 19, doi: / Andrews, D. A., Bonta, J., & Hoge, R. D. (1990). Classification for effective rehabilitation: Rediscovering psychology. Criminal Justice and Behavior, 17, Andrews, D., Bonta, J., & Wormith, S. (2006). The recent past and near future of risk and/or need assessment. Crime and Delinquency, 52, doi: / Boer, D. P., Hart, S. D., Kropp, P. R., & Webster, C. D. (1997). Manual for the Sexual Violence Risk-20: Professional guidelines for assessing risk of sexual violence. Vancouver, BC: British Columbia Institute Against Family Violence. Bonta, J., & Andrews, D. A. (2007). Risk-need-responsivity model for offender assessment and rehabilitation (Corrections Research User Report No ). Ottawa, Ontario: Public Safety Canada. Bonta, J., Law, M., & Hanson, K. (1998). The prediction of criminal and violent recidivism among mentally disordered offenders: a meta-analysis. Psychological bulletin, 123, doi: / Bonta, J., Wallace-Capretta, S., & Rooney, J. (2000). A quasi-experimental evaluation of an intensive rehabilitation supervision program. Criminal Justice and Behavior, 27, doi: / Collaborative Outcome Data Committee. (2007a). The Collaborative Outcome Data Committee s guidelines for the evaluation of sexual offender treatment outcome research Part 2: CODC guidelines (Corrections Research User Report No ). Ottawa, Ontario: Public Safety Canada. Collaborative Outcome Data Committee. (2007b). Sexual offender treatment outcome research: CODC guidelines for evaluation Part 1: Introduction and overview (Corrections Research User Report No ). Ottawa, Ontario: Public Safety Canada. Eke, A. W., Seto, M. C., & Williams, J. (2011). Examining the Criminal History and Future Offending of Child Pornography Offenders. Law and human behavior, 35, doi: /s French, S. A., & Gendreau, P. (2006). Reducing prison misconducts: What works! Criminal Justice and Behavior, 33, doi: / Gray, N. S., Brown, A. S., MacCulloch, M. J., Smith, J., & Snowden, R. J. (2005). An implicit test of the associations between children and sex in paedophiles. Journal of Abnormal Psychology, 114, doi: / X Gress, C. L. (2005). Viewing time measures and sexual interest: Another piece of the puzzle. Journal of Sexual Aggression, 11, doi: / Hanson, R. K. (1997). The development of a brief actuarial risk scale for sexual offense recidivism. (User Report 97-04). Ottawa, ON: Department of the Solicitor General of Canada. Hanson, R. K., Bourgon, G., Helmus, L., & Hodgson, S. (2009). The Principles of Effective Correctional Treatment Also Apply To Sexual Offenders: A Meta-Analysis. Criminal Justice and Behavior, 36, doi: / Hanson, R. K., & Morton-Bourgon, K. E. (2009). The accuracy of recidivism risk assessments for sexual offenders: A meta-analysis of 118 prediction studies. Psychological Assessment, 21, doi: /a Hanson, R. K., & Thornton, D. (2003). Notes on the development of Static Corrections Research (User Report No ). Ottawa, ON: Department of the Solicitor General of Canada. Heidanus, N. A. (2013). De raadsman en de TBS. In Ontmoetingen: Voordrachtenreeks van het Lutje Psychiatrisch-Juridisch Gezelschap [The counselor and TBS. In Encounters: Lectures of the Lutje Psychiatric-Judicial Society], 17, 21 29, Groningen, Netherlands: University of Grongingen. Helmus, L., Thornton, D., Hanson, R. K., & Babchishin, K. M. (2012). Improving the predictive validity of Static-99 and Static-2002 with older sex offenders: Revised age weights. Sexual Abuse: A Journal of Research and Treatment, 24, doi: / Jolliffe, D., & Farrington, D. P. (2004). Empathy and offending: A systematic review and meta-analy- 145

13 CHAPTER 7 sis. Aggression and violent behavior, 9, doi: /j.avb Landenberger, N. A., & Lipsey, M. W. (2005). The positive effects of cognitive-behavioral programs for offenders: A meta-analysis of factors associated with effective treatment. Journal of Experimental Criminology, 1, doi: /s Lowenkamp, C. T., & Latessa, E. J. (2002). Evaluation of Ohio s community based correctional facilities and halfway house programs [Technical report]. Cincinnati, OH: University of Cincinnati. Lowenkamp, C. T., Latessa, E. J., & Holsinger, A. M. (2006). The risk principle in action: What have we learned from 13,676 offenders and 97 correctional programs? Crime and Delinquency 52, doi: / Marlowe, D. B., Festinger, D. S., Lee, P. A., Dugosh, K. L., & Benasutti, K. M. (2006). Matching judicial supervision to clients risk status in drug court. Crime & Delinquency, 52, doi: / Marshall, W. L. (2014). Phallometric Assessments of Sexual Interests: An Update. Current psychiatry reports, 16, 1 7. doi: /s Michel, S. F., Riaz, M., Webster, C., Hart, S. D., Levander, S., Müller-Isberner, R.,... & Hodgins, S. (2013). Using the HCR-20 to Predict Aggressive Behavior among Men with Schizophrenia Living in the Community: Accuracy of Prediction, General and Forensic Settings, and Dynamic Risk Factors. International Journal of Forensic Mental Health, 12, doi: / Ponseti, J., Granert, O., Jansen, O., Wolff, S., Mehdorn, H., Bosinski, H., & Siebner, H. (2009). Assessment of Sexual Orientation Using the Hemodynamic Brain Response to Visual Sexual Stimuli. The journal of sexual medicine, 6, doi: /j x. Serin, R. C., Lloyd, C. D., Helmus, L., Derkzen, D. M., & Luong, D. (2013). Does intra-individual change predict offender recidivism? Searching for the Holy Grail in assessing offender change. Aggression and Violent Behavior, 18, doi: /j.avb Seto, M. C. & Hanson, R. K. (2011). Introduction to special issue on internet-facilitated sexual offending. Sexual Abuse: A Journal of Research and Treatment, 23, 3 6. Storey, J. E., Watt, K. A., Jackson, K. J., & Hart, S. D. (2012). Utilization and implications of the Static-99 in practice. Sexual Abuse: A Journal of Research and Treatment, 24, doi: / Wilson, D. B., Bouffard, L. A., & Mackenzie, D. L. (2005). A quantitative review of structured, group-oriented, cognitive-behavioral programs for offenders. Criminal Justice and Behavior, 32, doi: / Wormith, J. S., Hogg, S., & Guzzo, L. (2012). The predictive validity of a general risk/needs assessment inventory on sexual offender recidivism and an exploration of the professional override. Criminal Justice and Behavior, 39, doi: /