HTA Report Treatment for Convicted Adult Male Sex Offenders

Home , Anaphrodisiac, Chemical castration

HTA Report

Treatment for Convicted Adult Male Sex Offenders

July 2010 Institute of Health Economics The Institute of Health Economics (IHE) is an independent, not-for-profit organization that performs research in health economics and synthesizes evidence in health technology assessment to assist health policy making and best medical practices.

IHE Board of Directors Chair Dr. Lorne Tyrrell – Chair, Institute of Health Economics and Professor and CIHR/GSK, Chair in Virology, University of Alberta Government Mr. Jay Ramotar – Deputy Minister, Alberta Health and Wellness Ms. Annette Trimbee – Deputy Minister, Advanced Education and Technology Dr. Jacques Magnan – Interim President and CEO, Alberta Innovates Health Solutions Academia Dr. Renée Elio – Associate VP Research, University of Alberta Dr. Tom Feasby – Dean of Medicine, University of Calgary Dr. Philip Baker – Dean of Medicine, University of Alberta Dr. Tom Noseworthy – Professor and Head, Community Health Sciences, University of Calgary Dr. James Kehrer – Dean of Pharmacy, University of Alberta Dr. Herb Emery Svare – Chair, Health Economics, University of Calgary Dr. Doug West – Chair, Department of Economics, University of Alberta Industry Mr. Terry McCool – Vice President, Corporate Affairs, Eli Lilly Canada Inc. Mr. Gregg Szabo – Executive Director, Corporate Affairs, Merck Frosst Canada Ltd. Dr. Bernard Prigent – Vice President & Medical Director, Pfizer Canada Inc. Mr. Grant Perry – Director, Public Affairs, GlaxoSmithKline Inc. Mr. William Charnetski – Vice President, Corporate Affairs, AstraZeneca Canada Inc. Other Mr. Chris Mazurkewich – Executive Vice President & CFO, Alberta Health Services CEO Dr. Egon Jonsson - Executive Director and CEO, Institute of Health Economics, Professor, University of Alberta, University of Calgary IHE Mr. Doug Gilpin – Chair, Audit & Finance Committee Board Secretary Dr. Egon Jonsson – Executive Director and CEO, Institute of Health Economics, Professor, University of Alberta, University of Calgary Treatment for convicted adult male sex offenders

Prepared by:

Paula Corabian, BSc, MPH

Maria Ospina, BSc, MSc

Christa Harstall, BSc, MLS, MHSA

Reproduction, redistribution, or modification of the information for any purposes is prohibited without the express written permission of the Institute of Health Economics

Treatment of Convicted Adult Male Sex Offenders i Acknowledgments The Institute of Health Economics is most grateful for the following individuals for review and provision of information and comments on the draft report. The views expressed in the final report are those of the Institute. —— Dr. R. Karl Hanson, Corrections Research, Public Safety Canada, Ottawa, Ontario —— Dr. David Moher, Ottawa Hospital Research Institute, Ottawa Methods Centre, The Ottawa Hospital General Campus, Ottawa, Ontario

Information Services Support The literature search for the review was undertaken by Ms Liz Dennett, Information Specialist, Institute of Health Economics, University of Alberta, Edmonton, Canada.

Competing interest The author(s) of this report declared no competing interest.

Production of this document has been made possible by a financial contribution from Health and Wellness and under the auspices of the Alberta Health Technologies Decision Process initiative: the Alberta Model for health technology assessment and policy analysis. The views expressed herein do not necessarily represent the official policy of Alberta Health and Wellness.

ii Treatment of Convicted Adult Male Sex Offenders EXECUTIVE SUMMARY Background Sexual offending has become a major challenge for social policy because of the high human and financial costs to victims and the social and health services as well as the high public investment in policing, prosecuting, and incarcerating sex offenders. There is an expectation that the correctional systems should make reasonable efforts to reduce the potential that convicted sex offenders will reoffend. One common approach to sex offenders’ management in countries with developed market economies is to provide specialized treatment programs. A number of different sex offender treatment (SOT) programs have been developed and are currently operating, but there continues to be controversy regarding how well they work.

Objectives To evaluate the effectiveness of psychotherapy and pharmacotherapy interventions delivered within SOT programs to reduce the likelihood of reoffending in convicted adult male sex offenders.

Results Eight systematic reviews (SRs) conducted on the effectiveness of SOT interventions and programs met the inclusion criteria of this overview. All eight SRs focused on the use of psychotherapy for convicted sex offenders, whereas one also included surgical castration and hormonal medication. In these studies there was considerable variability in how interventions were classified, the types of sex offenders involved, and the definition of outcomes. According to the reviewed evidence, studies in the area of SOT outcome research have improved over the past 10 years. However, the need for more rigour remains. The following are highlights from the reviewed evidence. —— Although the debate in the scientific literature on what SOT interventions and programs are most effective for convicted adult male sex offenders remains, the results from seven moderate- to-high quality SRs show small but statistically significant reductions in sexual and general recidivism rates among convicted adult male sex offenders treated with various cognitive behavioural therapy (CBT) approaches. —— The most recently published high quality SR found that SOT programs using CBT approaches that adhered to the risk/need/responsivity (RNR) model for offender assessment and rehabilitation were most effective in reducing the risk of recidivism in convicted male sexual offenders. On average, programs that followed all three RNR principles reported recidivism rates that were less than half the recidivism rates for comparison groups. In contrast, there was no effect on recidivism rates for programs that did not follow the RNR model.

Treatment of Convicted Adult Male Sex Offenders iii —— Confidence in these findings, however, must be tempered as the available evidence is based mostly on poor quality primary research studies. —— Although one SR of moderate quality reported promising results on the use of hormonal treatments as an adjuvant to psychotherapy, well- conducted and reported controlled studies are needed to establish the effectiveness of adjuvant hormonal treatment to reduce the risk of recidivism among sex offenders. —— Overall, the results reported by the selected SRs provide little direction regarding how to improve current treatment practices. • It is still not clear whether all sex offenders require treatment or whether current interventions are more appropriate for certain subgroups and typologies of offenders. • There are still uncertainties regarding the most useful elements and components of a SOT program for convicted adult male sex offenders. • There is no clear answer on whether the setting of the SOT program affects its impact on recidivism rates. Conclusions While the evidence from seven moderate-to-high quality SRs suggests that SOT has the potential to reduce sexual and nonsexual recidivism, the reported findings provide stronger support for the effectiveness of CBT approaches and for programs adhering to the RNR model. Any conclusions drawn from this overview of SRs remain tentative. Given the methodological problems of the available primary research, it is difficult to draw strong conclusions about the effectiveness of SOT programs using various CBT approaches for such a heterogeneous population. The reviewed evidence does not provide clear answers to what are the components of an optimal SOT program and to whether where the program is delivered matters. All SRs concluded that more and better research is needed to clearly answer these questions.

Methodology SRs that by virtue of design and quality of reporting were most likely to provide high levels of evidence, were selected to formulate the evidence base for this overview. All SRs were identified through a systematic search of the relevant scientific literature published in English between January 1998 and June 2010. The searched databases included: MEDLINE, EMBASE, The Cochrane Library, The Campbell Collaboration Library, CRD Databases (HTA, DARE), PsycINFO, Violence and Abuse Abstracts,

iv Treatment of Convicted Adult Male Sex Offenders Criminal Justice Abstracts, Sociological Abstracts, SocINDEX with Full Text, Social Work Abstracts, Social Services Abstracts, Gender Studies Database, National Criminal Justice Reference Service Abstracts, and Web of Science. Also searched were the University of Alberta library catalogue, clinical trials websites, websites of pertinent agencies and departments of corrections, health technology assessment (HTA) agency websites. In addition, the Internet was searched using the Google search engine. The methodological quality of the SRs included in this overview was appraised independently by two reviewers using the AMSTAR quality assessment tool. One reviewer extracted the data from the selected SRs and a second reviewer verified the data extraction. The evidence was qualitatively synthesized and is presented in summary tables.

Treatment of Convicted Adult Male Sex Offenders v ABBREVIATIONS All abbreviations that have been used in this report are listed below unless the abbreviation is well known, has been used only once, or is a nonstandard abbreviation used only in figures, tables, or appendices in which case the abbreviation is defined in the figure legend or at the end of the table. AHS-MH — Alberta Health Services-Mental Health AMSTAR – Assessment of Multiple Systematic Reviews CI – confidence interval CSC – Correctional Service of Canada CBT – cognitive behavioural therapy CODC – Collaborative Outcome Data Committee CPA – ciproterone acetate DSM-IV – Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition ES – effect size HTA – health technology assessment ICD-10 - International Classification of Diseases, 10th Edition LHRH – luteinizing hormone-releasing hormone MeSH – medical subject headings of MEDLINE MPA – medroxyprogesterone acetate OR – odds ratio RCT – randomized controlled trial RP – relapse prevention SR – systematic review SOT – sex offender treatment SOTEP – Sex Offender Treatment and Evaluation Project SOTP – sex offender treatment program SSRI – selective serotonin reuptake inhibitor

vi Treatment of Convicted Adult Male Sex Offenders GLOSSARY OF TERMS The glossary terms listed below were obtained and adapted from the following sources: ATSA Professional Issues Committee, Association for the Treatment of Sexual Abusers. Practice Standards and Guidelines for Members of the Association for the Treatment of Sexual Abusers, 2003. Center for Sex Offender Management. Glossary of terms used in the management and treatment of sexual offenders. Maryland: Center for Sex Offender Management, 1999 (cited 2010 Jan 19). Available from: http://www.csom.org/pubs/glossary.pdf Institute of Health Economics. Glossary. Edmonton, AB: IHE (cited 2010 Jan 19). Available from: http://www.ihe.ca/documents/Glossary.pdf. Actuarial risk assessment: An objective algorithmic method of estimating the risk for reoffending based on empirically identified risk factors. Actuarial risk assessment methods are generally more accurate than methods based on clinical judgment alone. Admission criteria: The specific characteristics and level of risk which can be treated and managed safely and effectively in a treatment program. Androgen: A steroid hormone producing masculine sex characteristics and having an influence on body and bone growth and on the sex drive. Anti-androgen: A substance that blocks the production of male hormones. Anti-androgens reduce endogenous levels of testosterone and thereby can reduce sex drive. Among the most commonly prescribed anti-androgens are medroxyprogesterone acetate and leuprolide acetate. Assessment: The process of collecting and analyzing information about an offender so that appropriate decisions can be made regarding sentencing, supervision, and treatment. An assessment does not and cannot determine guilt or innocence, and it cannot be used to determine whether an individual fits the “profile” of an offender who will commit future offences. Assessments lay the groundwork for conducting an evaluation. Chemical castration: The use of medications to inhibit the production of hormones in the sex glands. Child pornography: Any audio, visual, or written material that depicts children engaging in sexual activities or behaviours, or images that emphasize genitalia and suggest sexual interest or availability. Civil commitment: The confinement and treatment of sex offenders who are especially likely to reoffend in sexually violent ways following the completion of their prison sentence. Commitment is court ordered and indeterminate.

Treatment of Convicted Adult Male Sex Offenders vii Clinical polygraph: A diagnostic instrument and procedure designed to assist in the treatment and supervision of sex offenders by detecting deception or verifying the truth of statements of persons under supervision or treatment. Polygraph examinations involve the posing of different kinds of questions while physiological measurements are taken. Clinical significance: A conclusion that an intervention has an effect that is of practical meaning to patients and health care providers. Even if an intervention has a statistically significant effect, the effect might not be clinically significant. For example, in a trial with a large number of patients, a small difference between treatment and control groups may be statistically significant but clinically unimportant. Conversely, in a trial with few patients, it may be possible to observe an important clinical difference that does not achieve statistical significance. In this case, a larger trial may be needed to confirm the statistical significance of the difference between treatment and control groups. Cognition: The mental processes such as thinking, visualizing, and memory functions that are created over time based on experience, value development and education. Conviction: The judgment of a court based on the verdict of guilty, the verdict of a judicial officer, or the guilty plea of the defendant, that the defendant is guilty of the offence. Criminogenic needs: Factors that can change over time and that are related to risk for reoffending. Criminogenic needs are therefore important targets for treatment and community supervision of sexual abusers. Examples of criminogenic needs include association with antisocial peers, deviant sexual fantasizing, and substance use. Criminogenic needs are also referred to as dynamic risk factors. Denial: Applied to the sex offences field, denial is commonly defined as denial of fact (the offender may act shocked or indignant over the allegations of sexual abuse). Denial is usually interpreted as the failure of sexual abusers to accept responsibility for their offences. Increasing acceptance of responsibility is often considered to be an important part of sex offender treatment. Deviant arousal (deviant sexual interests): A pattern of being sexually aroused to deviant sexual themes. Not all sex offenders have deviant arousal patterns. Deviant sexual arousal is the most obvious manifestation of deviant sexual interests (also “deviant sexual preferences” or paraphilic interests), which are defined as sexual interests that (a) are rarely observed among individuals who have not engaged in criminal sexual behaviour and (b) would infringe on the rights of others if acted upon.

viii Treatment of Convicted Adult Male Sex Offenders DSM-IV/ICD-10: DSM-IV is an abbreviation for the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition and ICD-10 is an abbreviation for the International Classification of Diseases, 10th Edition. These are compendiums of diagnoses and their definitions that are utilized universally in psychiatry and related professions. Dynamic risk factors: Risk factors that can change over time and that are therefore important targets for treatment and community supervision. Examples of dynamic risk factors include ongoing associations with antisocial peers, deviant sexual fantasizing, and substance use. These risk factors can be distinguished from static risk factors, which are historic or very unlikely to change (e.g., number of prior offences, diagnosis of psychopathy). They are also referred to as criminogenic needs. Effectiveness, effective: The benefit (to health outcomes) of using a technology for a particular problem under general or routine conditions (that is, by a physician in a community hospital or by a patient at home). Empathy: A capacity for participating in the feelings and ideas of another. Evaluation: The systematic application of criteria and the forming of judgments based on an examination of psychological, behavioural, and social information about an individual. In the context of sexual abuse, sex offender assessments precede sex offender evaluations. The purpose of an evaluation is to formulate an opinion regarding a sex offender’s amenability to treatment, risk, dangerousness, and other factors in order to facilitate case management. Evidence-based medicine: The use of current best evidence from scientific and medical research to make decisions about the care of individual patients. It involves formulating questions relevant to the care of particular patients, searching the scientific and medical literature, identifying and evaluating relevant research results, and applying the findings to patients. Exclusion criteria: Specific offender characteristics and level of risk that determine whether an individual can or cannot be treated and managed safely and effectively in a treatment program. Follow-up: Observation over a period of time of an individual, group, or initially defined population whose relevant characteristics have been assessed in order to observe changes in health status or health-related variables. Health technology assessment (HTA): The systematic evaluation of the properties, effects, and other impacts of health care technology. Its primary purpose is to provide objective information to support health care decisions and policy-making at the local, regional, national, and international level. Incest: Sexual relations between close relatives, such as father and daughter, mother and son, or sister and brother. This also includes other relatives, stepchildren, and children of common-law marriages.

Treatment of Convicted Adult Male Sex Offenders ix Incidence: The number of new cases of illness commencing, persons falling ill, or adverse events during a specified time period in a given population. Level of risk: The degree of dangerousness a sex offender is believed to pose to potential victims or the community at large. Meta-analysis: The use of statistical methods to combine results from different studies to obtain a quantitative estimate of the overall effect of a particular intervention or variable on a defined outcome. The combined results may produce a stronger conclusion than can be provided by any individual study. Also known as data synthesis or quantitative overview. Outcome data: Data that demonstrate clear, relevant, and undisputed information regarding the effect of supervision and/or treatment on sex offender recidivism rates. Outcome evaluation: Assessment conducted after discharge from a program, typically by tracking all sex offenders to determine rates of recidivism. Paraphilia/paraphilic interests: Paraphilia is a psychosexual disorder. Paraphilic interests are defined in the DSM-IV as “recurrent, intense, sexually arousing fantasies, sexual urges, or behaviours” pertaining to a particular class of objects or activities. Some paraphilic interests are illegal if acted upon, such as pedophilia (a paraphilic interest in prepubescent children), hebephilia (sexual interest in, or arousal to, adolescents), exhibitionism (exposing one’s genitalia to others for purposes of sexual arousal), and voyeurism (observing unsuspecting individuals, usually strangers, who are naked, in the act of dressing or undressing, or engaging in sexual activities). Parole: A method of prisoner release on the basis of individual response and progress within the correction institution, providing the necessary controls and guidance while serving the remainder of their sentences within the free community. Pedophile: An individual with a paraphilic interest in prepubescent children. Phallometry (phallometric assessment or penile plethysmography): A device used to measure penile erection in male response to presentations of both appropriate and inappropriate sexual stimuli (which can be audio, visual, or a combination of both). The measured changes reflect physiological sexual arousal and provide information about deviant sexual interests. Phases of assessment: There are several phases and types of sex offender assessments, including investigative assessment, risk assessment, treatment planning assessment, clinical assessment, formal and informal assessments of progress in treatment, graduation or discharge readiness assessment, classification assessment, and outcome evaluation.

x Treatment of Convicted Adult Male Sex Offenders Positive treatment outcome: A treatment outcome that includes a significantly lower risk of the sex offender engaging in sexually abusive behaviour as a result of attaining a higher level of internal control. Positive treatment outcomes include a lack of recidivism; a dramatic decrease in behaviours, thoughts, and attitudes associated with sexual offending; and other observable changes that indicate a significantly lower risk of reoffending. Prevalence: The proportion of persons with a particular disease, risk factor, or adverse event within a given population at a given time. Prosocial: This term describes attitudes, beliefs, values, and behaviours that are associated with compliance with important societal norms and criminal laws. In contrast, antisocial individuals have little regard for the feelings of others, focus on their own needs, and are willing to engage in behaviours that cause harm to others. Psychometric testing: A series of questions, problems, or practical tasks that provide a measurement of aspects of an individual’s personality, knowledge, ability, or experience. These include measures of intelligence, academic achievement, personality traits, attitudes and beliefs, and mood. Measures may not be specific to sexual abusers but may nonetheless provide useful information for the assessment and management of sexual abusers. Psychopathy: An extreme manifestation of antisocial personality traits, including callousness, glibness, need for stimulation, impulsivity, irresponsibility, and grandiosity. Individuals with psychopathic traits are very likely to have serious criminal histories, likely to engage in violence, and likely to reoffend. Psychopathy overlaps with antisocial personality disorder but the two clinical diagnoses are not synonymous. Psychopathy Checklist-Revised: The clinical instrument to assess the degree to which an individual has characteristics of psychopathy. It is a 20-item instrument that is scored by the evaluator based on collateral information and typically an interview of the offender. Psychophysiological testing: Measures of physiological responses, including phallometry, polygraphy, and viewing time measures. Psychophysiological assessments can be useful in the assessment and treatment of sexual abusers because their results can be used to corroborate self-reported measures of behaviour. Quality assessment (QA) of systematic reviews (SRs): The extent to which the design and conduct of a study are likely to have prevented systematic errors (bias). Variation in quality can explain variation in the results of studies included in a systematic review. Usually the process of assessing the quality of an SR involves independent reviewers (at least two) who apply QA criteria to the SR and use consensus methods to resolve any disagreement.

Treatment of Convicted Adult Male Sex Offenders xi Rape: The crime of engaging in sexual acts, especially involving vaginal, oral, or anal penetration, with a person who has not consented. Recidivism: The commission of a crime after the individual has been criminally adjudicated for a previous crime (reoffence). Recidivism refers to multiple occurrence of any of the following key events in the overall criminal justice process: commission of a crime whether or not followed by arrest, charge, conviction, sentencing, or incarceration. A distinction can be made between different forms of recidivism, ranging from any recidivism (a new offence of any kind) to specifically sexual recidivism (a new sexual offence). Recidivism rates among sexual offenders will vary according to the sample of sexual offenders being followed, the length of follow-up, and the outcome measure being used (e.g., self-reported offences versus new charges versus new convictions). Relapse: A re-occurring sexually abusive behaviour or sex offence. Relapse prevention: A multidimensional model that incorporates cognitive and behavioural techniques to treat sexually abusive or aggressive behaviours. Adapted from the addictions literature, relapse prevention techniques are used to assist clients in developing strategies to cope with the precursors (thoughts, feelings, behaviours, and events) that have typically preceded their sexual offences, as identified in the client’s offence chain. For example, clients may learn to cope with dysphoria by talking with their spouse or engaging in prosocial activities rather than engaging in deviant sexual fantasies. Risk factors: A set of internal stimuli or external circumstances that threaten a sex offender’s self-control and thus increases the risk of lapse or relapse, characteristics that have been found to be associated with an increased likelihood of recidivism for known sex offenders. Risk factors are typically identified through risk assessment instruments. An example of a sex offender risk factor is a history of molesting boys. Risk level: The determination by evaluation of a sex offender’s likelihood of recommitting an offence; if the offender reoffends, the extent to which the offence is likely to be traumatic to potential victims. Based on these determinations, offenders are assigned a risk level consistent with their relative threat to others. Sex offenders who exhibit fewer offences, less violence, and none or few collateral issues (e.g., substance abuse, cognitive deficits, learning disabilities, neurological deficits, or use of weapons) are considered low-risk offenders compared to those whose profile reflects more offences and greater violence in their actions. Risk level is changeable, depending on the behaviours exhibited within a treatment program. Disclosures of additional, previously unknown offences or behaviours may also alter the assessment of the offender’s level of risk. Risk management: A term used to describe services provided by corrections personnel, treatment providers, community members, and others to manage xii Treatment of Convicted Adult Male Sex Offenders the risk presented by sex offenders. Risk management approaches include supervision and surveillance of sex offenders in a community setting (risk control) and require sex offenders to participate in rehabilitative activities (risk reduction). Risk reduction: Activities designed to address the risk factors contributing to the sex offender’s sexually deviant behaviours. These activities are rehabilitative in nature and provide the sex offender with the necessary knowledge, skills, and attitudes to reduce the likelihood of a reoffence. Safety, safe: A judgment of the acceptability of risk (a measure of the probability of an adverse outcome and its severity) associated with using a technology in a given situation (for a patient with a particular health problem, by a clinician with certain training, or in a specified treatment setting). Selective serotonin reuptake inhibitors (SSRIs): A class of antidepressant drugs that increase active levels of serotonin in the brain by blocking serotonin reabsorption by neurons. SSRIs are antidepressants that are associated with diminished sexual arousal and diminished sexual motivation by affecting hypothalamic (inhibitory) serotonin receptors. Sexual dysfunction side effects (diminished arousal, inhibited or delayed orgasm) may be associated with spinal nerve serotonin pathways and are not directly related to hypothalamic serotonergic effects. Sex offender: The term most commonly used to define an individual who has been charged and convicted of illegal sexual behaviour. Sexual abusers: The term most commonly used to describe individuals who have engaged in sexual behaviour that is considered to be illegal. The term refers to individuals who may have been charged with a sex crime but have not been convicted. In some cases, the target person is unable to give legal consent (e.g., because of their young age or diminished mental capacity) or is unaware that the behaviour is taking place (e.g., voyeuristic activity). In other cases, the target person does not consent and surprise, threat, or force is used by the sexual abuser to engage in the sexual behaviour (e.g., rape, sexual assaults of children, exhibitionism). Sexual assault: Forced or manipulated unwanted sexual contact between two or more persons. Sexual deviancy: Sexual thoughts or behaviours that are considered abnormal, atypical, or unusual.

Treatment of Convicted Adult Male Sex Offenders xiii Social support network: A group of individuals who communicate with each other and can help sexual abusers refrain from reoffending while they are living in the community. Members of the social support network (support persons) can help sexual abusers to cope with risky situations and can help monitor their compliance with treatment or community supervision requirements. Support persons can include family members, church officials, employers, probation or parole officers, and treatment professionals. Static risk factors: Risk factors that are historical in nature or that are very unlikely to change over time, such as a number of prior offences or a history of exhibitionism. Static risk factors can be contrasted with dynamic risk factors, which may change over time and are important targets for treatment and community supervision. Statistical significance: A conclusion that an intervention has a true effect, based upon observed differences in outcomes between the treatment and control groups that are sufficiently large for it to be unlikely that the differences occurred by chance (as determined by a statistical test). Statistical significance indicates the probability that the observed difference was due to chance if the null hypothesis is true; it does not provide information about the magnitude of the treatment effect. Statistical significance is necessary but not sufficient for determining clinical significance.) Study design: An analytic approach to conduct an epidemiological investigation. There are different types of study design, but for particular research questions certain study designs result more appropriate. For example, randomized controlled trials are considered to be essential for addressing questions regarding therapeutic efficacy and effectiveness. Case-control studies and cohort studies are better for deciding questions relating to etiologic or risk factors.

xiv Treatment of Convicted Adult Male Sex Offenders TAble of contents Executive Summary ...... iii Abbreviations/Glossary of terms ...... vi Introduction ...... 1 Objective and Scope ...... 1 Background ...... 2 Sexual offending and sex offenders ...... 3 Etiology, diagnostic issues, and epidemiology ...... 4 Diagnostic issues ...... 5 Epidemiology of sexual offending ...... 5 Management of sex offenders ...... 6 Sex Offender Treatment for Convicted Adult Males ...... 6 Psychotherapy ...... 7 Pharmacotherapy ...... 8 Canadian Context ...... 10 Statistics on sexual offending and sex offenders ...... 10 Management of sex offenders ...... 11 Results ...... 11 Available evidence ...... 11 Description of the selected systematic reviews . . . . .12 Methodological quality of the selected systematic reviews ...... 17 Results reported by the selected systematic reviews . . . . . 18 Treatment characteristics ...... 25 Offender characteristics ...... 26 Study design quality characteristics ...... 27 Ongoing research ...... 31 Discussion ...... 32 Limitations ...... 38 Conclusions ...... 39 Appendix A: Methods ...... 40

Treatment of Convicted Adult Male Sex Offenders xv Appendix B: Excluded Studies ...... 54 Appendix C: Selected Systematic Reviews ...... 59 References ...... 77 Figures and Tables Figure 1: Study selection process ...... 51 Table 1: Summary of selected systematic reviews ...... 13 Table 2: Systematic reviews that conducted moderator analyses ...... 34 Table A1: Search strategy ...... 40 Table B1: Excluded research studies ...... 54 Table B2: Multiple publications ...... 58 Table C1: Systematic reviews ...... 61 Table C2: Methodological quality appraisal using AMSTAR tool ...... 72

xvi Treatment of Convicted Adult Male Sex Offenders INTRODUCTION Worldwide, sexual offending has gained great importance among the public as a result of high-profile cases portrayed in the media. Public concern has been matched by clinical, political, and legislative actions.1-8 A variety of sex offender treatment (SOT) interventions and programs of different levels of complexity and sophistication have been developed and used for the management of sex offenders. These therapeutic interventions are the source of a great deal of study and debate about how to achieve an effective reduction of recidivism in this population. A health technology assessment (HTA) study has been recently undertaken in response to a request from Alberta Health Services–Mental Health (AHS-MH) to determine from the published literature the efficacy/ effectiveness of psychotherapy and/or pharmacotherapy interventions delivered within SOT programs for convicted adult male sex offenders. To scope the HTA, the AHS-MH request has been carefully reviewed and framed with input from an Expert Advisory Group to identify and examine the best available scientific evidence in this field using a scientifically rigorous approach.

OBJECTIVE AND SCOPE The specific aim of the HTA (which was prospectively designed to assist policy-makers in Alberta in sorting through the evidence-based choices) has been to respond to the following predefined questions: 1. What are the most effective psychotherapy or pharmacotherapy interventions provided within a SOT program, either alone or in combination, for convicted adult male sex offenders to reduce (the risk of) sexual/non-sexual recidivism measured at 2 years or more after completion of SOT? 2. Which of the available SOT programs currently used for treating convicted adult male sex offenders are the most effective in terms of reduced (risk of) sexual/non-sexual recidivism measured after 2 years or more from completion of SOT? 3. Does the setting of the SOT program affect its impact on the outcomes of interest? To answer these questions, this HTA study used a systematic and structured approach to overview the systematic reviews published in English between January 1998 and June 2010 that reported on: Population – convicted adult male sex offenders (18 years and older; exhibitionists, internet-related sex offenders, voyeurs, child molesters, and/or rapists) without neurodevelopmental disorders;

Treatment of Convicted Adult Male Sex Offenders 1 Interventions – psychotherapy and/or pharmacotherapy (alone or combined) provided in inpatient and/or outpatient settings; Comparators – no therapy, placebo, usual care, and/or other therapy; Outcome – violent or nonviolent recidivism, sexual or nonsexual recidivism. The date restriction was applied to ensure that the evidence collected was current and clinically relevant since many changes in sex offender treatment and care occurred during the mid 1990s. This overview of systematic reviews does not cover the use of pharmacotherapy and/or psychotherapy interventions provided within a SOT program for other types of sex offenders, such as juvenile sex offenders, female sex offenders, or sex offenders with neurodevelopmental disorders. More details on the methodological approach used for the present study are provided in Appendices A and B. Appendix A describes the literature search strategy and the study selection process and also summarizes the methodological approach for data extraction and data synthesis and analysis used for the overview of the selected systematic reviews. Appendix B lists the excluded research studies and the main reasons for their exclusion.

BACKGROUND Sexual offending is both a social problem and a public health issue. It has become a major challenge for social policy because of the high human and financial costs to victims and the social and health services, and the high public investment in policing, prosecuting, and incarcerating sex offenders.1,2,4,6,7,9-16 There is an expectation that the correctional systems should make reasonable efforts to reduce the chance that identified sexual offenders will reoffend. Sexual offences are generally defined as any violation against established legal or moral codes with respect to sexual behaviours.1-3,5-7,14-18 They can vary from non-contact offences such as exhibitionism, voyeurism, and Internet-related (online) sex offences to contact offences such as rape and child molestation. Issues relating to gender, age, relationship, aggression, definition of consent, and location influence whether a particular sexual behaviour is considered to be legal or illegal. Thus, what constitutes an unacceptable or illegal sexual behaviour varies between societies and within the same society over time. As a result, each country and local jurisdiction manages sexual offence crimes in its own way.1-3,5-9,14-16 Different legal jurisdictions have different sexual offence laws as well as policies about enforcement, arrest, conviction, and plea bargaining. These practices can vary widely from one jurisdiction to another, especially as they relate to prosecution of some types of sexual offences such as sex between younger and older adolescents. One common approach to managing sex offenders in developed countries is to provide specialized

2 Treatment of Convicted Adult Male Sex Offenders treatment programs. A number of different treatment programs have been established, but there has been controversy concerning how well they work.

Sexual offending and sex offenders Research on the nature of sexual offending and on the characteristics of sex offenders has concentrated on understanding child molestation and rape and is based primarily on reports from victimization survey studies and studies of incarcerated sex offenders, which provide complementary information.2-8,14,19-28 Although there are methodological limitations associated with each of these types of research studies, to date they have reached a number of consistent conclusions. —— The overwhelming majority of sex offenders are men. —— Most sexual offences against children and adults are committed by men connected to their victims as family members, close friends, or acquaintances. —— The majority of sexual offences (80 to 90%) are never reported to the police. —— Most sex offenders are not mentally ill. However, many sex offenders have abnormal personality traits or personality disorders, some may have a diagnosis of paraphilia, and others may have learning disabilities or biological factors that contribute to their offending. —— Although a history of sexual abuse during childhood may increase the risk of developing sexually abusive behaviour as an adult, most victims of sexual abuse never become sex offenders, and most sex offenders have never been sexually abused. About one-third of all sex offenders reported physical or sexual abuse during childhood, and rates of childhood victimization differ depending on the type of sex offender (for example, child molesters are twice as likely to report having experienced childhood sexual abuse than rapists). —— When compared with other adult male offenders, sex offenders are in the older age range, with child molesters being significantly older than rapists. Recidivism rates decline with age among adult male sex offenders. —— Sex offenders tend to have versatile criminal careers, with sexual offending being embedded in more general offending behaviours. There is research evidence suggesting that sexual offending may be a lifelong problem for many sex offenders.2,4,7,29,30 Although the term “sex offender” might suggest that individuals who offend sexually are all alike, they are in fact a heterogeneous population in terms of demographic and social makeup. There is no “typical” sex offender or profile of a sex offender.2-7,14,17,20,25,31,32 Even though sex offenders are commonly grouped in categories based on victim or offence characteristics

Treatment of Convicted Adult Male Sex Offenders 3 (i.e., child offenders or rapists), they exhibit different patterns and precursors of offending and differ in their level of impulsiveness, persistence, risks posed to the public, and desire to change their behaviour. New developments and capabilities provided by the Internet have changed the ways in which sex offending can be committed. Recently, interconnections between sexual abuse through prostitution and trafficking and the Internet started to be reported worldwide.15-18,33-36 The Internet provides new tools to assist in the sale of children and adolescents, creates space to facilitate the commission of sex-related crimes, and provides access to vulnerable people to victimize. Internet may also be attractive to exhibitionists, who can use web cameras for offending sexually. Various computer-related communication tools (such as e-mail, newsgroups, instant messaging, and short messaging service) can operate as vehicles that facilitate Internet-related sex offences. The research regarding the nature and scope of Internet-initiated sexual offending is very limited.15,16,18,33-36 Available research has focused upon those who produce, collect, and distribute child pornography, and currently there is an ongoing debate on whether consumers of child pornography pose a risk for contact sexual offences.15-17,36,37 Little is known about those who groom children and adolescents online and the boundary between online and contact sexual offending. There is evidence to suggest that Internet-initiated sexual offences involving adults and youth more often fit a model of statutory rape - adult offenders who meet, develop relationships with, and openly seduce underage adolescents - than a model of forcible sexual assault or pedophilic child molesting.15 Knowledge about those who use “extreme pornographic sexual images” depicting adults is very limited, and it is yet to be considered how this type of offender may be categorized and assessed.

Etiology, diagnostic issues, and epidemiology Many etiological theories aim to explain why some individuals offend sexually.2-7,14,23,37 However, none of these theories applies to all (or even a majority) of sex offenders. Sexual offending is a very complex, multifaceted problem, and its causation appears to be multi-determined. There is no single cause for sexual offending, which is not likely to be fully explained by any one single trait, motivation, or characteristic, but, rather, by a combination of interacting risk factors that may increase an individual’s tendency to offend sexually. Researchers have examined multiple factors and traits of large samples of sex offenders and found several characteristics that seem to be common.2,4,6,14,19,20,38,39 These characteristics include sexual deviance (deviant sexual arousal, interests, or preferences); cognitive distortions or pro–offending attitudes; previous criminal history; substance abuse; history of childhood victimization; social or interpersonal problems with issues such as ineffective communication skills, social isolation, general social skills deficits, or problems in intimate relationships; and poor coping or self management skills. Although some of

4 Treatment of Convicted Adult Male Sex Offenders these characteristics can also be found in samples of other criminals, or within the general population, it is believed that they may somehow be related to sexual offending, particularly when these factors interact with other variables and circumstances. Some of these characteristics (such as sexual deviance, pro-offending attitudes, and intimacy deficits) may also be related to recidivism among known sex offenders.

Diagnostic issues Diagnostic issues are thought to be relevant to the conceptualization of the problems presented by sex offenders.2-4,6,7,14,15,17,23,37,40-43 Some clinicians and researchers apply a diagnosis of paraphilia to their sexual offending clients, which typically derives from the criteria outlined in the International Classification of Diseases (ICD-10) or the Diagnostic and Statistical Manual (DSM-IV). However, despite the existence of paraphilia as a diagnosis in both the ICD-10 and DSM-IV, many clinicians do not regard sexual deviance as a psychiatric entity, and there are concerns that the unclear boundary between sexual deviance and paraphilic disorders makes it difficult to make reliable diagnoses of sexual psychopathology.5,14,23,37,41,44 The knowledge about the association between psychopathology and sexual offending is still limited, and these diagnostic criteria might not provide a sound basis for a universally agreed upon classification that might guide treatment. Psychiatric diagnoses other than paraphilias are complicating factors in trying to understand the genesis of sexual offending, making it difficult to determine the extent to which mental disorder is associated with sexual offending as opposed to offending generally.

Epidemiology of sexual offending There is a lack of clear data regarding the prevalence and incidence of sexual offending in the community, making it difficult to understand the real human and financial costs that are associated with this problem.2,4-8,14,20,45 Statistical and other information about sexual offending are limited to what is known to the authorities as they are based on reported cases and on the activities of convicted sex offenders. Sex offenders who are in the criminal justice system at any time (either arrested, incarcerated, or under supervision in the community) represent only a small fraction of those who have committed sexual offences. Establishing reliable estimates of recidivism rates is also difficult because many offences are undetected or not reported to police, and recidivism rates vary across cohorts and jurisdictions.1,2,4,6-8,14,20,38 Important issues include the way recidivism is defined and measured, length of follow-up, and the large gap between official statistics and real counts. The international research literature shows that base rates of sexual recidivism are low compared to other types of offence, averaging between 10% and 24% over a 5- to 6-year follow-up period.2,4,6-9,20,39,46,47 Various types of sex offenders recidivate at different rates, and the risk for reoffending varies, depending on the types of victims they target, prior conviction for sexual offences, and other variables.

Treatment of Convicted Adult Male Sex Offenders 5 When arrested or convicted again, sex offenders tend to have committed nonsexual offences rather than new sexual offences.2,4-8,20,47

Management of sex offenders Management of sex offenders in institutions (prison or in-patient forensic psychiatry settings) or in the community (on probation or parole) makes use, in varying degrees, of various approaches, including specialized treatment, supervision strategies, substance abuse screening, polygraphy, tracking with global positioning systems, and community notification.1,2,4-6,14,45,48-50 Although all of these strategies can and should be evaluated for their effectiveness alone and in combination, doing so is beyond the scope of this study. This study addresses only SOT delivered within programs available for convicted adult male sex offenders.

SEX OFFENDER TREATMENT FOR CONVICTED ADULT MALES SOT is one component of the sex offender management process and is still a developing field.1,2,4,6,7,22,45,50,51 It involves the provision of specialized treatment designed to promote offenders’ accountability and enhance their skills and competencies. SOT has been delivered within programs located in a variety of settings to manage sex offenders at various stages in the criminal justice system and has been combined with other management strategies. Although the purpose of and the manner in which treatment is implemented and delivered can vary from one setting and jurisdiction to another, the main goal of SOT is to protect the community by reducing the likelihood of reoffending. In addition to the treatment goals, there are some other specific characteristics of SOT programs that include work with involuntary clients, a victim and community focus, a limited confidentiality agreement, being guided by treatment providers, and need for collaboration among professions. The effects of SOT programs would be expected to vary according to many factors related to the offenders’ characteristics (i.e., age, offence type, treatment behaviour, presence of personality disorders), the program’s characteristics (i.e., setting, selection criteria, type of treatment, treatment format and intensity, timing and duration of treatment, treatment providers’ professional background, training, and attitudes), and treatment integrity (whether treatment providers adhere to the planned program of treatment).2,4,6,7,11-14,22,32,45,50-52 An important consideration is given to the setting of the program, timing, and duration of treatment, and treatment integrity. There is still uncertainty regarding the most appropriate setting in which to offer SOT programs, the best time to provide SOT, and the optimal length of the program, which may be driven by the specific program’s objectives and the sex offenders’ needs. The timing of SOT is often related to the availability of treatment services. Practical issues of treatment

6 Treatment of Convicted Adult Male Sex Offenders expense, refusal, dropout rates, sources of referral, program reputation, and ease of access are factors that may influence the outcomes of interest. Because sex offenders represent a heterogeneous population, some of them will respond well to various SOT interventions and others will not. Prior to treatment, sex offenders usually undergo a comprehensive evaluation to identify the static and dynamic factors that influence their behaviour, with the overall goal to guide treatment.2,4,5,7,21,22,29,32,38,43,51 A formal risk assessment conducted by qualified professionals offers, with moderate accuracy, estimates of the risk posed by a sex offender to the community. The development and refinement of risk assessment tools is one of the most active areas of sex offender research.14,38,43 The debate about what is the proper model for conducting risk assessment began many decades ago, and it still continues.14,38,43 Treatment strategies for sex offenders have been available for many decades, and over time there have been many changes that corresponded to a concomitant shift in the nature and scope of the programs themselves.2,4,6,7,10,29,45,50,51,53-55 Treatment of sex offenders has moved from the early desire to cure “sexual disorders” to the general goal of rehabilitation through training in order to manage and control their behaviour so that they do not commit another offence. Psychotherapy has emerged as the principal approach, and there has been an increasing acceptance of pharmacotherapy as an adjunct to psychotherapy.

Psychotherapy Psychotherapy for sex offenders includes psychological and behavioural treatments that can be delivered in group sessions, individual sessions, or both.5,6,10,14,21,29,32,42,44,50,54,56-59 The goal of these treatments is to change sex offenders’ belief systems, eliminate inappropriate behaviour, and increase appropriate behaviour by modifying reinforcement contingencies so that offensive behaviour is no longer reinforced. However, psychotherapy interventions in SOT might have harmful effects, unintentionally increasing recidivism and thereby harming victims, sex offenders, their respective families, and society.10,58,60 For example, recounting offence details in acceptance of responsibility and relapse prevention (RP) exercises might expose other offenders to new sexual content and new methods for accessing victims.50,60 The approach to psychotherapy for sex offenders has changed over time.2,5- 7,10,14,32,50,51,55 The focus has shifted from treatment aimed at offenders’ gaining insight into why they commit sexual offending (early in the 1980s) to a more structured cognitive-behavioural approach (late 1980s and early 1990s) which focuses on a range of static and dynamic risk factors for future offending. Late in the 1980s, the cognitive-behavioural model adopted RP techniques from drug and alcohol research, and applied them through the development of individually tailored SOT programs. Since then, cognitive-behavioural therapy

Treatment of Convicted Adult Male Sex Offenders 7 (CBT) with RP has emerged as the principal strategy for the psychotherapeutic treatment of sex offenders and has been used in various combinations with skills-based interventions, empathy training, and other approaches. CBT for sex offenders addresses criminogenic needs (i.e., cognitive distortions and denial) and is focused on changing both the offenders’ problematic way of thinking and their actions.5,10,14,29,50,53,55,58,59,61-64 This approach has shifted over time toward a more flexible, individually tailored approach. It became less confrontational and started taking into consideration the relationship between therapist characteristics, group climate variables, and other process-related and contextual factors. Although CBT continues to emphasize an RP approach, it has evolved to include the acquisition of self-regulation skills (such as managing both positive and negative emotions, known as the “self-regulation model”, or SRM) and positive aspects of the human experience, usually referred to as the “good lives model”, or GLM. Overall, the emphasis on process-related variables, positive treatment goals, and engaging strategies reflects a more positive psychotherapeutic approach to SOT. Some of the currently available SOT programs using CBT employ the risk/need/responsivity (RNR) principles, which are the principles of effective correctional services for general offenders, to inform decisions about assessment, treatment, and supervision of their clients.13,50,65-67 According to these principles, treatments are most likely to be effective when they treat offenders who are likely to reoffend (moderate or higher risk), target characteristics that are related to reoffending (criminogenic needs), and match treatment to the offenders’ learning styles and cultures (responsivity). Currently, a group treatment approach is the preferred format for delivering psychotherapy interventions, as it can be offered to a greater number of offenders, it capitalizes on peer feedback and support, and it is cost-effective.2-4,6,29,32,45,50,51 However, just as the group members can have a positive influence on each other, they can also influence each other in antisocial ways. Based on the offender’s specific needs, individual counselling sessions may be an adjunct to group treatment. There is still debate about the level of intensity for delivering these interventions.

Pharmacotherapy Pharmacotherapy for sexual offenders is based on the assumptions that the behaviour is sexually motivated and that the suppression of sexual drive will reduce sexually deviant behaviour.2-6,14,32,45,48-50 The goal of pharmacotherapy is to preserve normophilic sexual interests and behaviours while reducing deviant or paraphilic behaviours and ultimately reduce recidivism. This would result in deviant sexual fantasy, urges, and behaviour being suppressed while non-deviant fantasy, urges, and behaviour would be preserved. Pharmacological options for sex offenders are classified under three categories of drugs: (1) anti-androgens and hormonal agents that reduce the level and

8 Treatment of Convicted Adult Male Sex Offenders activity of the male hormone testosterone, such as medroxyprogesterone acetate (MPA) (available in the United States) and cyproterone acetate (CPA) (available in Canada); (2) luteinizing hormone–releasing hormone (LHRH) agonists that can lower testosterone to castration levels; and (3) a class of antidepressants called selective serotonin reuptake inhibitors (SSRIs) that can cause sexual dysfunction in terms of reduced libido, delayed orgasm, and reduction in the frequency and intensity of sexual fantasy, urges, and arousal.2,3,5,29,32,40,48-50,57,68-70 Hormonal medication is indicated primarily for cases in which sexual arousal plays a central role in offending.3,5,6,11,21,32,40,48-50,57,68,70,71 However, the potential for significant side effects limits the utility of these agents, which are usually restricted to sexual offenders with moderate- to high-risk for committing contact sexual offences. The side effect profile for SSRIs is less potentially serious than that of anti-androgen and hormonal agents. The anti-androgen and hormonal treatments can be safely prescribed for only short periods of time and can interfere with conventional sexual drive, whereas the SSRIs are relatively safe for prolonged use, and conventional sexual drive is preserved in most offenders. However, optimal duration of treatment with any of hormonal and non-hormonal agents is not known due to lack of long-term clinical evaluation. The prescriptive use of any of these agents is recommended to be combined with psychotherapy. An important issue in pharmacotherapy is the offender’s compliance as these treatments are effective only while the drug is being taken.2,3,5,6,14,32,44,48-51,68,72 A comparison of treatment compliance and acceptance of serotonergic and anti-androgen agents showed that SSRIs are much more frequently selected by sex offenders.32 Pharmacotherapy for sex offenders with hormonal and non-hormonal agents may be successful in reducing recidivism rates through the reduction of sexual arousal and behaviour.2,3,5,6,14,29,32,40,48,49,57 However, knowledge about the neurobiological basis of sexual arousal and sexual behaviour is limited, much being based on animal studies that are of uncertain relevance to humans. There is some support for the efficacy of CPA, MPA, LHRH analogues, and SSRIs in reducing the frequency and intensity of sexual drive in paraphilias. The research supporting these results, however, is methodologically weak and consists mostly of open trials with small samples and short follow-up periods or case studies. Randomized controlled trials (RCTs) involving sex offenders are lacking. Despite the lack of well-conducted clinical trials of the efficacy of SSRIs for the treatment of sexual offenders,3,5,32,49 such medications have been reported to be the most commonly used agents in North America.49,50 Programs in Canada and the United States use hormonal agents much less than SSRIs, and the use of the MPA agents appears to be declining in the United States.50

Treatment of Convicted Adult Male Sex Offenders 9 CANADIAN CONTEXT

Statistics on sexual offending and sex offenders Similar to other countries, quantifying sexual offences in Canada is a challenge as only a small fraction of sexual assaults are reported to police (less than 10%), and the numbers of sexual assaults reported to police are markedly lower compared to numbers from victimization surveys.31,73-80 According to self- reported victim data from the 2004 General Social Survey (GSS), there were 1977 incidents of sexual assault per 100,000 population aged 15 years and older, a rate not statistically different from that of the 1999 GSS (2058 per 100,000 population). Police-reported data indicate that an estimated 24,200 sexual offences were brought to police attention in 2007.76 At a rate of 73 per 100,000 population, police-reported sexual offences were down 3% over the previous year but still accounted for 8% of all police-reported violent crime in 2007. According to data from Statistics Canada, rates for sexual assault and other sexual offences in 2008 remained almost the same as those in 2007.75 Both victimization and police-reported data indicate that the majority of sexual offences in Canada (more than 80%) are of a less serious nature with the least physical injury, such as unwanted sexual touching.73,74,76-80 Clearance rates by police and conviction rates in adult criminal courts are lower for sexual offences than for other types of violent crime.31,76 However, adults convicted for sexual offences generally tend to receive harsher sentences than those found guilty of other violent offences. On 31 December 2004 a review of Correctional Service of Canada’s offender management system81 revealed that: —— sex offenders under federal jurisdiction represented about 16% of the total federal offender population; —— sex offenders incarcerated in federal institutions represented about 20% of the total federal incarcerated population; —— sex offenders under community supervision represented 11.8% of the total federal conditional release population; —— the majority of the listed sex offenders were men (99.6%); —— the average age of sex offenders under federal jurisdiction was 44 years; —— the average age of sex offenders at admission was 40 years; and —— the majority of sex offenders (69.1%) were Caucasian and there was somewhat a larger proportion of Native sex offenders relative to their proportion of all federal offenders. Over a 10-year period (31 December 1994 to 31 December 2004) the total sex offender population decreased by 21.4%, the sex offender population in institutions decreased by 27.0%, and the sex offender population under community supervision decreased by 6%.81

10 Treatment of Convicted Adult Male Sex Offenders According to data reported by Statistics Canada and the Canadian Centre for Justice Statistics, rates of reported sexual offences in Alberta have fluctuated over time. In 1993 Alberta had the second-highest rate of sexual offences in Canada.80 Between 1996 and 1997 the rate of sexual offences declined and Alberta had the fourth-lowest rate in Canada.9 In 2003, Edmonton ranked 10th and Calgary ranked 19th out of 25 Canadian cities in the frequency of sexual offences.80 Between 2007 and 2008, rates for reported sexual assault and other sexual offences in Alberta increased slightly from 64.3 and 6.2 per 100,000 population, respectively, to 67.2 and 8.5 per 100,000 population, respectively.75 While various governmental agencies collect data on sexual offences and on child pornography offences in Canada (i.e., Statistics Canada, the Canadian Centre for Justice Statistics, Correctional Service of Canada), Internet-specific data are not currently available.33

Management of sex offenders The Correctional Service of Canada (CSC) has continually implemented more SOT programs since it began offering this type of treatment in 1973.9,82 The implementation of SOT programs by CSC has put Canada at the forefront of research and knowledge about SOT, with many of the Canadian SOT programs reporting promising results. One of these programs is the Phoenix Program, which operates at the Alberta Hospital Edmonton, a specialized psychiatric hospital that provides intensive care (inpatient forensic services) (www.albertahealthservices.ca).9,24,41,83-87 In the community, the Alberta Hospital Edmonton provides comprehensive services through the Forensic Assessment and Community Services (FACS).

Results The literature search conducted for both the HTA study (see Appendix A) identified 1690 citations. After screening of titles and abstracts, 1619 citations were excluded from the final selection process. The full text of 71 potentially relevant articles was retrieved and further evaluated for inclusion in the overview. The application of the selection criteria to all retrieved full-text articles resulted in 61 being excluded (main reasons for their exclusions are listed in Table B1, Appendix B). Figure 1 in Appendix A outlines the study retrieval and selection process for the HTA study. The following commentary summarizes the reviewed research evidence.

Available evidence Eight SRs10-13,58,88-90 that by virtue of design and quality of reporting were most likely to provide high levels of evidence were selected for the overview. Two other SRs91,92 were identified as multiple publications of two SRs included in the overview.11,13 Although the multiple publications91,92 were not included as unique studies, any relevant information that the authors provided was included when appropriate.

Treatment of Convicted Adult Male Sex Offenders 11 Details of all selected SRs and the results of the critical appraisal of the methodological quality of the reviews are provided in Appendix C (Tables C1 and C2).

Description of the selected systematic reviews Of all selected SRs, four were conducted by reviewers from North America12,13,88,90 and four were conducted by reviewers from Europe and Australia.10,11,58,89 Objectives and selection criteria varied across the nine SRs, and as a result, there was little overlap among their included primary research studies. Only papers reporting results from the long-term RCT conducted by Romero and Williams and papers reporting preliminary and final findings from the California’s Sex Offender Treatment and Evaluation Project (SOTEP) study were included in most of the SRs.10-13,58,88,90 The number of published and unpublished primary research studies included in the selected SRs ranged from nine10 to sixty-nine11 (see Table 1 below and Table C.1 in Appendix C), most of which were based on American and Canadian samples. All SRs10-13,58,88-90 included only controlled or comparative studies (reported between 1953 and 2009)) that evaluated the effect of various psychotherapy interventions compared to no treatment, drug treatment, standard care, “treatment as usual”, alternate treatment, or treatment refused. Few of the controlled/comparative studies used rigorous/strong research designs, and there were even fewer studies with strong designs examining interventions consistent with current or contemporary standards as most of the evaluated SOT programs were developed and implemented before the mid 1980s. None of the selected SRs provided relevant details on elements of interest for the treatment programs that were evaluated by their included individual primary research studies (e.g., treatment concept; duration; selection criteria; risk assessment procedures and tools; type, frequency, and duration of sessions; timing of treatment; and treatment providers). Most of the primary research included in the selected SRs involved a combination of individuals convicted for different types of sexual offences (see Table 1 below and Table C1 in Appendix C). Child molestation (including incest offence) was the most frequently addressed sexual offence, followed by rape, and only a few studies included adult exhibitionists and voyeurs. It appears that no primary research study referred exclusively to rapists or voyeurs. It also appears that no primary research study focused on or involved Internet-related sex offences. None of the selected SRs reported separate results on the effectiveness of any SOT interventions for different sex offender typologies. The majority of the reviews did not provide relevant details on the characteristics of the offenders involved in their included primary research studies. Recidivism was measured by reconviction, reoffending, reincarceration, or rearrest in most studies using various definitions (e.g., return to prison,

12 Treatment of Convicted Adult Male Sex Offenders readmissions to institutions, parole violations, unofficial community reports, or all of these) and sources (e.g., criminal justice records, state/provincial records, child protection records, self-reports, or all of these). None of the selected SRs reported on adverse events for any of the interventions they evaluated. Table 1: Summary of selected systematic reviews

Included primary research studies Systematic review* and authors’ conclusions*

88 Aos et al., 2006 Included primary research: 18 studies reported between 1983 and 2005 Objective: to assess interventions provided within adult Study design: two RCTs, 16 quasi-experimental and corrections programs non-experimental evaluation studies Standards of scientific rigour: Participants: adult sex offenders outcome evaluation studies rated Intervention: CBT, psychotherapy/counseling, behavioural at least 3 on a 5-point rigour scale therapy, mixed treatments provided within adult SOTPs located (based on Maryland scale) in prison or in the community QA scores: moderate quality on Comparator: treatment as usual AMSTAR tool (4/11) Outcome: recidivism Follow-up: NR Authors’ conclusion: “We found that cognitive-behavioural treatments are, on average, effective at reducing recidivism, but other types of sex offender treatment fail to demonstrate significant effects on further criminal behaviour.” Cognitive- behavioural programs for sex offenders on probation “demonstrated the largest effect observed in our analysis”.

89 Bilby et al., 2006 Included primary research: 21 studies reported between 1991 and 2004 Objective: to examine quasi-experimental research on Study design: 21 quasi-experimental studies, most with psychological interventions for matched controls adult sex offenders and adults Participants: adults (≥18 yr); males and females; rapists, incest showing abusive sexual behaviours offenders, pedophiles, child molesters, child abusers, child Standards of scientific rigour: offenders, exhibitionists, voyeurs control trials, matched Intervention: cognitive behavioural, cognitive self-change, and or non-matched; covert sensitization programs located in institutions (most often QA scores: low quality on prisons) or in the community AMSTAR tool (2/11) Comparator: standard care, standard punishment and prison, no treatment, drug treatment Outcome: reconviction, recidivism, rearrest, reoffending Follow-up: NR Authors’ conclusion: “As with the review of the experimental literature, there is little consensus on the capabilities of interventions from varying theoretical backgrounds in altering offenders’ behaviour or attitudes.”

Treatment of Convicted Adult Male Sex Offenders 13 10 Brooks-Gordon et al., 2006 Included primary research: nine studies reported between 1983 and 2005 Objective: to examine experimental research on Study design: 9 RCTs (category B); only 1 long-term RCT psychological interventions for (conducted in the community) adult sex offenders and individuals Participants: convicted adult male sex offenders: pedophiles, showing abusive sexual behaviours exhibitionists, and sexual assaulters (1 long-term RCT) Standards of scientific rigour: Intervention: group psychotherapy plus probation RCTs described as category A or (one long-term RCT) B using Cochrane Collaboration Handbook guidelines (2002) Comparator(s): probation (standard care) (one long-term RCT) QA scores: moderate quality on Outcome: rearrest for sex offence (one long-term RCT) AMSTAR tool (7/11) Follow-up: up to 10 yr Authors’ conclusion: Group psychotherapy provided in the community “increased re-arrest at 10 years.” “The re-arrest rate was not statistically significantly increased in the therapy group… compared to the no group therapy control. If there were only a few more arrests in the intervention group, it could be suggested that the therapy was less effective than doing nothing to prevent re-arrest.”

12 Hanson et al., 2002 Included primary research: 68 studies reporting results from 43 treatment programs between 1977 and 2000 Objective: to examine effectiveness of psychological Study design: 4 random-assignment studies, 17 incidental treatment for sex offenders assignment studies, 10 studies comparing completers vs. dropouts, six studies comparing any treatment attendance vs. Standards of scientific rigour: treatment refusers, six studies assigning treatment based on need controlled or comparative studies using same recidivism criteria Participants: sex offenders; adults and juveniles; for all subjects; research design males and females classification based on how Intervention: “current” treatments (behavioural, other offenders were assigned to groups psychotherapeutic, mixed delivered between 1998 and 2000 (using CODC coding manual) or CBT delivered after 1980) and “non-current” treatments QA scores: moderate quality on (behavioural, other psychotherapeutic, mixed delivered AMSTAR tool (5/11) before 1980) Comparator: no treatment, alternate/alternative treatment Outcome: sexual/general reconviction, rearrest Follow-up: 1 to 16 yr (median of 46 months) Authors’ conclusion: “We believe that the balance of available evidence suggests that current treatments reduce recidivism, but that firm conclusions await more and better research.” For adults, “the treatments that appeared effective were recent programs providing some form of cognitive-behavioural treatment”. “Further research is needed in order to make reliable distinctions between types of treatment and types of offenders”.

14 Treatment of Convicted Adult Male Sex Offenders 13,92 Hanson et al., 2009 Included primary research: 23 studies reported between 1980 and 2009 Objective: to examine if principles of effective interventions for general Study design: four randomized trials, one study using offenders also apply to SOT and “researcher assigned nonrandomized” design, six retrospective to assess SOT effectiveness using cohort studies, six studies using “need, volunteer, and dropout” only studies with minimum design, six studies using “other concurrent comparison study quality group” design Standards of scientific rigour: Participants: sex offenders; adolescents and adults; controlled/comparative studies males and females rated “weak”, “good”, and “strong” Intervention: psychological interventions (institution- and/or with CODC guidelines community-based) QA scores: high quality on Comparator: no treatment AMSTAR tool (10/11) Outcome: reconviction, rearrest Follow-up: 1 to 21 yr (median of 4.7 yr) Authors’ conclusion: “Given the consistency of the current findings with the general offender rehabilitation literature, we believe that the RNR principles should be a major consideration in the design and implementation of treatment programs for sexual offenders.” “Cognitive-behavioural treatments are the norm…, and in the current review many of the programs examined also made special efforts to engage sexual offenders in the treatment. Further research is needed concerning how best to apply the risk principle to sexual offenders.”

90 Polizzi et al., 1999 Included primary research: 13 studies reported between 1988 and 1996 Objective: to evaluate prison- and non-prison-based SOTPs Study design: 13 impact evaluation studies (four rated as level ‘4’, two rated as level 3, and seven rated as level 2) Standards of scientific rigour: studies on impact of SOT rated at Participants: child molesters, high-risk sex offenders, adult least 2 on a 5-point rigour scale rapists, exhibitionists (Maryland scale) Intervention: CBT and RP (prison- and non-prison-based), QA scores: moderate quality on intensive residential CTP (prison-based), SSOSA (prison-based) AMSTAR tool (5/11) Comparator: no treatment Outcome: rearrest for sexual, felony, or violent offences; reconviction for sexual, violent of both offences; reincarceration Follow-up: up to 31 yr for prison-based SOT; up to 11 yr for community-based SOT Authors’ conclusion: “…non-prison-based sex offender treatment programs using cognitive behavioural treatment methods are effective in reducing the sexual offense recidivism of sex offenders.” “Prison-based treatment programs were judged to be promising, but the evidence is not strong enough to support a conclusion that such programs are effective. Too few studies focused on particular types of sex offenders to permit any type of conclusions about the effectiveness of programs for different sex offender typologies.”

Treatment of Convicted Adult Male Sex Offenders 15 11,91 Schmucker and Losel, 2008 Included primary research: 69 studies reported between 1953 and 2004 Objective: to systematically review controlled outcome evaluations of Study design: 6 reports of uncompromised random design psychosocial and biological SOT (level ‘5’), 6 reports of studies using “matching, statistical control” design (level ’4’), 17 reports of studies with “equivalence Standards of scientific rigour: assumed” (level ‘3’), 37 reports studies for which group SOT evaluation studies rated at equivalence could not be assumed (level ‘2’) least 2 on a 5-point rigour scale (adapted from Maryland scale) Participants: rapists, child molesters, incest offenders, exhibitionists; adults and adolescents QA scores: high quality on AMSTAR tool (8/11) Intervention: CBT, classical behavioural, insight oriented, therapeutic community, other psychosocial (unclear), hormonal medication, surgical castration Comparator: no treatment available, treatment refused, other Outcome: recidivism defined by arrest, conviction, charge, lapse behaviour Follow-up: 1 to 10 yr (median of 5.22 yr) Authors’ conclusion: “...our results indicate that sexual offender treatment can significantly reduce recidivism rates. The size of the effect is small to moderate but it is in accord with what we know from the larger research literature on general offender treatment evaluation. However, the evidence is based on studies that mostly apply a weak methodological standard.” “Hormonal medication, cognitive-behavioural, and behavioural approaches also revealed a positive effect”. “Non-behavioural treatments did not show a significant impact.” “Overall, findings are promising but more differentiated evaluations of high quality are needed.”

58 White et al., 1998 Included primary research: three studies reported between 1983 and 1994 Objective: to evaluate anti-libidinal techniques to assist people with Study design: three RCTs (category B); only one long-term RCT disorders of sexual preference and Participants: “mixed” convicted sex offenders those convicted of sexual offences (one long-term RCT) Standards of scientific rigour: Intervention: group psychotherapy plus probation RCTs described as category A or (one long-term RCT) B using Cochrane Collaboration Handbook guidelines (1997) Comparator(s): probation (standard care) (one long-term RCT) QA scores: high quality on Outcome: rearrest for sex offence (one long-term RCT) AMSTAR tool (10/11) Follow-up: up to 10 yr Authors’ conclusion: “Considering the widespread use of group therapy, the findings of the largest and longest study in this review must be considered disturbing. That it reports no effects on recidivism over a long period of time may suggest that nondescript group therapy may have to give way to a more focused treatment such as response prevention.”

*Only information and authors’ conclusions regarding the population, intervention(s) and outcome of interest are summarized AMSTAR – Assessment of Multiple Systematic Reviews; CBT – cognitive behavioural therapy; CODC – Collaborative Outcome Data Committee; CTP – Correctional Treatment Program; NR – not reported; QA - quality assessment; RCT – randomized controlled trial; RNR principles –Risk/Need/Responsivity principles; RP – relapse prevention; SOT – sex offender treatment; SOTP – sex offender treatment program; SSOSA – Special Sex Offender Sentencing Alternative; SR – systematic review; yr – year(s)

16 Treatment of Convicted Adult Male Sex Offenders Methodological quality of the selected systematic reviews Methodological quality varied across the selected SRs (see Table 1 above and Table C.2 in Appendix C). The AMSTAR tool rated one SR89 as low quality, four SRs as moderate quality,10,12,88,90 and three SRs11,13,58 as high quality. All selected SRs used multiple electronic databases in their literature searches and other sources such as handsearching, scanning reference lists, and consulting experts to locate the most relevant primary research. In two SRs10,89 the literature search and study selection were not limited by publication language. One SR11 searched for studies reported in five languages (English, French, German, Dutch, and Swedish), one12 conducted a search for studies published in English and French, and in another SR13 researchers reviewed studies reported in English, French, and German. It appears that the study selection in three SRs58,88,90 was restricted to those reported in English. In only three SRs10,11,58 was the likelihood of publication bias assessed. Seven SRs10-13,58,88,89 provided detailed description of their inclusion criteria in terms of participants, interventions, outcomes, and study design. The standards for scientific rigor used for the selection and analysis of primary research varied across the nine SRs (see Table 1 above and Table C1 in Appendix). To account for the differences in the study design of their included primary research, three SRs used the Maryland Scale of Scientific Rigor or a 5-point rigour scale adapted from this scale.11,88,90 Two SRs10,58 used the guidelines from The Cochrane Collaboration and one SR13 made decisions regarding study quality based on the guidelines of the Collaborative Outcome Data Committee (CODC). Consequently, these SRs often disagreed in terms of which studies were identified to be rigorous or good enough to be included in their systematic review. Four of the selected SRs10,12,13,58 reported that data extraction from the included primary research studies was performed by two reviewers working independently, whereas three 10,13,58 reported that two reviewers had independently assessed the methodological quality of the included studies. Four of the selected SRs11,12,88,90 categorized their included primary research studies according to a level of evidence hierarchy, but it is not clear from their reporting whether they also performed a formal critical appraisal to determine how studies within each category compared in terms of their methodological quality. Four SRs11-13,88 analyzed and synthesized the evidence both qualitatively and quantitatively (meta-analyses). Because of the large degree of heterogeneity in terms of study design, participants, interventions, and outcome measures in the reviewed primary research, they conducted statistical analyses of study heterogeneity for their pooled data and moderator analyses.

Treatment of Convicted Adult Male Sex Offenders 17 The conclusions of all reviewed SRs appeared to follow from their results, and it was appropriate that all of the SRs highlighted the need for further research and more rigorous evaluations of SOT interventions. Three SRs13,58,90 provided information on whether their authors had a conflict of interest and only two SRs58,90 mentioned their sources of funding. None of the SRs reported the sources of funding of their included primary research studies. Seven SRs,10-13,58,88,90 which were rated as moderate- to high-quality on the AMSTAR tool, qualified for data analysis and synthesis, and their findings are summarized in the following commentary. Although the remaining review89 did not qualify for data analysis and synthesis, its details were collected and summarized in Table C1 (Appendix C).

Results reported in the selected systematic reviews The review by White et al.58 appears to be the first meta-analysis of RCTs that presented findings relevant to the management of sex offenders. In 1998 the reviewers reported the results from a Cochrane SR on anti-libidinal treatment for adults who have been convicted of sexual offences or who have disorders of sexual preference. They included three RCTs: one on anti-libidinal medication (MPA), one on an RP program, and one on group psychotherapy. The same three studies were also reviewed by Brooks-Gordon et al.,10 who, in 2006 reported the results from the most comprehensive search for experimental research in the area of sexual offending. They qualitatively reviewed data from nine RCTs (all reported before 1998) as part of an SR of the literature on psychological interventions for adult sexual offenders and adult individuals showing abusive sexual behaviours. Of all experimental research reviewed by White et al.58 and by Brooks-Gordon et al.,10 the RCT on group psychotherapy (conducted by Romero and Williams and published by Romero in 1983) is the only RCT reporting recidivism (measured as rearrest for a sex offence) data over a follow-up period of 2 years or more for all offenders. This RCT reports a 10-year follow-up of a mix of 231 convicted adult sex offenders (144 sexual assaulters, 39 exhibitionists, and 48 pedophiles; 170 of these sex offenders had previous arrests for sex offences) randomly assigned to group psychotherapy plus probation (1 hour of group therapy per week for 40 weeks, plus one probation home visit per month) or standard care (one report to probation per month plus one home visit per month). Results reported by this RCT did not show group psychotherapy to be beneficial and even suggested the potential for harm at 10-year follow-up.10,58 Group psychotherapy plus probation increased the rearrest rate by 10 years when compared to standard care. However, the study found no significant difference in rearrest rate for those allocated to group psychotherapy plus probation (14%) and those receiving standard care (7%) (odds ratio [OR] = 1.87, 95% confidence interval [CI]: 0.8 to 4.37).10,58

18 Treatment of Convicted Adult Male Sex Offenders According to Brooks-Gordon et al.10 “If there were only a few more arrests in the intervention group, it could be suggested that the therapy was less effective than doing nothing to prevent re-arrest”. White et al.58 stated that “no effects on recidivism over a long period of time may suggest that nondescript group therapy may have to give way to a more focused treatment such as response prevention”. The SRs by White et al.58 and Brooks-Gordon et al.10 highlighted that RCTs can be employed as an evaluative tool in this difficult area of research. However, their findings do not provide clear guidance regarding how to improve current practice. In 1999 Polizzi et al.90 published the results of an SR of sex offender prison- and non-prison-based treatment programs. The reviewers included 13 impact evaluation studies, all of them completed within the past 10 years before 1999, although the treatment might have been developed and implemented earlier. The included 13 studies (eight studies on prison-based treatment and five studies on non-prison-based treatment) varied with regard to the type of sex offender population (e.g., child molesters, high-risk offenders, adult rapists, and exhibitionists) and the outcome measures used (e.g., sexual recidivism or nonsexual recidivism). Of the eight studies examining prison-based treatment, only two were rated as sufficiently rigorous (level 4 on the 5-point methodological rigour scale), and from these data the reviewers drew their conclusions.90 Overall for the prison-based programs, only one level 4 study (published in 1995), using a CBT approach for 296 high-risk sex offenders, found statistically significant differences between the treated offenders and the controls (untreated 283 matched sex offenders) for sexual reconviction rates (14.5% and 33.2%, respectively; mean follow-up of 6 years). The effect size (ES) for reconvictions for sexual offence was moderate (0.45). Findings for nonsexual reconviction rates were shown to be not statistically significant (ES = 0.06). The other level 4 study evaluating prison-based treatment (published in 1993) examined recidivism rates for 197 child molesters released from maximum-security prisons between 1958 and 1974.90 The follow-up period for both treated and untreated child molesters extended up to 31 years (not clear on the type of treatment and how many offenders were treated and untreated). Recidivism was determined as a re-conviction for a sexual offence, violent offence, or both. The study found that offenders in the treatment program (location of the program not clearly stated) had fewer reconvictions (44%) than offenders who were incarcerated prior to the inception of the treatment program (48%) (ES = 0.8) but not compared to offenders who were sentenced to the same institution at the same time as the treatment group but did not participate in treatment (33%) (ES = -0.23). These differences are not statistically significant.

Treatment of Convicted Adult Male Sex Offenders 19 Four of the non-prison-based treatment studies were judged to be of scientific merit (rated at a level 3 or above), and these results were used by the reviewers in their analyses.90 Two of these studies (published in 1988 and 1991) found significantly lower recidivism rates for the treated group compared to the untreated group (ES were moderate to large: 0.51 and 0.70, respectively). In both studies, the treatment program was based on CBT approaches used for 126 child molesters (setting not mentioned) who were followed for up to 11 years in the 1988 study (rated as level 4) and for 61 exhibitionists (in a Sexual Behaviour Clinic) who were followed for 4 years in the 1991 study (rated as level 3). A third study (rated as level 4) found that 98 child molesters and adult rapists who received cognitive-behavioural and RP treatment and completed the program had lower recidivism in comparison to untreated offenders in the control groups (97 in the volunteer group and 96 in the non- volunteer group). However, this difference was not statistically significant (offenders in all groups were followed up for approximately 5 years). This study was published in 1994 and reported preliminary findings from the longitudinal RCT conducted at a state hospital in California, known as the SOTEP study. The fourth study evaluating non-prison-based treatment (rated as level 3) included 23 exhibitionists who participated in a program at a Sexual Behaviour Clinic and 21 exhibitionists who did not participate in treatment.90 The program attempted to modify the deviant sexual preferences of 23 treated sex offenders (no other details). Both the treated and untreated sex offenders were followed for approximately 9 years. Exhibitionists participating in the program were reconvicted or charged with a sexual offence at a lower rate than untreated exhibitionists (39.1% versus 57.1%, respectively) (ES = 0.36). However, the differences were not statistically significant. From their analyses, Polizzi et al.90 concluded that non-prison-based SOT programs using CBT approaches are effective in reducing sexual recidivism of sex offenders. Prison-based treatment programs were judged to be promising, but the reviewers cautioned that the evidence was “not strong enough to support a conclusion that such programs are effective”. The findings reported by Polizzi et al.90 provide evidence on the overall effectiveness of CBT approaches delivered within older SOT programs as all of the included primary research studies were reported before 1998. Since the programs were not described sufficiently and relevant offenders’ characteristics were not provided, the findings do not offer clear guidance regarding which SOT interventions work best, in what setting, for which type of sex offenders, and under what circumstances. According to the reviewers, “there were too few studies with specific types of sex offenders to give us enough information to draw conclusions about the effectiveness of programs that targeted specific offender typologies.” In 2002 Hanson et al.12 published the first report from a Collaborative Data Research Project established by the Association for the Treatment of Sexual

20 Treatment of Convicted Adult Male Sex Offenders Abusers (ATSA). The reviewers conducted a meta-analysis that combined data on 43 treatment programs using psychological interventions (23 offered in institutions, 17 in the community, and three in both settings) involving more than 9000 treated and untreated or differently treated adult male sex offenders. The evaluated interventions were delivered between 1965 and 1999, mostly in specialized SOT programs, with approximately 80% of the offenders receiving “current” treatment (defined as CBT offered after 1980 or behavioural, other psychotherapeutic, and/or mixed treatments delivered between 1998 and 2000). Recidivism was defined by reconviction in eight studies, rearrest in 11, and 20 studies used broad definitions, including parole violations, readmissions to institutions, unofficial community reports, or all of these. Thirteen programs reported only on sexual recidivism, five reported only on general recidivism, and 25 reported on both. Averaged across all types of treatments and research designs, the findings indicate an overall positive effect of treatment versus no treatment in terms of sexual recidivism rates (12% versus 17%, respectively) and general (any) recidivism rates (28% versus 39%, respectively).12 The findings were based on a median follow-up period of 46 months for both treatment and comparison groups. Most of the evidence for treatment effectiveness came from incidental assignment studies which, on average, were associated with significant reductions in sexual recidivism (OR = 0.62, 95% CI 0.50 to 0.77, 17 studies, N = 2948) and, in general, recidivism (OR = 0.52, 95% CI 0.40 to --0.68, 10 studies, N = 1176) with statistically significant heterogeneity across the studies. Overall, the only four random-assignment studies were associated with non-significant odd ratios for both sexual (OR = 1.03, 95% CI 0.67 to 1.59, three studies, N = 694) and general recidivism (OR = 0.92, 95% CI 0.69 to 1.22, four studies, N = 897), with statistically significant heterogeneity across the studies. When data from the random- and incidental-assignment studies were combined, “current” treatments were associated with significant reductions in both sexual (from 17.3% to 9.9%) and general recidivism (from 51% to 32%) for adult and adolescent sex offenders whereas the “older” treatments appeared to have little effect.12 According to the reviewers, “these reductions were not large, but they were statistically reliable and large enough to be of practical significance”.12 However, they advised on interpreting the findings with caution, given the small number of studies and the significant variability across these studies (see Table C1, Appendix C). For both sexual and general recidivism, studies comparing a “current” treatment to an alternate treatment tended to find larger effects (OR = 0.28, 95% CI 0.15 to 0.54, six studies, N = 708) than did studies that compared a “current” treatment to an untreated comparison group (OR = 0.64, 95% CI 0.51 to 0.81, 10 studies, N = 2753), with non-statistically significant heterogeneity across the studies. According to Hanson et al.,12 “studies comparing treatment completers to dropouts consistently found higher sexual and general recidivism rates for

Treatment of Convicted Adult Male Sex Offenders 21 the dropouts, regardless of the type of treatment provided. Even in studies where there was no difference between the treatment group and the untreated comparison groups, the treatment dropouts did worse”. The reviewers also reported that “offenders who refused treatment were not at higher risk for sexual recidivism than offenders who started treatment. Treatment refusers, however, were at relatively high risk for general recidivism”.12 Offenders assigned to treatment based on perceived need had significantly higher sexual recidivism rates than the offenders considered not to need treatment and showed similar general recidivism rates for both the treated and untreated groups. “Current” treatments appeared to be effective for adult sex offenders in terms of reduced sexual recidivism (OR = 0.61; 95% CI 0.48 to 0.76; 12 studies, N = 2779).12 Both treatments delivered within institutional programs (six studies, N = 1771) and those offered in community programs (six studies, N = 1008) were associated with significant reductions in sexual recidivism (OR = 0.62, 95% CI 0.48 to 0.80; and OR = 0.57, 95% CI 0.34 to 0.95, respectively). “Current” treatments also significantly reduced general recidivism for adults (OR of 0.59; 95% CI 0.45 to 0.78; five studies, N = 1101). However, treatments delivered in the community (two studies, N = 330) appeared to have a stronger effect on general recidivism (OR = 0.21, 95% CI 0.12 to 0.37) than treatment provided in institutions (OR = 0.82; 95% CI 0.60 to 1.13; three studies, N = 771). Four studies (one random-assignment and three incidental-assignment studies) of sex offender specific treatment for adults found a significant reduction in general recidivism (OR = 0.61, 95% CI 0.45 to 0.82, Q = 32.79; df = 3; P <:001). The results reported by Hanson et al.12 suggest that treatments that appeared effective for adult sex offenders were “current” programs providing some form of CBT. They also suggest no effect depending on setting, as both institution- based and community-based programs for adults were associated with reductions in sexual and general recidivism of adult sex offenders. However, these findings provide evidence about the overall effectiveness of CBT delivered within older SOT programs as only 23% of the reviewed studies were reported after 1999. They do not identify specific program models that were found to be superior. Neither do the findings provide guidance on the effectiveness of any SOT interventions for different sex offender typologies. Since the treatment programs were not sufficiently documented in the included studies, the treatment quality could not be rated with any confidence. Aos et al.88 conducted an SR on correction programs for adult offenders, including SOT programs, to determine what works, if anything, to lower the criminal recidivism rates. Of all 291 primary research studies that met their standards for scientific rigor, 18 were outcome evaluation studies of various adult SOT programs (located in prison or in the community). Eleven SOT programs provided specialized cognitive-behavioural treatment (CBT) and reported significant reductions in recidivism.

22 Treatment of Convicted Adult Male Sex Offenders —— CBT in prison: Data pooled from five studies indicate that specialized CBT programs provided within a prison setting (which may also include behavioural reconditioning to discourage deviant arousal and modules addressing RP) for adult sex offenders (no details on subjects’ characteristics) achieve, on average, a statistically significant 14.9% reduction in recidivism rates compared with “treatment as usual” (no description provided). One of the five studies was the final report on the SOTEP study, a longitudinal RCT with an 8-year follow-up period, showing small but not statistically significant effects on recidivism. —— CBT in the community for low-risk offenders on probation: Data pooled from six studies indicate that specialized CBT programs provided in the community for low-risk adult sex offenders on probation (no details on subjects’ characteristics) achieve, on average, a statistically significant 31.2 % reduction in recidivism rates compared to “treatment as usual” (no description provided). These community-based CBT programs were described as being similar to the specialized CBT programs provided in prison and may also include behavioural reconditioning to discourage deviant arousal and modules addressing RP. The other seven SOT programs provided psychotherapy (three studies), behavioural therapy (two studies), and mixed treatments for adult sex offenders (two studies).88 Based on analyses of data pooled from these seven studies, the reviewers found that the evaluated SOT programs did not demonstrate a statistically significant reduction in recidivism compared to “treatment-as-usual” (no description provided). The findings reported by Aos et al.88 do not provide guidance concerning which SOT interventions work best, in what setting, and for which sex offenders. These findings also provide evidence about the overall effectiveness of CBT delivered within older treatment programs (only one of the reviewed primary research studies reported after 1999 showed no statistically significant treatment effect for the evaluated program). Since most treatment programs were not sufficiently described and no relevant information was provided on the involved offenders’ characteristics, the results from this meta-analysis do not provide information relevant for current practice. Schmucker and Losel11,91 conducted the first meta-analysis of both published and unpublished sex offender psychological and biological treatment (hormonal treatment and surgical castration) outcome studies reported in five languages. They reviewed 69 studies with more than 22,000 subjects using 80 independent comparisons between treated and untreated sex offenders (see Table C1, Appendix C). About one-third of these studies were reported since 2000, but the actual program implementation started earlier (1990s). Although most studies were conducted in North America, eight studies from German-speaking countries, eight from Great Britain/United Kingdom, and

Treatment of Convicted Adult Male Sex Offenders 23 five from other countries were also located. Unpublished evaluations comprised 36% of the study pool. The majority of comparisons involved only adult sex offenders (see Table C1, Appendix C).11,91 Nearly one-half of the comparisons addressed CBT programs (46%) and 14 comparisons addressed physical treatment, six of which dealt with hormonal medication. Although most treatments were specifically designed for sex offenders, the reviewers found it difficult to rate whether treatment was implemented reliably, as three-quarters of the studies (49 comparisons) did not provide information on program integrity. Residential treatment was somewhat more frequent than outpatient, with approximately one-half of the comparisons taking place in an institutional setting. Although a group format was most frequently used, almost 50% of the programs included at least some individualized treatment. An explicit extension of treatment through specific after-care services was reported for 15 comparisons. In more than 50% of the primary research studies the authors were affiliated with the evaluated treatment. The definition of recidivism varied from arrest (24%), conviction (30%), and charges (19%), to lapse behaviour (4%).11,91 Recidivism was recorded after an average follow-up period of more than 5 years (see Table C1, Appendix C). Sexual recidivism outcomes were reported in 74 of the 80 comparisons. Data regarding (nonsexual) violent reoffending were provided for 20 comparisons, and in 49 studies the authors presented data on overall recidivism. Most programs combined offenders convicted for different types of sex offences.11,91 Most frequently, programs involved child molesters (including incest offenders), followed by rapists and exhibitionists. Nine programs addressed child molesters only, and four addressed exhibitionists only. No program referred exclusively to rapists. Other contact and non- contact offences were included, but the specific types were not defined. Treated offenders participated voluntarily in most studies; 30% of the comparisons referred to offenders who were at least partially obliged to attend treatment. Results of an analysis that integrated the individual ES according to the random-effect model showed that the absolute difference in sexual recidivism between treatment and control groups was 6.4 percentage points.11,91 According to Schmucker and Losel, this represented a 37% reduction from the base rate of the control group.11,91 For violent recidivism, the average recidivism rate for treated offenders was 5.2 percentage points lower than that for untreated offenders (a 44% reduction from the base rate of the control group). The corresponding rate of general recidivism for treated offenders was 11.1 percentage points lower than for untreated offenders (a 31% reduction from the base rate of the control group). However, except for violent recidivism, the ES distributions showed considerable heterogeneity (see Table C1, Appendix C). The reviewers conducted moderator analyses in order to isolate variables that might account for these differences.11,91 The moderator analyses were restricted

24 Treatment of Convicted Adult Male Sex Offenders to sexual recidivism as an outcome. The following commentary summarizes the findings from these analyses.

Treatment characteristics The various treatment approaches differed considerably in ES.11,91 In total, biological treatment had higher effects (OR = 7.37, 95% CI 4.14 to 13.11) than psychosocial interventions (OR = 1.32, 95% CI 1.07 to 1.62); Q (1, k = 66) = 30.47, P < 0.001. This was particularly due to the very large mean ES for surgical castration. Hormonal treatment also showed a higher effect (OR = 3.08, 95% CI 1.40 to 6.79) than two psychosocial interventions, which also had a significant impact on sexual recidivism (OR = 1.45, 95% CI 1.12 to 1.86, for CBT evaluated in 35 comparisons; and OR = 2.19, 95% CI 1.22 to 3.92, for classic behaviour therapy evaluated in seven comparisons). The ORs of the other psychosocial approaches (OR = 0.98, 95% CI 0.51 to 1.89, for insight-oriented treatment evaluated in five comparisons; and OR = 0.86, 95% CI 0.54 to 1.35, for therapeutic community treatment evaluated in eight comparisons) were close to one, indicating no significant difference in recidivism rates between treated and untreated offenders.11,91 According to the reviewers, although promising results were reported for other pharmacological treatments, there were no controlled evaluations to determine their usefulness in terms of reduced recidivism.11,91 In the further moderator analyses, the studies on surgical castration were excluded for various reasons.11,91 Although the setting variable revealed no significant difference, there was a strong tendency for relatively larger effects in outpatient treatment (OR = 1.93, 95% CI 1.35 to 2.77; evaluated in 27 comparisons) and smaller effects in institution-based treatments (OR = 1.16, 95% CI 0.84 to 1.60, not statistically significant for prison-based programs evaluated in 21 comparisons; OR = 1.10, 95% CI 0.62 to 1.94, not statistically significant for hospital-based programs evaluated in eight comparisons).11,91 Mixed settings (evaluated in 10 comparisons) had an intermediate ES (OR = 1.37, 95% CI 0.78 to 2.41, not statistically significant). While these differences did not reach significance, an ordering of the setting variable from institutional to outpatient treatment revealed a significant correlation of r= 0.27 (P = 0.02). However, the reviewers assumed this variable was somewhat confounded with the treatment approach. Schmucker and Losel11,91 found that whether the treatment program was specifically designed for sexual offenders or it was a program for general offenders that also incorporated sexual offenders made a difference in sexual recidivism rates. This difference was also found in outcomes of general recidivism (P = 0.05). Although the decade in which a program was implemented related significantly to ES, the reviewers found no linear relationship.11,91 More “modern” programs did not generally prove to be particularly successful

Treatment of Convicted Adult Male Sex Offenders 25 (OR = 1.27, 95% CI 0.86 to 1.87, not statistically significant for treatments implemented in the 1990s; OR = 1.38, 95% CI 1.08 to 1.77, for treatments implemented in the 1980s; OR = 2.03, 95% CI 1.34 to 3.09, for treatments implemented in the 1970s; OR = 0.56, 95% CI 0.32 to 0.98, for treatments implemented before 1970). The year of publication as another indicator of recently showed similar results (r = 0.08, P = 0.51). Whether the treatment was delivered in an individual or a group format did not result in significant outcome differences (OR = 1.12, 95% CI 0.76 to 1.66, not significant for only group treatment; OR = 1.57, 95% CI 1.02 to 2.42, for mainly group treatment; OR = 2.45, 95% CI 1.36 to 4.40, for mixed treatment; OR = 1.40, 95% CI 0.77 to 2.53, not statistically significant for mainly individual treatment; OR = 2.88, 95% CI 1.14 to 7.24, for only individual treatment).11,91 However, in this category, the reviewers assumed confounding with various content variables. Another finding related to the implementation of the programs was that evaluations in which the study author(s) were in some way involved in the program delivery showed clearly significant treatment effects, but programs that were evaluated by independent researchers did not11,91 (OR = 1.92, 95% CI 1.44 to 2.56, for 32 comparisons in which authors were affiliated with the evaluated programs; OR = 0.99, 95% CI 0.76 to 1.29, not statistically significant, for 30 comparisons in which the authors were not affiliated with the evaluated programs). A related finding was that model programs revealed better outcomes than programs implemented in everyday routine. However, the difference was not as clear-cut, and both kinds of implementation revealed significantly positive mean effects. There was no separate reporting for adult sex offenders in any of the moderators analyses summarized above.

Offender characteristics Schmucker and Losel11,91 found programs that specifically addressed juvenile sex offenders had a higher significant effect (OR = 2.35, 95% CI 1.01 to 5.43, for programs including adolescents only, evaluated in seven comparisons) than those for adult offenders (OR = 1.43, 95% CI 1.08 to 1.90, for programs including adults only evaluated in 36 comparisons). However, for the programs evaluated in these studies it was not clear their settings, the type of interventions used, type of therapeutic sessions, investigators’ involvement or authors’ affiliation, type of sex offences, sex offenders’ characteristics, how they defined recidivisms, and how data were collected. Although the impact on specific offender groups is important in treatment practice, according to the reviewers the description of offenders’ characteristics in the included primary research was often insufficient, and only some studies differentiated the results by type of sexual offence.11,91 These comparisons showed significant effects for all categories (OR = 4.91, 95% CI 1.64 to 14.68,

26 Treatment of Convicted Adult Male Sex Offenders for rapists; OR = 2.15, 95% CI 1.11 to 4.16, for extrafamilial child molesters; OR = 3.72, 95% CI 1.27 to 10.93, for exhibitionists) except that of intrafamilial child molestation (OR = 1.02, 95% CI 0.58 to 1.80, not statistically significant, for incest offenders). There was no separate reporting on adult sex offenders. The reviewers cautioned that the latter finding was mainly due to the low recidivism base rate of incest offenders.11,91 The relatively high effect for rapists was based on only five studies and should be regarded with caution. As an indicator of therapy motivation, programs with voluntarily participating offenders were compared with programs that involved a more or less coerced treatment. Only the voluntary treatment showed a significant mean treatment effect (OR = 1.45, 95% CI 1.08 to 1.93). However, the reviewers cautioned about the heterogeneity within any of the categories (each P < 0.02), and the between differences did not reach statistical significance.11,91 Whether treatment was terminated regularly or prematurely had an impact on sexual recidivism.11,91 Completers showed better effects than the control groups (OR = 1.58, 95% CI 1.23 to 2.05, for treatment completed regularly). However, dropouts did significantly worse (OR = 0.51, 95% CI 0.39 to 0.67). Dropping out of treatment doubled the odds of relapse (the odds to not recidivate for dropouts was only half the odds for offenders who were not treated at all), and this effect was homogeneous (Q = 11.52, df = 13, P = 0.57). In contrast, ES that referred to completers revealed considerable heterogeneity (Q = 100.20, df = 43, P < 0.001). There was no separate reporting on adult sex offenders.

Study design quality and characteristics Overall, study design quality, which indicates how much one can trust its results, did not yield a significant moderator effect.11,91 Comparisons of equivalent treatment and control groups (Maryland scale level 3 and above) fared somewhat better (OR = 1.69, 95% CI 1.26 to 2.28) than comparisons on level 2 (OR = 1.16; 95% CI 0.90 to 1.50). However, there was no linear relationship between design quality and the ES (r = 0.11, P = 0.36). RCTs did not differ from the other comparisons (OR = 1.48, 95% CI 0.74 to 2.96); Q (1, k = 66) = 0.07, P = 0.79. Whether the comparison group consisted of offenders who had refused treatment or not had a small but not statistically significant influence.11,91 Studies with control groups containing treatment refusers revealed relatively large treatment effects (OR = 1.96, 95% CI 1.20 to 3.20). However, according to the reviewers, these effects did not differ significantly from studies using other control groups.11,91 The length of follow-up time did not correlate with the ES (r = 0.00).11,91 Neither did different indicators of reoffending (i.e., reconviction, rearrest, etc.) relate systematically to outcome variation (Q (6, k = 60) = 3.45, P = 0.49). The sources used to gather the respective information had a significant impact on the ES (Q (2, k = 62) = 7.91, P = 0.02), with comparisons using self-reported

Treatment of Convicted Adult Male Sex Offenders 27 recidivism alongside official criminal records showing larger effects than studies that relied on official data only. However, according to the reviewers, this variable was confounded with the type of treatment because all studies on hormonal medication included self-reported recidivism. Sample size had a clear relation to the ES (r = -0.26, P = 0.03) that resulted mainly from large treatment effects in trials with very small samples (N ≤ 50) and could not be attributed to a publication bias only.11,91 Features of descriptive validity, which address the quality of study reporting, were also related to the ES.11,91 In particular, a lack of reporting details on the treatment concept and on outcome statistics correlated significantly with the ES (r = -0.33, P < 0.001; and r = -0.24, P = 0.03). Although no significant difference was found between the findings from published and unpublished studies, a significant mean treatment effect appeared in published studies only (N = 40, OR = 1.62, 95% CI 1.23 to 2.13).11,91 There were no significant ES differences between the various groups of countries in which the studies were performed (Q (4, k = 66) = 2.46, P = 0.65). Overall, Schmucker and Losel11,91 noted a positive outcome with SOT programs, with CBT approaches and hormonal treatment reported as the most promising interventions. However, their reported aggregated data on evaluations of heterogeneous modes of treatment did not allow a differentiated analysis and limited the extent to which the utility of one particular treatment and/or program could be isolated for examination. The reported results were difficult to interpret in terms of what specific CBT approaches and which hormonal agents were most effective, for what categories of sex offenders, and under what circumstances. In 2009 Hanson et al.13,92 reported the findings from a meta-analysis of 23 SOT outcome studies. The primary question was whether the principles of effective correctional interventions for general offenders (risk/need/ responsivity [RNR]) also apply to psychological treatment for sexual offenders. A secondary objective was to assess the effectiveness of psychological treatment for sexual offenders using only studies that met a minimum level of study quality established by the Collaborative Outcome Data Committee (CODC) guidelines93,94 and examine the extent to which the study results varied based on study design. The reviewers originally planned to assess both psychological and physical interventions, but none of the surgery or pharmacotherapy studies met the minimum standard of acceptability. Most of the included primary research studies were based on Canadian or American samples and were focused on adult male sex offenders (see Table C1, Appendix C).13,92 Of the 23 treatment programs, 10 were offered in institutions, 11 in the community, and two in both settings. Treatment was delivered between 1965 and 2004, with approximately 90% of the offenders receiving treatment after 1980 (most studies were reported between 1980 and 2009,

28 Treatment of Convicted Adult Male Sex Offenders 16 being reported after 1998). Fourteen of the 16 studies reported after 1998 examined specialized treatment programs for sex offenders. Recidivism was defined as reconviction in 10 studies and rearrest in 12 studies, and the most common source of recidivism information was national criminal justice records, followed by state or provincial records. The average follow-up periods ranged from 1 to 21 years (median of 4.7 years). The extent to which each of the 23 programs evaluated in the primary research adhered to the RNR principles was rated based on all available information, including program manuals, research articles, reports of accreditation panels, and, in some cases, site visits.13,92 Programs were rated as adhering to the “risk” principle if they provided intensive interventions to higher risk offenders and little or no treatment to low-risk offenders. Adherence to the “need” principle was met if the primary treatment targets were among those reported as being significantly related to sexual or general recidivism in previous meta-analyses. Programs were considered to meet the “responsivity” principle when they provided treatment in a manner and style matched to the learning style of the offenders. Hanson and colleagues found that the sexual, violent (including sexual), and general recidivism rates for the treated sexual offenders were lower than the rates observed for the comparison groups (based on unweighted averages, 10.9% versus 19.2% for sexual recidivism; 22.9% versus 32% for sexual or violent recidivism; and 31.8% versus 48.3% for general recidivism).13,92 However, reviewers mentioned that confidence in their findings was tempered by the observation that most studies used “weak” research designs. Of the included 23 studies, 18 were rated as weak and five were rated as good, and the effects tended to be stronger in the weak research designs compared to the good research designs. For 22 studies examining the sexual recidivism rate, results from both fixed-effect and random-effects analyses indicated significantly lower sexual recidivism rates in the treatment groups than in the comparison groups (see Table C1, Appendix C).13,92 For the reduction of sexual recidivism, treatment (no details provided) appeared to be effective for both adults (OR = 0.71, 95% CI 0.53 to 0.95, random-effect model; OR = 0.79, 95% CI 0.67 to 0.94, fixed-effect model), and adolescents (OR = 0.38, 95% CI 0.10 to 0.41, random-effect model; OR = 0.47, 95% CI 0.22 to 0.98, fixed-effect model), and did not depend on whether the program was delivered in the community or institution (both fixed-effect and random-effect comparisons showed no significant differences).13,92 Using a fixed-effect model, the treatment effects on sexual recidivism were smaller in the good-quality studies (OR = 0.94, 95% CI 0.74 to 1.20) than the weak studies (OR = 0.64, 95% CI 0.51 to 0.81). This comparison was not significant using random-effect comparisons. No significant differences were noted in the treatment effects for the published or unpublished studies.

Treatment of Convicted Adult Male Sex Offenders 29 For 10 studies that examined violent (including sexual) recidivism, according to both fixed- and random-effect analyses, the combined sexual and violent recidivism rates were not significantly lower for the treatment groups relative to the comparison groups (see Table C1, Appendix C).13,92 There were no differences in the effects according to whether the studies involved adults or adolescents, or were delivered in the community or in an institution.13,92 The effect of treatment was lower in the good-quality studies (OR = 1.11, 95% CI 0.83 to 1.48, random-effect model; OR = 1.08, 95% CI 0.88 to 1.32, fixed-effect model) than in the weak studies (OR = 0.66, 95% CI 0.41 to 1.08, random-effect model; OR = 0.70, 95% CI 0.54 to 0.90, fixed-effect model). There were no differences in the effects according to whether the studies were published or unpublished. General (any) recidivism was examined in 13 studies (see Table C1, Appendix C).13,92 Results from both fixed- and random-effect analyses indicated that the recidivism rate was statistically significantly lower in the treatment groups than in the comparison groups. Treatment appeared to be effective for both adults (OR = 0.71, 95% CI 0.56 to 0.90, random-effect model; OR = 0.79, 95% CI 0.69 to 0.90, fixed-effect model) and adolescents (OR = 0.24, 95% CI 0.09 to 0.65, random-effect model; OR = 0.31, 95% CI 0.17 to 0.56, fixed-effect model). There were no differences in the treatment effects according to whether the research design was good or weak, whether the study was published or unpublished, or whether the treatment was delivered in the community or in an institution. The results reported by Hanson et al.13,92 suggest that the RNR principles are relevant to the treatment of sexual offenders. According to the reviewers, the pattern of results was consistent with the direction predicted by the RNR principles “in both the full set of included studies as well as in the published studies with better design”.13 When adherence to the RNR principles was considered, there was relatively little residual variability. According to the reviewers, programs that did not adhere to any of the principles had consistently low treatment effects. For programs adhering to all three principles, the treatment effects were consistently large. Only for programs adhering to two principles was the variability significant. For the reduction of sexual recidivism (22 studies), fixed-effect comparisons showed that programs were more effective if they targeted criminogenic needs (need principle) and were delivered in a manner that was likely to engage the offenders (responsivity principle).13,92 The OR values for the high-risk samples were not significantly different than those for the other samples, although, according to the reviewers, the direction of the treatment effect was consistent with the risk principle, with stronger treatment effects for the high risk offenders. Both the fixed-effect and random-effect models found that effectiveness of treatment increased according to the total number

30 Treatment of Convicted Adult Male Sex Offenders of principles adhered to (none, only one, any two, all three, corresponding to OR = 1.17, 95% CI 0.77 to 1.77; OR = 0.64, 95% CI 0.42 to 0.92; OR = 0.63, 95% CI 0.38 to 1.08; and OR = 0.21, 95% CI 0.70 to 0.64, respectively, using random-effect estimates). For the 10 studies that examined sexual and violent recidivism, there were no significant differences based on adherence to RNR principles although, according to the reviewers, all effects were in the expected directions.13,92 For the 13 studies that examined general (any) recidivism, the fixed-effect model found stronger effects for treatments adhering to the responsivity principle as well as for treatments adhering to all three principles.13,92 According to the reviewers, all the effects were in the direction predicted by the RNR principles, although none were statistically significant using the random-effect model. Using the fixed-effect model, the effectiveness of treatment increased significantly according to the number of RNR principles adhered to (b =-0.15, 95% CI -0.27 to -0.03). The same pattern was found using the random-effect estimates, although the confidence interval included zero (b = -0.19, 95% CI -0.39 to 0.01), indicating that the comparison was not statistically significant. Hanson et al.13,92 also reported that recent treatments were more effective, on average, than the treatments delivered in previous decades. The starting date for the treatment ranged between 1965 and 1997 (M = 1986, SD = 8.5 years, median of 1989). For all outcomes, the linear association was statistically significant for both the fixed- and random-effects models. According to the reviewers, although in the evaluated studies CBT approaches were “the norm”, the treatments examined in the better design studies were diverse.13,92 However, they did not report separate results on the effectiveness of any CBT approaches for different sex offender typologies. No details are provided on the characteristics of the SOT programs and psychological interventions examined by any of the 23 the primary research studies included in their analyses. Neither do they provide details on the included sex offenders’ characteristics or separate information on the length of follow-up in the selected studies that specifically addressed adult sex offenders.

Ongoing research The Criminal Justice Research Centre with the Ohio State University is currently sponsoring a research project entitled “Sex Offender Treatment: Does it Work?” (http://cjrc.osu.edu/researchprojects/sexoffendertx.html). This is an SR and meta-analysis that aims to identify treatment programs that are most effective in reducing recidivism for male and female sexual offenders. This project also aims to identify conditions of this treatment and under what circumstances these programs are effective. The Criminal Justice Research Centre with the Ohio State University was contacted for more details about the methodological approach of this study

Treatment of Convicted Adult Male Sex Offenders 31 and its current stage. However, at the time this report was completed, no response had been received.

Discussion During the past decade the number of programs established to provide treatment to individuals who have sexually offended has grown significantly.1,2,7,9-11,13,14,51,58,82,90 At the same time, the field has developed increasingly sophisticated theories, classification systems, and specialized treatment interventions to understand and treat these individuals. Despite this growth, research on the efficacy/effectiveness of SOT interventions and programs has been slow to mature, and results have been contradictory. Accordingly, the perceived efficacy/effectiveness and value of SOT programs and the views on how best to manage adult male sex offenders have been inconsistent. This overview addressed two questions: Which of the available SOT interventions and programs are effective in terms of reducing the risk of reoffending by convicted adult male sex offenders? and What are the optimal characteristics of a successful SOT program for this population? In this report, results from the best quality SRs on this topic are presented to answer these questions.

Which of the available SOT interventions and programs are effective? According to the results reported by seven moderate- to high-quality SRs,10- 13,58,88,90 the debate in the scientific literature on what SOT interventions and programs work for adult male sex offenders remains divided. Five SRs11-13,88,90 have concluded that psychological treatment using CBT approaches reduces the risk of recidivism in this population whereas the authors of two SRs10,58 have concluded that the evidence is insufficient to draw such a conclusion. Although some of the selected SRs suggest a positive effect for CBT on both sexual and general recidivism, methodological problems, inconsistent results, and a lack of high-quality primary research studies included in the SRs raise uncertainty about which of the available approaches work for adult male sex offenders. The primary research included in the moderate-to-high quality SRs10-13,58,88,90 was focused on answering the question: Does SOT works or not? While the absolute recidivism rates reported in the SOT literature are subject to issues of under reporting, the focus of these studies was on the relative rates of the treated sex offenders compared to the untreated ones. It is therefore apparent that this research provides evidence for some efficacy of treatment in the reduction of recidivism among convicted adult male sex offenders. Therefore the answer to the question, Does SOT using CBT approaches work? is a cautious “Yes” as, overall, the best available evidence shows small but statistically significant reductions in sexual and general recidivism in adult male sex offenders after undergoing CBT.

32 Treatment of Convicted Adult Male Sex Offenders The pharmacotherapy treatments are promising but tend to be the most controversial.3,5,6,11,14,48,49 Hormonal treatment showed encouraging results in the meta-analysis by Schmucker and Losel.11 However, according to Schmucker and Losel, “it is necessary to collect more solid knowledge on circumstances and modes that may prove hormonal treatment to be reasonable”.11 This medication may be helpful as an adjunct to psychotherapy,3,11,32,48,49,68,72 given the offender’s informed consent and ability to withdraw from the medication.6,68,72,95,96 However, it is unlikely that such treatment alone will meet all the needs of most sex offenders.3,5,6,11,32,48,49,68,72 As some sex offenders may commit their offences for reasons other than sexual gratification, simply reducing the libido for these offenders would not necessarily be sufficient to control their potential to reoffend. The associated potential serious side effects, noncompliance, and treatment dropout are common problems for hormonal therapy, which is effective only while the drugs are being taken. After cessation of therapy recidivism rates showed a marked increase.11 The most comprehensive and contemporary review of the role for SSRI anti-depressants for sexual offenders3 highlighted significant methodological flaws in the available primary research on this topic and indicated that further investigation of treatment of sexual offenders with SSRI antidepressants is warranted. The studies identified by this review did not provide sufficient data to prove that SSRIs alone or in combination with psychotherapy are effective in terms of reduced recidivism.

What are the optimal SOT programs for convicted adult male sex offenders? A large number of SOT programs using CBT approaches for convicted adult male sex offenders were evaluated in the primary research included in the moderate- to high-quality SRs.10-13,58,88,90 However, since the evaluated programs were not sufficiently documented (see Table C1 in Appendix C), it was not possible to identify if any characteristics or elements contributed more or less to the success or failure of a program and who of the involved offenders were most likely to benefit from or be harmed by treatment. SOT programs typically work within a broad CBT framework but may vary in terms of resources, philosophy of a program and its treatment objectives, timing, duration, format, intensity, and content of treatment, level of worker expertise and treatment fidelity/integrity as well as the referred sex offenders’ characteristics and selection criteria for participation in the program (which can be based on various risk assessment modalities or no risk assessment at all).1,2,4-7,11-14,45,51 Because of the lack of information on these variables, three of the moderate- to high-quality SRs11-13 the researchers conducted analyses to determine which of the treatment and sex offender characteristics might moderate the treatment effects observed in their primary research studies (see Table 2 below).

Treatment of Convicted Adult Male Sex Offenders 33 Table 2: Systematic reviews that conducted moderator analyses

Moderator variables

Systematic Treatment review characteristics Offender characteristics

Hanson et al., Treatment approach: CBT showed Age: CBT approaches delivered after 12 2002 significant reduction in sexual and 1980 appeared to be equally effective general recidivism for adults and adolescents (significantly reduced sexual and general recidivism Specific treatment: sex offender compared to alternate or no treatment) specific treatment for adults was associated with significant reduction in Treatment participation: dropouts had general recidivism higher sexual and general recidivism rates than completers; in contrast to Recency*: current treatments findings for sexual recidivism, those (CBT delivered after 1980) who refused SOT were more likely to Setting: institutional and community recidivate with any offence programs using current treatment were both associated with significant reductions in sexual and general recidivism compared to alternate or no treatment; community programs appeared to have a stronger effect on general recidivism than institutional programs

Schmucker and Treatment approach: hormonal Age: programs that specifically 11,91 Losel, 2008 medication, CBT, and classic behaviour addressed adolescents had a higher therapy had a significant impact on effect than those for adult offenders; sexual recidivism; non-behavioural the difference was not treatments (e.g., insight-oriented, statistically significant therapeutic community) had no Offence type: comparisons showed significant impact on recidivism significant effects on recidivism for all Specific treatment: only program types of sex offences (extra-familial designed specifically for sex offenders child molesters, exhibitionists, and had a significant effect on recidivism rapists) except for incest offenders Recency: more “modern” programs Treatment participation: whether (1990s) did not prove to be successful treatment was completed or terminated in reducing recidivism prematurely had an impact on sexual recidivism, with completers showing Setting: although setting variable better effects than controls and showed no significant difference in dropouts showing worse effects than recidivism, there was a strong tendency controls (the odds to not recidivate for larger effects for outpatient settings for dropouts was half the odds for and smaller effects for institutions untreated offenders) Treatment format: whether treatment Therapy motivation: voluntary was delivered in an individual or group participation showed a significant format did not result in significant effect on recidivism and obligatory differences in recidivism participation and mixed conditions Author affiliation: when study authors resulted in no effect were involved in program there was a larger effect on recidivism than when the program was evaluated by independent researchers

34 Treatment of Convicted Adult Male Sex Offenders Hanson et al., Treatment approach: CBT that adhere Age: overall, treatments offered to 13,92 2009 to RNR principles adults and those offered to adolescents had similar effects in terms of sexual, Recency: recent treatments sexual or violent, or general recidivism (implemented/delivered after 1980s) were more effective, on average than older treatments Setting: there were no differences in the effects on sexual recidivism, sexual or violent recidivism, or general recidivism according to whether programs were delivered institutional and community settings (setting did not moderate treatment effects for sex offenders)

* Time of treatment implementation CBT – cognitive behavioural therapy; RNR –Risk/Need/Responsivity; SOT – sex offender treatment However, these reviews11-13 reported their findings differently in terms of the outcome of interest (which was defined and measured differently), and they did not reach similar conclusions for some of the evaluated characteristics (see Table 2 above). Their reported aggregate findings did not allow a differential analysis for determining which is a model of effective SOT program, who is likely to benefit, and under what circumstances. All moderate- to high-quality SRs10-13,58,88,90 concluded that more and better conducted and reported primary research studies are needed to focus on which specific treatment works with whom and under what circumstances. Few primary research studies in this field used strong research designs (i.e., random assignment), and even fewer studies with strong research designs examined interventions consistent with contemporary standards. Only papers reporting during the early 1980s on the long-term RCT conducted in the community by Romero and Williams and several papers reporting between 1994 and 2005 preliminary and final findings from the RCT conducted in a hospital setting (the SOTEP study) were included in most SRs.10-13,58,88 While findings from these two RCTs do not necessarily support the efficacy of treating convicted adult male sex offenders with group psychotherapy and CBT/RP, in the interpretation of their findings it must be recognized that the programs they evaluated were developed and delivered during the 1970s and 1980s. The design and implementation of these programs predated the advances in research and practice, and therefore the findings reported by these two RCTs, considered within the context of the more recent developments in the field, may be less than optimal. The authors of the SOTEP study97 cited a number of factors that might have impacted their overall null findings. These include a program design that might have not been considered the state of the art when evaluated against current standards, a less-than-optimal individualization of treatment based on risk and needs, and the lack of a more developed and collaborative after-care component.

Treatment of Convicted Adult Male Sex Offenders 35 The SOTEP study97 provides some evidence of the differential impact of CBT/RP treatment on different types of offenders. It reported that child molesters who met the treatment goals reoffended at lower rates than those who did not demonstrate as much progress in treatment. Similarly, high-risk sex offenders who understood the program concepts, or “got it”, and evidenced more progress in treatment had lower rates of sexual recidivism than high-risk sex offenders who “did not get it” and made less progress in treatment. These findings are consistent with those from one of the most recently published best quality SRs conducted by Hanson et al.,13 which revealed better recidivism outcomes when sex offenders were matched differentially to psychological interventions based on identified levels of risk and needs. Hanson et al.13 suggest considering SOT as a special case in the treatment of general offenders as one approach to making decisions concerning the quality of SOT programs. They believe that the research evidence supporting the “what works” RNR principles for general offenders is sufficient, so they should be a primary consideration in the design and implementation of SOT programs using CBT approaches for sexual offenders. Drawing on their data set and analyses, the reviewers suggest that policy-makers and researchers concerned with improving SOT programs would benefit from carefully considering the RNR principles and the importance of “selecting staff based on relationship skill, using treatment manuals, training staff, and starting small”.13 According to Hanson et al.,13 although most contemporary SOT programs using CBT already conform to some aspects of the responsivity principle, many do not meet the test for the need principle. Attention to this principle would create the largest changes in the therapeutic interventions currently given to sexual offenders. In their included primary research, many of the evaluated programs made special efforts to engage sexual offenders in treatment. The reviewers mentioned that, although much remains to be known about how best to apply the risk principle to sexual offenders, treatment providers should be aware that noticeable reductions in recidivism are not to be expected among the lowest risk offenders. Other treatment goals, such as meaningful reintegration into the community, may be appropriate for these cases. Transferring the RNR principles into “real-world” SOT program settings appears to be very challenging, given that the correctional agencies/facilities have a diverse workforce in terms of education, values and experience, and policies and management practices.61,63-65,67,98-101 The RNR model has been criticized regarding an apparent failure to appreciate the totality of client needs, specifically with respect to offender responsivity concerns.29,59,63,64,100,101 Recently it has been suggested that sex offender rehabilitation would benefit from an integration of the RNR and good lives models in attempting to maximize the treatment gain and reduce recidivism.63,64,99-101 In offering effective SOT interventions, consideration of treatment readiness is a necessity, as is attention to approaches that seek to engage offenders.

36 Treatment of Convicted Adult Male Sex Offenders Does setting/location of the SOT program affect its impact on outcomes of interest? This question could not be answered with certainty since the selected SRs reached slightly different conclusions on the impact of setting based on their included primary research, which did not provide sufficient details on the evaluated SOT programs and interventions. Of the three SRs that analyzed setting or location of the program as a variable that might account for changes in treatment effect for adult and adolescent sex offenders, two12,13 reported that it did not moderate the treatment effect, while the third one11 reported a trend for larger treatment effect for outpatient settings than for in-patient settings. According to Hanson et al.,13 the available literature on SOT “does not provide strong tests of whether program location matters, given that no studies have directly compared the same treatment in both settings”.

What do we know? —— SOT programs neither cure sexual offending nor guarantee a complete cessation of offending, and they represent one element in a comprehensive risk management strategy designed for convicted adult male sex offenders. —— Not all SOT interventions and programs are effective in reducing sexual/non-sexual recidivism in this population. —— SOT programs using CBT approaches that follow the RNR principles have the potential to reduce the risk of reoffending among adult male sexual offenders. —— Although pharmacotherapy (using hormonal or non-hormonal agents) may be helpful as an adjunct to psychotherapy treating adult male sex offenders, better designed clinical studies are needed to establish its role in SOT programs for this population. —— There are still uncertainties regarding the timing of treatment, optimum intensity and length of treatment, how it should be delivered, which offenders benefit most, and how to assess the risk of reoffending pre- and post-treatment.

What don’t we know yet? Future well-designed, -conducted, and -reported research is warranted as several questions remain unanswered about SOT interventions and programs. —— What are the defining characteristics that identify sex offenders who would benefit most from psychotherapy and/or pharmacotherapy? —— Which of the available CBT approaches and/or pharmacotherapy options (if any) are most effective and for which types of sex offenders? —— Which subgroups of sex offenders are mostly likely to be prematurely terminated or drop out from treatment?

Treatment of Convicted Adult Male Sex Offenders 37 —— What are the optimal components of SOT programs that affect treatment outcomes for convicted adult male sex offenders? —— Does setting or location of the SOT program affect recidivism rates?

Limitations Although the present study used a systematic and structured approach to overview the available SRs, it has several limitations. The literature review was confined only to secondary research written in English and published from 1998 onward. The focus on secondary research ensured the inclusion of data prior to 1998, as the primary research studies included in the selected SRs were published as far back as 1953. On the other hand, the studies that were published decades ago may no longer be relevant due to the major changes that have occurred during the past two decades in the SOT field in terms of goals, content, methods, style, and dose. The studies were initially selected by examining the abstracts of these articles, and it is possible that some studies were inappropriately excluded prior to the examination of the full-text article. However, where details were lacking, ambiguous papers were retrieved as full text to minimize this possibility. Proprietary reports were excluded. Only full-text articles were included because abstracts provide insufficient details to allow an accurate, unbiased assessment and comparison of the study results. The authors of the abstract-only publications were not contacted for full details of their studies. Outcome-reporting bias and inclusion criteria bias are unlikely as the reviewers had no detailed knowledge of the topic literature, and the scope of the review and its methodological approach were defined a priori. The review scope was developed with the assistance of a group of experts in the assessment and treatment of sex offenders from both the Northern and Southern Forensic Psychiatry services in Alberta. However, most of the primary research studies reviewed by the selected SRs were conducted outside Alberta, and therefore their generalizability to this Canadian province’s context may be limited and needs to be considered. This review was confined to an examination of the efficacy/effectiveness and safety of the therapeutic interventions of interest and did not consider ethical or legal considerations associated with these interventions.

38 Treatment of Convicted Adult Male Sex Offenders CONCLUSIONS During the past decade there has been a multitude of descriptive and meta-analytic reviews which synthesized and analyzed the growing body of research on sex offenders and the efficacy/effectiveness of increasingly sophisticated interventions and programs designed to treat and manage them. Their results have provided useful information on SOT and possible markers for effective interventions and programs. The findings reported by moderate- to high-quality SRs indicate a small statistically significant effect size in reducing sexual and nonsexual recidivism rates by SOT programs in convicted adult male sex offenders. Of all available SOT programs, the reported findings provide a stronger support for the effects of those using CBT approaches and adhering to the RNR model. However, these results should be interpreted with caution as the available evidence is based on weak study designs. When analyses were restricted to the few available RCTs, a comparable mean effect was shown, but it was not statistically significant. Despite inconsistent evidence, there are important reasons for continuing to provide and evaluate SOT interventions and programs, including the fact that even small reductions in recidivism are significant in terms of reduction of harm to potential victims and the community and savings in intangible and tangible costs to society. However, the results reported by the selected SRs provide little direction concerning how to improve current treatment practice. It is not clear whether all sex offenders require treatment or whether current therapeutic interventions are appropriate for all subgroups of offenders. There also remains much disagreement concerning what are the most useful components and elements of SOT programs that would ensure meaningful rehabilitation for convicted adult male sex offenders and limit the number of future victims. Neither does the available research on SOT provide clear answers on whether program location matters. More and better designed, conducted, and reported primary research is warranted to resolve these uncertainties.

Treatment of Convicted Adult Male Sex Offenders 39 Appendix A: METHODS

Search strategy The search strategy was created and carried out prior to the study selection process. A comprehensive literature search with no date and language limits was conducted by the IHE Research Librarian in January 2009 and updated in June 2010. Studies about adolescent or female offenders were eliminated from the results, and so were studies about sex offenders with neurodevelopmental disorders. The Internet sites of pertinent agencies, institutions, and departments of corrections and their links for relevant material were searched. To identify more unpublished work, local experts in the field of sexual offender treatment were contacted and asked if they knew or had personally conducted further evaluations. In addition to the strategy outlined below, reference lists of retrieved articles were reviewed for potential studies.

Table A1: Search strategy † See below for limits

Edition or date Database searched Search terms ††

Core Databases

Cochrane Database of 3 June 2010 #1 MeSH descriptor Paraphilias explode all trees Systematic Reviews #2 (sex* AND (offender* OR abuser* OR perpetrator* ) http://www. ):ti,ab,kw thecochranelibrary.com #3 (child AND molest* ):ti,ab,kw #4 (pedophil* OR paedophil* OR rapist ):ti,ab,kw #5 (#1 OR #2 OR #3 OR #4)

(4 results before removing duplicates)

The Campbell Library 17 June 2010 Sex* AND offend* in All Text http://www. campbellcollaboration. (Browsed through 46 results) org/library.php (2 results)

40 Treatment of Convicted Adult Male Sex Offenders Table A1: Search strategy (CONTINUED) † See below for limits

Edition or date Database searched Search terms ††

Core Databases

MEDLINE includes in- 1950 to 1. exp paraphilias/ process citations (OVID 3 June 2010 2. paraphili*.tw. Interface) 3. exhibitionist*.tw. 4. sadist*.tw. 5. masochist*.tw. 6. voyeur*.tw. 7. rapist*.tw. 8. (pedophil* or paedophil*).tw. 9. frotteur*.tw. 10. necrophil*.tw. 11. (sex* adj2 delinquen*).tw. 12. (child adj2 molest*).tw. 13. (public adj3 masturbat*).tw. 14. sexually abusive.tw. 15. sex* offend*.tw. 16. sexual abuser*.tw. 17. or/1-16 18. exp Sex Offenses/ 19. (offender* or abuser* or deviant* or molester* or pervert* or perpetrat* or commit*).tw. 20. 18 and 19 21. 17 or 20 22. therap*.tw. 23. dt.fs. 24. th.fs. 25. tu.fs. 26. intervention*.tw. 27. treatment*.tw. 28. (inpatient* or outpatient* or aftercare).tw. 29. (program or programs).ti. 30. Cognitive Therapy/ 31. Psychotherapy/ 32. (victim adj3 empathy).tw. 33. cognitive restructuring.tw. 34. relapse prevention.tw. 35. life planning.tw. 36. goal attainment.tw. 37. anger management.tw. 38. (pharmacotherap* or pharmaceuticals or prescription* or medicat* or drug therapy).tw. 39. (anaphrodisiac or antiandrogen or castrat* or antilibidinal).mp. 40. placebo.mp. 41. or/22-40 42. exp Recurrence/ 43. recidivi*.mp.

Treatment of Convicted Adult Male Sex Offenders 41 Table A1: Search strategy (CONTINUED) † See below for limits

Edition or date Database searched Search terms ††

Core Databases

44. reoffen*.mp. 45. reconvict*.mp. 46. convictions.mp. 47. (relapse* or lapse*).mp. 48. rearrest*.mp. 49. rehabilitat*.ti. 50. risk reduction.mp. 51. exp treatment outcome/ 52. outcome*.mp. 53. ((effective or success* or ineffective or unsuccessful adj2 (treatment* or therapy)).tw. 54. (efficacy or effective* or ineffective or unsuccessful or success*).ti. 55. (plethysmograph* or phallometric).mp. 56. (deviant adj2 arousal).tw. 57. or/42-56 58. ((juvenile* or adolescent* or youth*) not adult).ti. 59. (women not (men or male)).ti. 60. 21 and 41 and 57 61. 60 not (58 or 59) 62. limit 61 to case reports 63. 61 not 62

(478 results before removing duplicates)

CRD Databases (DARE, 3 June 2010 # 1 MeSH Paraphilias EXPLODE 1 HTA & NHS EED) # 2 sex* AND (offender* OR abuser* OR perpetrator* ) # 3 child AND molest* # 4 pedophil* OR paedophil* OR rapist* # 5 #1 OR #2 OR #3 OR #4

(27 results before removing duplicates)

EMBASE (Ovid 1980 to 2010 1. exp Sexual Deviation/ Interface) week 21 2. paraphili*.tw. 3. exhibitionist*.tw. 4. sadist*.tw. 5. masochist*.tw. 6. voyeur*.tw. 7. rapist*.tw. 8. (pedophil* or paedophil*).tw. 9. frotteur*.tw. 10. necrophil*.tw. 11. (sex* adj2 delinquen*).tw. 12. (child adj2 molest*).tw. 13. (public adj3 masturbat*).tw. 14. (sex* adj3 perpetrat*).tw. 15. sexually abusive.tw.

42 Treatment of Convicted Adult Male Sex Offenders Table A1: Search strategy (CONTINUED) † See below for limits

Edition or date Database searched Search terms ††

Core Databases

16. sex* offend*.tw. 17. sexual abuser*.mp. 18. or/1-17 19. exp sexual abuse/ 20. (offender* or abuser* or deviant* or molester* or pervert* or commit*).mp. 21. 19 and 20 22. 18 or 21 23. therap*.tw. 24. dt.fs. 25. th.fs. 26. intervention*.tw. 27. treatment*.tw. 28. (inpatient* or outpatient* or aftercare).tw. 29. (program or programs).ti. 30. exp psychotherapy/ 31. (victim adj3 empathy).tw. 32. cognitive restructuring.tw. 33. relapse prevention.tw. 34. life planning.tw. 35. goal attainment.tw. 36. anger management.tw. 37. exp drug therapy/ 38. (anaphrodisiac or antiandrogen or castrat* or antilibidinal).tw. 39. Placebo/ 40. or/23-39 41. recidivi*.mp. 42. reoffen*.tw. 43. reconvict*.tw. 44. convictions.tw. 45. (relapse* or lapse*).mp. 46. rearrest*.tw. 47. rehabilitat*.ti. 48. risk reduction.mp. 49. exp treatment outcome/ 50. outcome*.tw. 51. ((effective or success* or ineffective or unsuccessful) adj2 (treatment* or therapy)).tw. 52. (efficacy or effective* or ineffective or unsuccessful or success*).ti. 53. (plethysmograph* or phallometric).mp. 54. (deviant adj2 arousal).tw. 55. or/41-54 56. ((juvenile* or adolescent* or youth*) not adult).ti.

Treatment of Convicted Adult Male Sex Offenders 43 Table A1: Search strategy (CONTINUED) † See below for limits

Edition or date Database searched Search terms ††

Core Databases

57. (women not (men or male)).ti. 58. (22 and 40 and 55) not (56 or 57) 59. Case Report/ 60. 58 not 59

(522 before removing duplicates)

PsycINFO (Ovid 1806 to 1. exp paraphilias/ interface) June week 1, 2. paraphil*.tw. 2010 3. (masochist* or sadist* or voyeur* or frotteur* or necrophil*).tw. 4. (pedophil* or paedophil*).tw. 5. (sex* adj2 offender*).tw. 6. (sex* adj2 (deviant* or deviat*)).tw. 7. (sex* adj2 delinquen*).tw. 8. (child adj2 molest*).tw. 9. (public adj3 masturbat*).tw. 10. (sex adj3 perpetrat*).tw. 11. sexually abusive.tw. 12. sexual abuser*.tw. 13. rapist*.tw. 14. or/1-13 15. exp sex offenses/ 16. (offender* or abuser* or molester* or pervert or commit*).tw. 17. 15 and 16 18. 14 or 17 19. cognitive therapy/ or cognitive behavior therapy/ or cognitive restructuring/ 20. drug therapy/ 21. (anaphrodisiac or antiandrogen or castrat* or antilibidinal).tw. 22. placebo/ 23. exp psychotherapy/ 24. treatment/ 25. behavior modification/ 26. psychotherapeutic techniques/ 27. (program or programs).ti. 28. (inpatient* or outpatient* or aftercare).tw. 29. anger control/ 30. relapse prevention/ 31. (victim adj3 empathy).tw. 32. life planning.tw. 33. goal attainment.tw. 34. intervention*.tw. 35. treatment*.tw.

44 Treatment of Convicted Adult Male Sex Offenders Table A1: Search strategy (CONTINUED) † See below for limits

Edition or date Database searched Search terms ††

Core Databases

36. therap*.tw. 37. or/19-36 38. “Relapse (Disorders)”/ 39. recidivi*.mp. 40. reoffen*.tw. 41. reconvict*.tw. 42. convictions.tw. 43. rearrest*.tw. 44. rehabilitat*.ti. 45. risk reduction.mp. 46. exp treatment outcomes/ 47. “recovery (disorders)”/ 48. outcome*.mp. 49. (effective or success* or ineffective or unsuccess* adj2 (treatment* or therapy).tw. 50. plethysmography/ 51. (efficacy or effective* or ineffective or success* or unsuccess*).ti. 52. (deviant adj2 arousal).tw. 53. or/38-52 54. Treatment Effectiveness Evaluation/ 55. 18 and ((37 and 53) or 54) 56. ((juvenile* or adolescent* or youth*) not adult*).ti. 57. 55 not 56 58. (women not men).ti. 59. 57 not 58 60. limit 59 to (“0200 clinical case study” or 1000 mathematical model or 1400 nonclinical case study) 61. 59 not 60

(1208 before duplicates removed)

Violence and Abuse 2 June 2010 S1 TI (rapist* OR sex* offend* OR sexually abusive OR Abstracts (EBSCO) child molesters OR (sex* AND predator*) OR pedophile* OR paedophile* OR sex* abusers) or SU sex offenders S2 (therap* or treatment or rehabilitation or psychotherapy or pharmacotherapy OR intervention*) OR TI program* S3 outcome* or success* or efficacy or ineffective* or unsuccess* or effectiv* or recidivi* or reoffend* or rearrest* or reconvict* or relapse* S4 TI (juvenile* OR youth* or teen* or adolescent) not TI adult S5 TI (women or female) not TI (men or male) S6 (S1 AND S2 AND S3) NOT (S4 OR S5) (167 References before duplicate removal)

Treatment of Convicted Adult Male Sex Offenders 45 Table A1: Search strategy (CONTINUED) † See below for limits

Edition or date Database searched Search terms ††

Core Databases

Criminal Justice 2 June 2010 ((DE=”child molesters” OR DE=”sex offenders” OR Abstracts (CSA TI=(rapist* OR sex* offend* OR sexually abusive OR interface) child molesters OR (sex* AND predator*) OR pedophile* OR paedophile* OR sex* abuser*)) and(KW=(outcome* or success* or efficacy or ineffective* or unsuccess* or effectiv* or recidivi* or reoffend* or rearrest* or reconvict* or relapse*)) and (KW=(therap* or treatment or rehabilitation or psychotherapy or pharmacotherapy OR intervention*) OR TI=program*)) NOT ((TI=(women or female) NOT TI=(male or men)) OR (TI=(juvenile* OR youth* OR teen* OR adolescent*) NOT TI=adult*))

(407 References before duplicate removal)

Sociological abstracts 2 June 2010 ((TI=(rapist* OR sex* offend* OR sexually abusive OR child molesters OR (sex* AND predator*) OR pedophile* (CSA Interface) OR paedophile* OR sex* abuser*)) and (KW=(outcome* or success* or efficacy or ineffective* or unsuccess* or effectiv* or recidivi* or reoffend* or rearrest* or reconvict* or relapse*)) and(KW=(therap* or treatment or rehabilitation or psychotherapy or pharmacotherapy OR intervention*) OR TI=program*)) NOT ((TI=(women or female) NOT TI=(male or men)) OR (TI=(juvenile* OR youth* OR teen* OR adolescent*) NOT TI=adult*))

(63 References before duplicate removal)

SocINDEX with Full Text 2 June 2010 S1 (TI (rapist* OR sex* offend* OR sexually abusive (EBSCO) OR child molesters OR (sex* AND predator*) OR pedophile* OR paedophile* OR sex* abusers)) or DE “sex offenders” S2 (therap* or treatment or rehabilitation or psychotherapy or pharmacotherapy OR intervention*) OR TI program* S3 outcome* or success* or efficacy or ineffective* or unsuccess* or effectiv* or recidivi* or reoffend* or rearrest* or reconvict* or relapse* S4 TI (juvenile* OR youth* or teen* or adolescent) not TI adult S5 TI (women or female ) not TI (men or male) S6 (S1 AND S2 AND S3) NOT (S4 OR S5)

(732 References before duplicate removal)

46 Treatment of Convicted Adult Male Sex Offenders Table A1: Search strategy (CONTINUED) † See below for limits

Edition or date Database searched Search terms ††

Core Databases

Social Work Abstracts 2 June 2010 S1 (TI (rapist* OR sex* offend* OR sexually abusive OR (EBSCO) child molesters OR (sex* AND predator*) OR pedophile* OR paedophile* OR sex* abusers )) S2 (therap* or treatment or rehabilitation or psychotherapy or pharmacotherapy OR intervention* ) OR TI program* S3 outcome* or success* or efficacy or ineffective* or unsuccess* or effectiv* or recidivi* or reoffend* or rearrest* or reconvict* or relapse* S4 (S1 AND S2 AND S3) NOT TI (juvenile* OR adolescent*)

(37 References before duplicate removal)

Social Services 2 June 2010 ((TI=(rapist* OR sex* offend* OR sexually abusive OR Abstracts (CSA) child molesters OR (sex* AND predator*) OR pedophile* OR paedophile* OR sex* abuser*)) and (KW=(outcome* or success* or efficacy or ineffective* or unsuccess* or effectiv* or recidivi* or reoffend* or rearrest* or reconvict* or relapse*)) and(KW=(therap* or treatment or rehabilitation or psychotherapy or pharmacotherapy OR intervention*) OR TI=program*)) NOT ((TI=(women or female) NOT TI=(male or men)) OR (TI=(juvenile* OR youth* OR teen* OR adolescent*) NOT TI=adult*)) (117 References before duplicate removal)

Gender Studies 2 June 2010 S1 (TI (rapist* OR sex* offend* OR sexually abusive OR Database (EBSCO) child molesters OR (sex* AND predator*) OR pedophile* OR paedophile* OR sex* abusers )) S2 (therap* or treatment or rehabilitation or psychotherapy or pharmacotherapy OR intervention* ) OR TI program* S3 outcome* or success* or efficacy or ineffective* or unsuccess* or effectiv* or recidivi* or reoffend* or rearrest* or reconvict* or relapse* S4 TI (juvenile* OR youth* or teen* or adolescent ) not TI adult

S5 TI (women or female ) not TI (men or male) S6 (S1 AND S2 AND S3) NOT (S4 OR S5)

(138 before duplicate removal)

Treatment of Convicted Adult Male Sex Offenders 47 Table A1: Search strategy (CONTINUED) † See below for limits

Edition or date Database searched Search terms ††

Core Databases

National Criminal Justice 2 June 2010 ((TI=(rapist* OR sex* offend* OR sexually abusive OR Reference Service child molesters OR (sex* AND predator*) OR pedophile* Abstracts (CSA) OR paedophile* OR sex* abuser*)) and (KW=(outcome* or success* or efficacy or ineffective* or unsuccess* or effectiv* or recidivi* or reoffend* or rearrest* or reconvict* or relapse*)) and (KW=(therap* or treatment or rehabilitation or psychotherapy or pharmacotherapy OR intervention*) OR TI=program*)) NOT ((TI=(women or female) NOT TI=(male or men)) OR (TI=(juvenile* OR youth* OR teen* OR adolescent*) NOT TI=adult*))

(613 before duplicate removal)

Web of Science 2 June 2010 ((TI=(rapist* OR sex* offend* OR sexually abusive OR child molesters OR (sex* AND predator*) OR pedophile* ISI Interface Licensed OR paedophile* OR sex* abuser*)) and (TS=(outcome* Resource or success* or efficacy or ineffective* or unsuccess* or effectiv* or recidivi* or reoffend* or rearrest* or reconvict* or relapse*)) and (TS=(therap* or treatment or rehabilitation or psychotherapy or pharmacotherapy OR drug therapy OR intervention*) OR TI=program*)) NOT ((TI=(women or female) NOT TI=(male or men)) OR (TI=(juvenile* OR youth* OR teen* OR adolescent*) NOT TI=adult*)) (383 before duplicate removal)

NEOS Library 2 June 2010 sex offense AND (treatment OR therapy) http://www.library. ualberta.ca/catalogue

Sex offender websites

Center for Sex Offender 2 June 2010 Browsed publications (1 document)(no new results) Management: http:// www.csom.org/

Internet

Google 2 June 2010 Sex offender treatment meta-analysis OR systematic review (browsed first 100 results) http://www.google.com

Note:††“*”, “# “, and “?” are truncation characters that retrieve all possible suffix variations of the root word e.g. surg* retrieves surgery, surgical, surgeon, etc. Searches separated by semicolons have been entered separately into the search interface

48 Treatment of Convicted Adult Male Sex Offenders Study selection process The systematic reviews (SRs) were independently selected by two reviewers (PC and MO) using a predefined set of inclusion and exclusion criteria. Disagreements between reviewers during the study selection process were resolved by consensus. The two reviewers were not blinded to any aspects of the published papers being evaluated during the study selection process. The initial identification of potentially relevant secondary research studies was based on the screening of titles and abstracts retrieved by the literature searches. The selection was confined to studies published since January 1998 that reflect the current knowledge about sex offenders and recently reported shifts in the approach to SOT. The retrieval of potentially relevant articles was limited to studies written in English. The retrieved full-text copies of potentially eligible studies were then independently assessed for eligibility by the same two reviewers using the selection criteria listed below. Only those studies that met the inclusion criteria were used to formulate the evidence base for this HTA. Reasons for exclusion of studies from the review are documented in Table B1, Appendix B.

Inclusion criteria For the purpose of this HTA study, a sexual offence is an officially recorded sexual misbehaviour or criminal behaviour with sexual intent that results in some form of criminal justice intervention or official sanction.102 Only sexual offences against an identifiable victim such as a child or a non consenting adult victim (category “A” in the STATIC-99) will be considered in this review.102 Category “A” sex offences includes the following types of sexual offences: aggravated sexual assault, attempted sexual offences (i.e., attempted rape, attempted sexual assault), contributing to the delinquency of a minor (where the offence had a sexual element), exhibitionism, incest, indecent exposure, invitation to sexual touching, lewd or lascivious acts with a child under 14, manufacturing or creating child pornography where an identifiable child victim was used in the process, molest children, oral copulation, penetration with a foreign object, rape, sexual assault, sexual assault causing bodily harm, sexual battery, sexual homicide, sexual offences against animals, sexual offences involving dead bodies, sodomy, unlawful sexual intercourse with a minor, and voyeuristic activity. Likewise, Internet-related sexual offences were also considered. Published reports of secondary research studies were included in the review for data extraction if they met the following selection criteria: Population – adult male sex offenders (18 years and older) who have been convicted of a sexual offence; both hands-off/non-contact (e.g., exhibitionism, Internet offences, and voyeurism) and contact sex offences (e.g., child molestation, rape) were considered; studies including first time convicted and recidivists with low, medium and/or high risk to reoffend were also considered for this review. Treatment of Convicted Adult Male Sex Offenders 49 Intervention(s) – psychotherapy (psychological or psychosocial therapies) and/or pharmacotherapy (alone or combined), provided in inpatient and/or outpatient settings (i.e., institution-/prison-, hospital-, and/or community-based SOT programs). Comparator(s) – no therapy, placebo, usual care, other therapy (e.g., surgical therapy; alternate therapy, less intensive, and/or less specific therapy). Outcome – violent or nonviolent, sexual or nonsexual recidivism defined as the number of convictions for violent or nonviolent (re-)offences; number of convictions for (re-)offences with or without sexual element. Study design - SRs and HTAs reporting on the efficacy/effectiveness and safety of therapeutic interventions provided within SOT programs for convicted adult male sex offenders. Time frame - published from January 1998 onward. Based on the description by Cook et al.,103 a review or an HTA was selected as a systematic review if it met the following criteria: —— focused clinical question; —— explicit search strategy; —— use of explicit, reproducible, and uniformly applied criteria for article selection; —— critical examination of the included studies to explore the methodological quality differences as an explanation for heterogeneity in study results; —— qualitative or quantitative data synthesis. Background information: Were appropriate, relevant, published material, in the form of clinical reviews, overview articles, commentaries and discussion papers, was included as background information for the various sections of the report. Research studies recommended by professional experts or mentioned in the list of references of reviewed publications that were not identified by the search strategy were included. Only full, peer-reviewed articles were included as abstracts usually do not provide adequate detail on the study’s methodology and results. In the case of duplicate or multiple publications (different articles reporting results based on the same sample of sex offenders or overlapping samples), only the most recent and complete version was included. Only studies conducted in Canada, the United States of America, Australia, New Zealand, and European Union countries were included since the cultural and legal norms, and the access to health care in other countries are likely to be too different from those of Canada to be relevant.

50 Treatment of Convicted Adult Male Sex Offenders Exclusion criteria Published reports of secondary research studies were excluded from data extraction if: —— they were not accessible through our standard information services; —— the full-text article could not be obtained during the report preparation period; —— they evaluated individual therapies that were not part of a SOT program; —— they involved only juvenile sex offenders (between 12 and 18 years of age), female sex offenders, or sex offenders with psychotic disorders, intellectual deficits, or other neurodevelopmental disorders; —— they involved many types of offenders but did not report separate results for convicted adult male sex offenders in both treated and control/ comparison groups; —— they compared data on recidivism only for SOT program completers versus non-completers/dropouts or used only refusers and dropouts (combined) as a control/comparison group; or —— minimum follow-up period was shorter than 2 years after completing the SOT programs. Narrative and descriptive reviews, editorials, letters and comments were excluded from this HTA study. Figure 1 provides a summary of the study selection process. Figure 1: Study selection process

Total number of citations retrieved from literature searches selected for further examination of titles and abstracts N = 1690 Reasons for exclusions Does not meet SR criteria = 40 Full-text articles retrieved and evaluated for selection Not on effectiveness of SOTP = 4 n = 71 Not original research = 4 Not English language = 4 Articles included Excluded Not full text format = 3 n = 10 n = 61 No outcome data = 3 Not on convicted adult sex offenders = 2 Multiple publications Not retrieved = 1 Number of unique n = 2 studies included n= 8

Treatment of Convicted Adult Male Sex Offenders 51 Data extraction Data on study characteristics, targeted populations, interventions, and outcomes were extracted from the selected SRs by one reviewer (PC) using a predefined data extraction form. Data extraction was verified by a second reviewer (MO). Details from the selected SRs are summarized in Table C1 (Appendix C). For studies in which it was not clearly stated whether the literature searches were limited or not by publication language, their authors were contacted for further information. Only the authors who provided an e-mail address as contact information in the published reports of their studies were contacted. At the time of the completion of this report replies had been received from the authors of only four of the selected systematic reviews.10,12,13,89

Methodological quality assessment The methodological quality of the selected SRs was evaluated using the AMSTAR tool, which evaluates the overall methodological quality of systematic reviews on a scale from 0 to 11.104 The quality of the selected SRs was rated as low (score 0 to 3), moderate (score 4 to 7), or high (score 8 to 11). The AMSTAR tool has shown good face, content,104 and reliability105 for evaluating the methodological quality of SRs. Two reviewers (PC and MO) independently assessed the methodological quality of SRs. The reviewers were not blinded to any aspects of the reviews being evaluated. Disagreements among reviewers were solved by consensus. Prior to critically appraising the studies, the two reviewers discussed the AMSTAR tool with respect to the interpretation of the tool’s items. As a result, one item of the AMSTAR tool was modified (question 11). The modified version of the AMSTAR tool as well as the adapted guidelines for the tool are included in Appendix C. Critical appraisal results for all included SRs are presented in Table C2 (Appendix C).

52 Treatment of Convicted Adult Male Sex Offenders Analysis and synthesis of data Evidence tables were used to report information on study design, study population, treatment groups, outcomes, and results abstracted from all selected SRs. Separate tables summarized the identified issues of rigour. Reviews rated by AMSTAR tool to be of moderate- to high-methodological quality qualified for data analysis and synthesis. Data from these selected SRs were synthesized qualitatively.

External review of the overview of systematic reviews External reviewers with expertise in SOT and HTA methodology evaluated the draft report and provided feedback. In selecting reviewers, the practice of the Institute of Health Economics is to choose experts who are well recognized and published in peer-reviewed literature, and who can offer a provincial and/ or national perspective on treatment for convicted adult male sex offenders.

Treatment of Convicted Adult Male Sex Offenders 53 APPENDIX B: EXCLUDED STUDIES The application of the selection criteria described in Appendix A resulted in 60 studies excluded. The primary reasons for exclusion of studies from this HTA were as follows. 1. The study did not meet the SR criteria (n = 40). 2. The study did not report on the effectiveness of sex offender treatment programs (n = 4). 3. The article did not contain original research (n = 4) 4. The study was not published in English language (n = 4). 5. The study was not published in full-text format (abstract only) (n = 3). 6. The study did not report on outcomes of interest as defined in Appendix A (n = 3). 7. The study did not report on convicted adult sex offenders (n = 2) 8. The full text of the study was not retrieved (n = 1) Table B1 lists the excluded full-text reports of the retrieved research studies and the main reasons for their exclusion. Table B1: Excluded research studies Main reason for exclusion: The study did not meet the SR criteria (n = 39)

Abracen J, Looman J, Langton CM. Treatment of sexual offenders with psychopathic traits: recent research developments and clinical implications. Trauma Violence & Abuse 2008;9(3):144-66

Alexander MA. Sexual offender treatment efficacy revisited. Sexual Abuse: Journal of Research and Treatment 1999;11(2):101-16

Baerga-Buffler M, Johnson JL. Sex offender management in the Federal Probation and Pretrial Services System. Federal Probation: A Journal of Correctional Philosophy and Practice 2006;70(1):13-7

Becker JV, Murphy WD. What we know and do not know about assessing and treating sex offenders. Psychology, Public Policy, and Law 1998;4(1-2):116-37

Beenakkers EMT. Effectiveness of correctional treatment: a literature survey. The Hague: Netherlands Ministry of Justice; 2000

Briken P. Pharmacological treatments for paraphilic patients and sexual offenders. Current Opinion in Psychiatry 2007;20(6):609-13

Briken P, Hill A, Berner W. Pharmacotherapy of paraphilias with long-acting agonists of luteinizing hormone-releasing hormone: a systematic review. Journal of Clinical Psychiatry 2003;64(8):890-7

Craig LA, Browne KD, Stringer I. Treatment and sexual offence recidivism. Trauma, Violence, & Abuse 2003;4(1):70-89

Crighton D, Towl G. Experimental interventions with sex offenders: a brief review of their efficacy. Evidence-Based Mental Health 2007;10(2):35-7

Day A. Interventions to improve empathy awareness in sexual and violent offenders: conceptual, empirical, and clinical issues. Aggression and Violent Behavior 2010;15(3):201-8

54 Treatment of Convicted Adult Male Sex Offenders Donato R, Shanahan M. The economics of implementing intensive in-prison sex-offender treatment programs. Trends and Issues in Crime and Criminal Justice 1999;134:1-6

Doren DM, Yates PM. Effectiveness of sex offender treatment for psychopathic sexual offenders. International Journal of Offender Therapy and Comparative Criminology 2008;52(2):234-45

Dowden C, Antonowicz D, Andrews DA. The effectiveness of relapse prevention with offenders: a meta-analysis. International Journal of Offender Therapy and Comparative Criminology 2003;47(5):516-28

Drapeau M. Research on the processes involved in treating sexual offenders. Sexual Abuse: Journal of Research and Treatment 2005;17(2):117-25

Fonza MA. A review of sex offender treatment programs. ABNF Journal 2001;12(2):42-8

Friendship C, Beech AR. Reconviction of sexual offenders in England and Wales: an overview of research. Journal of Sexual Aggression 2005;11(2):209-23

Gallagher CA, Wilson DB, Hischfield P. A quantitative review of the effects of sex offender treatment on sexual reoffending. Corrections Management Quarterly 1999;3(4):19-29

Grossman LS, Martis B, Fichtner CG. Are sex offenders treatable? A research overview. Psychiatric Services 1999;50(3):349-61

Heilbrun K, Nezu CM, Keeney M, Chung S, Wasserman AL. Sexual offending: linking assessment, intervention, and decision making. Psychology, Public Policy, and Law 1998;4(1-2):138-74

Hill A, Briken P, Kraus C, Strohm K, Berner W. Differential pharmacological treatment of paraphilias and sex offenders. International Journal of Offender Therapy & Comparative Criminology 2003;47(4):407-21

Hollin CR. Treatment programs for offenders - meta-analysis, “what works”, and beyond. International Journal of Law & Psychiatry 1999;22(3-4):361-72

Kirsch LG, Becker JV. Sexual offending: Theory of problem, theory of change, and implications for treatment effectiveness. Aggression and Violent Behavior 2006;11(3):208-24

Lehne G. Treatment of sexual paraphilias: a review of the 1999-2000 literature. Current Opinion in Psychiatry 2000;13(6):569-73

Levenson JS, Macgowan MJ, Morin JW, Cotter LP. Perceptions of sex offenders about treatment: satisfaction and engagement in group therapy. Sexual Abuse: Journal of Research & Treatment 2009;21(1):35-56

Maletzky BM, Field G. The biological treatment of dangerous sexual offenders: a review and preliminary report of the Oregon pilot Depo-Provera program. Aggression & Violent Behavior 2003;8(4):391

Mann RE. Innovations in sex offender treatment. Journal of Sexual Aggression 2004;10(2):141-52

Marshall WL. Sexual offender treatment: a positive approach. Psychiatric Clinics of North America 2008;31(4):681-96

Marshall WL, Fernandez YM, Serran GA, Mulloy R, Thornton D, Mann RE, et al. Process variables in the treatment of sexual offenders: a review of the relevant literature. Aggression & Violent Behavior 2003;8(2):205

Moster A, Wnuk DW, Jeglic EL. Cognitive behavioral therapy interventions with sex offenders. Journal of Correctional Health Care 2008;14(2):109-21

Murphy WD, McGrath RJ, Christopher MG. Evidence for the development of the ATSA practice standards and guidelines for the treatment of adult male sexual abusers. Journal of Forensic Psychology Practice 2008;8(1):77-88

Treatment of Convicted Adult Male Sex Offenders 55 Nicholaichuk TP, Yates PM, Schwartz BK. Treatment efficacy: outcomes for the Clearwater sex offender program - the sex offender: current treatment modalities and systems issues. Volume 4. Kingston NJ: Civil Research Institute; 2002

Palermo GB. Young offenders and recidivism. International Journal of Offender Therapy and Comparative Criminology 2008;52(1):3-4

Pray RT. Sex offender therapy outcome: a meta-analysis. Dissertation Abstracts International: Section B: The Sciences and Engineering 2003;63(10-B)

Rice ME, Harris GT. The size and sign of treatment effects in sex offender therapy. Annals of New York Academy of Sciences 2003;989:428-40

Saleh FM, Berlin FS. Sex hormones, neurotransmitters, and psychopharmacological treatments in men with paraphilic disorders. Journal of Child Sexual Abuse 2003;12(3-4):233-53

Shingler J. Managing intrusive risky thoughts: what works? Journal of Sexual Aggression 2009;15(1):39-53

Stalans LJ. Adult sex offenders on community supervision: a review of recent assessment strategies and treatment. Criminal Justice & Behavior 2004;31(5):564

The American Psychiatric Association Task Force on Sexually Dangerous Offenders. Dangerous sex offenders: a task force report of the American Psychiatric Association. Washington, DC: American Psychiatric Association, 1999: 194

Ward T, Gannon T, Yates PM. The treatment of offenders: current practice and new developments with an emphasis on sex offenders. International Review of Victimology 2008;15(2):179-204

Weiss P. Assessment and treatment of sex offenders in the Czech Republic and in Eastern Europe. Journal of Interpersonal Violence 1999;14(4):411-21

Main reason for exclusion: the study did not report on the effectiveness of sex offender treatment programs (n = 4)

Hanson RK, Bussiere MT. Predicting relapse: a meta-analysis of sexual offender recidivism studies. Journal of Consulting and Clinical Psychology 1998;66(2):348-62

Hanson RK, Morton-Bourgon KE. The characteristics of persistent sexual offenders: a meta-analysis of recidivism studies. Journal of Consulting and Clinical Psychology 2005;73(6):1154-63

Marshall WL, McGuire J. Effect sizes in the treatment of sexual offenders. International Journal of Offender Therapy & Comparative Criminology 2003;47(6):653-63

Prendergast WE. Treating Sex Offenders: A Guide to Clinical Practice with Adults, Clerics, Children, and Adolescents, Second Edition. United States:2004. Available from: http://www.HaworthPress.com

Main reason for exclusion: The study did not contain original research (n = 4)

Abracen J, Looman J, Mailloux D, Serin R, Malcolm B. Clarification regarding Marshall and Yates’s critique of dosage of treatment to sexual offenders: are we overprescribing? International Journal of Offender Therapy and Comparative Criminology 2005;49(2):225-30

Brooks-Gordon B, Bilby C. Psychological interventions for treatment of adult sex offenders. BMJ 2006;333(7557):5-6

Schneider JE. A review of research findings related to the civil commitment of sex offenders. Journal of Psychiatry and Law 2008;36(3):463-83

Zourkova A. Use of lithium and depot neuroleptics in the treatment of paraphilias. Journal of Sex and Marital Therapy 2000;26(4):359-60

56 Treatment of Convicted Adult Male Sex Offenders Main reason for exclusion: The study is not published in English language (n = 4)

Brichcin S. H. Use of Androcur in ambulatory protective treatment of sexual delinquents. Ceska a Slovenska Psychiatrie 1998;94(1):28-33

Endrass J, Rossegger A, Noll T, Urbaniok F. Therapy effectiveness with violent and sex offenders. Psychiatrische Praxis 2008;35(1):8-14

Muzinic L. Therapy of sexual delinquency. Medicina 2005;41(4):329-33

Rudel A. Involuntary outpatient treatment of sexual offenders under court order. Journal fur Neurologie, Neurochirurgie und Psychiatrie 2009;10(4):34-40

Main reason for exclusion: The study was not published in full-text format (abstract only) (n = 3)

Fransblow J. Rational emotive behaviour therapy: effectiveness with sexual and violent offenders. International Journal of Psychology 2000;35(3-4):45

Schmucker M, Loesel F. Meta-analyzing sexual offender treatment efficacy: an integration of research syntheses and the effects of meta-analytic strategies. Paper presented at: American Society of Criminology Conference 2007;1

Schmucker M, Loesel F. Beyond Programs: Outcome Differences Between Cognitive-Behavioral Programs of Sexual Offender Treatment. Paper presented at American Society of Criminology Conference; 2006

Main reason for exclusion: The study did not provide outcome data (n = 3)

Adi Y, Ashcroft D, Browne K, Beech A, Fry-Smith A, Hyde C. Clinical effectiveness and cost-consequences of selective serotonin reuptake inhibitors in the treatment of sex offenders. Health Technology Assessment 2002;6(28):1-66

Bilby C. Psychological interventions for those who have sexually offended or are at risk of offending. Cochrane Database of Systematic Reviews 2008;(4)

Khan O. Pharmacological interventions for those who have sexually offended or are at risk of offending. Cochrane Database of Systematic Reviews 2009;(3)

Main reason for exclusion: The study did not report on convicted adult sex offenders (n = 2)

Maletsky BM, Field G. The biological treatment of dangerous sexual offenders, a review and preliminary report of the Oregon pilot “Depo-Provera” program. Aggression and Violent Behavior 2003;8(4):391-412

Wilson DB, Bouffard LA, Mackenzie DL. A quantitative review of structured, group-oriented cognitive- behavioural programs for offenders. Criminal Justice and Behaviour 2005;32:-172

Main reason for exclusion: The full text of the study was not retrieved (n = 1)

Schaffer M. Cognitive-behavioral therapy in the treatment and management of sex offenders. Journal of Cognitive Psychotherapy 2010;24(2):92-103

Treatment of Convicted Adult Male Sex Offenders 57 Multiple Publications of Studies Included in the Overview From 11 included articles, two were identified as multiple publications; that is, cases in which the same study was published more than once or part of the data from an original report was republished. The multiple publications were not considered to be unique studies, and any information that they provided was included with the data reported in the main study. Table B2: Multiple publications

Multiple publications of studies included in the review (n = 2)

Losel F, Schmucker M. The effectiveness of treatment for sexual offenders: a comprehensive meta-analysis. Journal of Experimental Criminology 2005;1(1):117-46. 91 Associated publication of Schmucker M, Losel F. (2008)

Hanson RK, Public SC. A meta-analysis of the effectiveness of treatment for sexual offenders: risk, need, and responsivity. Ottawa, ON: Public Safety Canada; 2009. 92 Associated publication of Hanson et al. (2009)

58 Treatment of Convicted Adult Male Sex Offenders Appendix C: SELECTED SYSTEMATIC REVIEWS Abbreviations ATSA – Association for the Treatment of Abusers BMA – British Medical Association BNF – British National Formulary CBT – cognitive-behavioural therapy CI – confidence interval CODC – Collaborative Outcome Data Committee df – degrees of freedom ES – effect size h – hour(s) IATSO – International Association for the Treatment of Sexual Offenders MAU - Millhaven Assessment Unit NCJRS - National Criminal Justice Reference System NOTA – the National Organisation for the Treatment of Abusers NR – not reported/stated n.s. – not statistically significant

Treatment of Convicted Adult Male Sex Offenders 59 NZ – New Zealand ON – Ontario OR – odds ratio(s) PCL-R – Psychopathy Checklist- Revised Q – Cochrane’s Q statistic QOL – quality of life RCT – randomized controlled trial RPC – Regional Psychiatric Centre RTC – Regional Treatment Centre SK – Saskatchewan SOT – Sex Offender Treatment SOTP – sex offender treatment program UK – United Kingdom US – United States of America vs. – versus yr – year(s)

60 Treatment of Convicted Adult Male Sex Offenders y ’s main findings* 18 studies, evaluations of SOTPs for adult sex Selected studies: 18 studies, evaluations of SOTPs offenders (not clear if all participants in programs were males; what sex offences were committed by study participants; measures were used for recidivism; and the follow-up range). Ten studies were reported before 1998. Only 2 of the 18 selected studies utilized randomly assigned controlled groups (not clear how the remaining 16 studies were rated according to 5-point rigour scale; follow-up range not clearly stated). Recidivism: three studies (N= 313); on average, Psychotherapy/counselling: showed no significant reduction in recidivism compared to treatment as usual; fixed effect model weighted mean ES of 0.134 statistically significant heterogeneity in the ES across (P = 0.179; the studies at P = 0.05 (homogeneity test p=0.038); random effects weighted mean ES P = 0.892); of 0.027 CBT for sex offenders in prison: five studies (N = 894); on average, significantly reduced recidivism by 14.9% compared to treatment as usual; fixed-effect model weighted mean ES of -0.144 (P = 0.005); non-statistically significant heterogeneity in the ES sex across the studies at p=0.05 (homogeneity test P = 0.173); fixed-effect offence outcomes reported in four studies (N = 705); model weighted mean ES non-statistically of -0.119 (P = 0.027); significant heterogeneity in the ES across studies, at P = 0.05 (homogeneity test P = 0.080) CBT in the community for low-risk sex offenders on probation: six studies (N = 359 subjects); on average, significantly reduced recidivism by 31.2% compared to treatment as usual; fixed effects model weighted mean ES of -0.391 (P = 0.00); non-statistically significant heterogeneity in the ES across the studies at P = 0.05 (homogeneity test p=0.438); sex offence outcomes reported in five St ud t eris t ics y ’s charac Included studies: adult corrections outcome and evaluation studies published in English between 1970 2005, rated at least 3 on the 5-point rigour scale to (with a non-treatment control group well matched treatment group; random assignment and/or non-random assignment studies; quasi-experimental studies included if sufficient information provided to demonstrate comparability between treatment and comparison groups history) criminal prior and gender, age, as such variables on Excluded studies: studies rated 2 or 1 on the 5-point rigour scale (studies with a single-group, pre-post research treatment group design; comparison studies in which made up solely of program completers) adult offenders, including sex Participants: offenders (no details on subjects’ characteristics) Intervention: interventions provided within adult corrections programs Comparator(s): treatment as usual Outcome(s) and outcome measures: reduction in recidivism (not defined) rates St ud t ic reviews 88 s t ema y le C 1: Sy le St ud Aos et al., 2006 Objective: provide a comprehensive assessment of adult corrections programs and policies that have a proven ability to affect crime rates Methodology: independent literature databases of research searches engines (not specified) using search (no details on key words used); reference lists of individual studies and other systematic narrative reviews; methodological quality of in - cluded studies assessed with a 5-point rigour scale (based on Maryland scale); quantitative and qualitative analysis Country: US Support/funding: NR Conflict of interest: NR b Ta

Treatment of Convicted Adult Male Sex Offenders 61 studies (N = 262); fixed effects model weighted mean ES of -0.357 studies (N = 262); fixed effects model weighted mean ES of -0.357 (P = 0.001); non-statistically significant heterogeneity in the ES across studies, at P = 0.05 (homogeneity test 0.846) Behavioural therapy: two studies (N =c130); on average, showed no significant reduction in recidivism compared to treatment as usual; fixed effects model weighted mean ES of -0.190 (P = 0.126); non-statistically significant heterogeneity in the ES across studies, at P = 0.05 (homogeneity test 0.635) on Mixed-treatment in the community: two studies (N = 724); average, showed no significant reduction in recidivism compared to treatment as usual; fixed effects model weighted mean ES of -0.176 (P = 0.001); statistically significant heterogeneity in the ES across studies, at P = 0.05 (homogeneity test 0.015) Adverse events: no reporting on adverse events elements: no detailsNOTE on SOTPs’ were provided on treatment concept, duration of program, selection criteria for participation in the program, assessment procedures and instru - ments, type of therapeutic sessions, frequency and duration therapeutic sessions, timing of treatment, and treatment providers

62 Treatment of Convicted Adult Male Sex Offenders Selected studies: 21 quasi-experimental studies (no details on study design); 12 studies were reported before 1998; (follow-up range not clearly stated) Recidivism: in 7 of 21 studies treatment had a statistically significant effect on treatment group in terms of recidivism (not clear were the interventions and/or how recidivism is defined and which comparators in these studies); data from 4 of 21 studies were not or they reported their results in a more qualitative useable or clear, manner; in 10 of 21 studies treatment had no statistically significant effect on offenders’ behaviour or attitudes (not clear how recidivism were the interventions and/or comparators was defined and which in these studies) events: no reporting on adverse events Adverse elements: no detailsNOTE on SOTPs’ were provided on treatment concept, setting, duration of program, assessment procedures and instruments, treatment offered, type of therapeutic sessions, frequency and duration of therapeutic timing treatment, and treatment providers controlled trials, matched or non- Included studies: controlled trials, matched (studies where allocation to an experimentalmatched group was carried out due to, for example, sentencing or non- decisions, and a control group, either matched was evident) matched, Excluded studies: not clearly stated adults ( ≥ 18 yr) treated in institutional Participants: (criminal justice or mental health care facility) or community resulted in conviction settings for sexual behaviours which or caution for sexual offences (incest, child prostitution, child molestation, pornography, rape, child sex tourism, indecent exposure) or offences with a sexual element or violent behaviours with a sexual element, and adults perceived to be at risk of offending (including those who had sought treatment voluntarily for behaviours which would be classified as illegal) interventions for adult sex Intervention: psychological offenders and those showing abusive sexual behaviours (behavioural, cognitive behavioural, psychodynamic) Comparator(s): (10) any drug (as defined in BNF; BMA, 2003) treatment administered outside of standard care specifically for sexual behaviour impulses and/or standard care (defined as the a person would normally trial; receive had they not been included in the research it incorporated “waiting list control groups”); other broad therapies class of psychological Outcome(s) and outcome measures: (10) recidivism (not defined) measured by any offence, time before reoffence, or nonsexual violent offence; adverse effects; death; global state; behaviour; mental state; engagement with services; prison and service outcomes; satisfaction with treatment; acceptance of treatment; leaving the study early; QOL; economic outcomes 89 Bilby et al., 2006 Objective: to examine quasi-experimental and qualitative into psychological research interventions for adult sex offenders and individuals showing abusive sexual behaviours Methodology: electronic searches Central Register of Controlled (Cochrane Groups Schizophrenia Cochrane Trials, Related to Forensic Register of Trials Mental Health Services, MEDLINE, NCJRS, CINAHL, PsycINFO, EMBASE, etc) and hand for quasi-experimental searches research (1861 to 2003) using terms relating and sexual offending, abusive behaviour, treatment programs; hand psychological for articles was carried out and searching requests for additional information were made to academics, practitioners, and members of appropriate professional organizations; the first author of included studies was contacted for information regarding unpublished data; and study selection not literature search limited by publication language; qualitative analysis (not clear information on quality assessment of included studies) Country: UK Support/funding: NR Conflict of interest: NR

Treatment of Convicted Adult Male Sex Offenders 63 nine RCTs, all reported before 1998 (all Selected studies: nine RCTs, did category B studies, with moderate risk of bias); 7 9 RCTs not include recidivism among outcomes of interest, 1 RCT reported and 1 RCT reported on re-arrest rates by on reconviction by 1 yr, conducted in hospitals; 1 RCT conducted in 10 yr); 8 of 9 RCTs the community (no further details) is the only RCT that included convicted adult male sex offenders and reported data on recidivism over ≥ 2 yr follow-up period Recidivism: data from 1 large long-term RCT (N = 231 convicted adult male sex offenders followed up for 10 yr) showed that the rearrest rate for group therapy plus probation (1 hour of therapy per week for 40 weeks plus 1 probation visit per month) was not statistically significantly increased at 10 yr (14%) when compared to standard care (one report to probation per month plus 1 home visit per month) (7%) events: no reporting on adverse events Adverse NOTE on SOTP’s elements: no details were provided on treatment concept, selection criteria for participation in the program, assessment procedures and instruments, timing of treatment, treatment providers RCTs (category A: explicit good Included studies: RCTs allocation concealment; category B: no explicit concealment) Excluded studies: category C studies (allocation process not described/implied) Participants: adults ( ≥ 18 yr) treated in institutional (criminal justice or mental healthcare facility) or community settings have resulted in conviction for sexual behaviours which or caution for sexual offences with a element or violent behaviours with a sexual element, and adults perceived to be at risk of offending (including those who had sought treatment voluntarily for behaviours which would be classified as illegal) interventions for adult sex Intervention: psychological offenders and those showing abusive sexual behaviours (behavioural, cognitive behavioural, psychodynamic) Comparator(s): any drug (as defined in BNF; BMA, 2003) treatment administered outside of standard care specifically for sexual behaviour impulses and/or standard care (defined as a person would normally receive had trial; it incorporated they not been included in research ‘waiting list control groups’); other broad class of therapies psychological Outcome(s) and outcome measures: recidivism (measured by any offence, time before reoffence, or nonsexual violent offence); adverse effects; death; global state; behaviour; mental state; engagement with services; prison and service outcomes; satisfaction with treatment; acceptance of treatment; leaving the study early; QOL; economic outcomes 10 Brooks-Gordon et al., 2006 Objective: to examine experimental research interventions for evaluating psychological adult sex offenders and individuals showing abusive sexual behaviours Methodology: electronic searches Central Register of (including Cochrane Schizophrenia Cochrane Controlled Trials, Related to Forensic Groups Register of Trials Mental Health Services, MEDLINE, NCJRS, CINAHL, PsycINFO, EMBASE, etc) and for experimental hand searches research (1861 to 2003) using terms relating and sexual offending, abusive behaviour, treatment program; requests psychological for additional data to academics, researchers, practitioners, and clinicians involved in SOT, registered with national and international the first or NOTA; ATSA, arms of IATSO, included study was contactedauthor of each for information regarding unpublished data; and study selection not literature search limited by publication language; retrieved trials allocated to three quality categories (A, B, C) by considering concealment of allocation as key aspect of methodology; qualitative analysis (not clear information on quality assessment of included studies Country: UK Support/funding: NR Conflict of interest: NR *Only main findings regarding the population, intervention(s) and outcome of interest are summarized *Only main findings regarding the population, intervention(s) and outcome of interest are Review (Issue 3, 2004)Note: A section of this study was published electronically as a Cochrane with the support of NHS and Development Programme Research

64 Treatment of Convicted Adult Male Sex Offenders y ’s main findings* 68 recidivism outcome studies (total of 5078 Selected studies: 68 recidivism outcome studies (total of 5078 untreated sex offenders; 38 of 68 treated sex offenders and 4376 studies reported before 1998) reporting results from 43 treat - ment programs (16 from Canada, 21 US, five from UK, one from NZ); most studies focused on adult male sex offenders (four studies examined adolescents; one study indicated <5% of total sample were females); 43 programs examined (23 offered in three in both; 40 of 43 were special - in community, institutions, 17 ized treatment programs for sex offenders); mean follow-up period ranged from 1 to 16 yr (median of 46 months for both treatment and comparison groups). Sexual recidivism: examined in 38 studies (N = 8164); across designs, there was a small all types of treatment and research advantage for treated vs. untreated offenders (OR = 0.81; 95% CI 0.71 to 0.94), with statistically significant heterogeneity across p < 0.001); “current” treatments studies (Q = 145.02; df 37; ) appeared to be effective (OR = for adults (12 studies, N = 2779 0.61; 95% CI with statistically 0.48 significant heterogene - to 0.76), df = 11; P < 0.05); ES of “current” ity across studies (Q = 21.17; was OR = institutional treatments for adults (six studies, N = 1771) 0.62; 95% (CI 0.48 to 0.80) with statistically significant heteroge - neity across studies (Q = 12.31; df 5; P < 0.05); ES of “current” community treatments for adults (six studies, N = 1008) was OR = 95% CI 0.340.57; to 0.95), with non-statistically significant hetero - df = 5; P > 0.05) geneity across studies (Q = 8.78; General (any) recidivism: examined in 31 studies (n = 6075); designs, treated across all types of treatments and research offenders had significantly lower general recidivism rates than untreated offenders (OR = 0.56; 95% CI 0.50 to 0.64), with statistically St ud t eris t ics y ’s charac Included studies: studies that compared recidivism rates (sexual or general) of treated sex offenders with a comparison group of sex offenders, using the same recidivism criteria for both groups, and reporting the samerates for approximately follow-up period for both groups; combined sample of at least 10 offenders (5 in group) each Excluded studies: NR (not clearly stated exclusion criteria) sex offenders (adults and adolescents; Participants: males and females) treatment Intervention: psychological Comparator(s): no treatment, alternate/alternative treatment, a form of treatment judged/considered to be inadequate or inappropriate Outcome(s) and outcome measures: sexual and/or general (any) recidivism (no other details provided) St ud t ic reviews (con t ’d) 12 s t ema y le C 1: Sy le St ud Hanson et al., 2002 Objective: to examine the effectiveness of treatment for sex offenders psychological of Methodology: computer searches PsycINFO and NCJRS up to May 2000 using key terms and words relating to sexual exhibition - molester, offending (rape, child ism, paraphilia, etc), treatment (program, intervention) and outcome (recidivism, failure, etc) examination of reference lists collected articles; request for additional or unpublished data/research from 30 established- research for studies published ers in the field; searched selected studies were in English and French; rated based on study design by using CODC manual; qualitative and quantitative analysis Country: Canada, US Support/funding: NR Conflict of interest: NR b Ta

Treatment of Convicted Adult Male Sex Offenders 65 significant heterogeneity across studies (Q = 120.08; df = 30; P < 0.001); “current” treatments or adults (five studies, N = 1101) appeared to be effective (OR = 0.59; with 95% CI 0.45 to 0.78), statistically significant heterogeneity across studies (Q = 33.00; df = 4; P < 0.001); ES of “current” community treatments for adults (two studies, N = 330) was OR with = 0.21 (95% CI 0.12 to 0.37) non-statistically significant heterogeneity (Q = 0.01; df 1; ns); ES of “current” institutional treatments for adults (examined in three was OR = 0.82; 95% CIstudies, N = 771) 0.60 to 1.13), with statistically significant heterogeneity across studies (Q = 15.76; df = 2; P < 0.001) events: no reporting on adverse events Adverse that included adults, NOTE on SOTP’s elements: for the SOTPs no details were provided on treatment concept, duration of pro - gram, assessment procedures and instruments, type of therapeutic sessions, frequency and duration of therapeutic timing treatment, and treatment providers * Only main findings regarding the population, intervention(s) and outcome of interest are summarized * Only main findings regarding the population, intervention(s) and outcome of interest

66 Treatment of Convicted Adult Male Sex Offenders y ’s main findings* Selected studies: 23 recidivism outcome studies (7 of reported before 1998; five as “good” 18 rated as “weak” quality, quality): 12 from Canada, five US, three from UK, two from NZ, one from Holland; most studies (19 of 23) focused on adult examinedmale sex offenders; 23 SOTPs (10 offered in institutions, two in both); most studies (19 of 23) examined 11 in community, specialized treatment programs for sex offenders; average follow-up ranged from 1 to 21 yr (median of 4.7 yr) offenders, Sexual recidivism: examined in 22 studies (N = 6746 of 22 studies, recidivism rate 3121 were treated); in 17 of which of treatment group was lower than that comparison (P = with a fixed-effect weighted 0.0085); OR ranged from 0.08 to 2.47, (95% CI 0.65 to 0.91), significant between-study mean of 0.77 df = 21, P < 0.001), and a random-effect variability (Q = 47.17, weighted mean of 0.66 (95% CI 0.49 to 0.89); according to both models, treatment appeared to be equally effective for adults and was delivered adolescents, and did not depend on whether SOTP in community or institution; a pooled analysis of 18 studies that included only adults (N = 6462) yielded fixed-effect weighted mean to 0.94) (95% CI and a random-effect weighted mean 0.67 of 0.79 of 0.71 (95% CI 0.53 to 0.95) (including sexual) recidivism: examinedViolent in 10 studies (N = 4823 2021 were treated); in 6 of 10 offenders, of which studies, recidivism rate for treatment group was lower than that of one-tailed); OR ranged from 0.04 comparison group (P = 0.377, to 1.34 to with a fixed-effect weighted mean of 0.92 (95% CI 0.78 statistically significant heterogeneity across the studies (Q 1.07), = 26.63, df 9, P < 0.005), and random-effect weighted mean of 0.81 (95% CI 0.58 to 1.14); according both models, there were no differences in treatment effects according to whether St ud t eris t ics y ’s charac Included studies: studies that compared treated sex offenders with a comparison group of sex offenders, which met a minimum level of study quality Excluded studies: studies involving illegal but consensual the only difference between sexual crimes; studies in which groups was the form of supervision; studies that included recidivism rates for different groups (including a treatment group) but did not intend to test treatment effectiveness sex offenders defined as with Participants: sexually motivated offences against an identifiable victim; adults or adolescents interventions Intervention(s): psychological Comparator(s): no treatment, an alternate less treatment, or established norms Outcome(s) and outcome measures: recidivism (not defined) rates (sexual, violent, or general) St ud t ic reviews (con t ’d) 13,92 s t ema y le C 1: Sy le St ud Hanson et al., 2009 Objective: to examine if principles of effec - tive interventions for general offenders also effective - and to assess SOT apply to SOT ness using only studies with a minimum study quality level (up to May Methodology: computer search 2008) of PsycINFO, of Science, Digital Web Dissertations, and NCJRS using key words molester, relating to sexual offence (rape, child exhibitionism, etc), treatment (program, inter - vention) and outcome (recidivism, failure, etc); reference lists of collected articles examined; reviewed studies published in English, French and German; quality of studies assessed using CODC guidelines; qualitative and quantitative analysis; Country: Canada Support/funding: NR Conflict of interest: NR b Ta

Treatment of Convicted Adult Male Sex Offenders 67 studies involved adult or adolescent offenders, were delivered in community or institution; a pooled analysis of eight studies that yielded a fixed-effect weighted included only adults (N = 4718) to 1.09)mean of 0.93 (95% CI and a random-effect weighted 0.79 mean of 0.86 (95% C 0.62 to 1.20) General (any) recidivism): examined in 13 studies (N = 4801 were treated); in 12 of 13 studies, recidi - 1979 offenders, of which one tailed); OR vism rate favoured treatment group (P = 0.0017, to 1.14 with a fixed-effect mean of 0.75 (95% CI ranged from 0.07 0.66 to 0.86), statistically significant heterogeneity across studies (Q = 29.82, df 12, P < 0.005), and random-effect weighted mean to 0.80);of 0.61 (95% CI according to both models treatment 0.47 appeared to be more effective for adolescent offenders than adult offenders, and there were no differences in treatment effects according to whether treatment was delivered in community or institution; a pooled analysis of 10 studies that included only adults (N = 4606) (95% CI yielded a fixed-effect weighted mean of 0.79 0.69 to 0.90) and a random-effect weighted mean of 0.71 (95% CI 0.56 to 0.90) events: no reporting on adverse events Adverse that included elements: for the SOTPs NOTE on SOTPs’ adults, no details were provided on treatment concept, duration of program, selection criteria for participation, assessment procedures and instruments, type of interventions, therapeutic sessions, frequency and duration of therapeutic sessions, timing treatment, and treatment providers * Only main findings regarding the population, intervention(s) and outcome of interest are summarized * Only main findings regarding the population, intervention(s) and outcome of interest (con t ’d)

68 Treatment of Convicted Adult Male Sex Offenders y ’s main findings* Selected studies: 13 impact evaluation studies, all published before 1998 provided in prison-based (eight studies evaluated SOT two were rated as level 4, one study 3, and setting, of which SOT provided in non-prison- five as level 2; studies evaluated two were rated as level 4, 3 based setting, of which and one study as level 2); follow-up covered up to 31 yr for prison- (follow-up and up to 11 yr for community-based SOT based SOT range not clearly identified for all studies) Recidivism: 6 of 13 studies showed statistically significant findings 4 of 6 studies that showed a positive in favor of evaluated SOTPs; treatment effect, incorporated a cognitive-behavioural approach. using CBTNon-prison-based SOTPs methods were deemed to be effective in curtailing future criminal activity (sexual and nonsexual molest - recidivism) as at least two studies (one included 126 child ers, rated as level 4, and one included 105 exhibitionists, level 3) demonstrated a significant reduction in reconviction rates for treated sex offenders when compared to untreated controls respectively). Prison-based (moderate to large ESs: 0.51 and 0.70, were judged to be promising, but the evidence was not SOTPs programs are strong enough to support a conclusion that such effective as only 1 study (rated level 4) provided evidence that treated high-risk offenders had significantly fewer sex offence recon - victions than untreated sex offenders (moderate ES: 0.44 to 0.45). events: no reporting on adverse events Adverse elements: no detailsNOTE on SOTPs’ were provided on treat - ment concept, duration of program, selection criteria for participa - tion in the program, assessment procedures and instruments, treat - ment offered, type of therapeutic sessions, frequency and duration of therapeutic sessions, timing treatment, and treatment providers St ud t eris t ics y ’s charac Included studies: study of the impact SOT Excluded studies: process or descriptive evaluations; studies not including an outcome measure of recidivism; studies rated 1 on the 5-point rigour scale sex offenders (no other details;Participants: study popu - lation of interest not clearly stated/described; not clear considered) were offenders sex male adult convicted only if Intervention: prison- and non-prison-based SOT interventions (no specific mentioned) Comparator(s): NR (comparator of interest not clearly stated/described; no specific comparators mentioned) Outcome(s) and outcome measures: reduced recidivism (sexual offence or any kind of future criminal behaviour) St ud t ic reviews (con t ’d) 90 s t ema y le C 1: Sy le St ud et al., 1999Polizzi Objective: to evaluate prison- and non- prison-based SOTPs Methodology: library database searches (publication date is not clearly for research stated); area were individuals working in SOT contacted to identify any other studies that were not included in analysis; elements of design for internal validity of each research selected study examined with a 5-point rigour scale identical to the Maryland scale; although not clearly stated if publication language was used as a selection criterion, it appears study selection was restricted to studies in English Country: US Support/funding: supported in part by the State Legislature Joint Audit and Washington Review Committee Conflict of interest: NR * Only main findings regarding the population, intervention(s) and outcome of interest are summarized * Only main findings regarding the population, intervention(s) and outcome of interest b Ta

Treatment of Convicted Adult Male Sex Offenders 69 y ’s main findings* Selected studies: 80 comparisons from 66 published and unpublished reports of 69 studies (total 9512 were in treated of 22,181 sex offenders, which from Canada, 31 US, eight from UK,groups): 17 eight from German speak - of 66 reports published before 1998);ing countries, five from other countries (37 45 comparisons examined only adult sex offenders; seven comparisons examined only adolescents; 29 comparisons referred to outpatient treatment; 25 prison setting, 14 referred to hospital setting, and 10 comparisons referred to a mixture of outpatient and residential settings; programs addressing sex-offender- specific treatment evaluated in 64 comparisons; follow-up period ranged from 1 to 10 yrs, and averaged 5.22 yrs comparisons; overall, average treatment Sexual recidivism: evaluated in 74 (95% CI:effect was OR 1.35-2.13, random-effect model), with statisti = 1.70 - 73, P< 0.001); recidivism rate df = cally significant heterogeneity (Q = 237.14, lower than in the control group; programs that of treated offenders was 37% specifically addressed adult sex offenders, evaluated in 36 comparisons, had a significant ES (OR = 1.43, 95% CI 1.08 to 1.90) recidivism: evaluated in 20 comparisons; overall, average treatment Violent effect was OR = 1.90 (95% CI 1.49 to 2.33; random-effect model) , with non- statistically significant heterogeneity (Q = 19.68, df = 19, n.s.); recidivism rate of treated offenders was 44% lower than in the control group; no separate reporting on adult sex offenders General (any) recidivism): evaluated in 49 comparisons; overall, average treat - (95% CIment effect was OR 1.33 to 2.08; = 1.67 random-effect model) with statistically significant heterogeneity (Q = 159.80, df = 48, P < 0.001); recidivism rate of treated offenders was 31% lower than in the control group; no separate reporting on adult sex offenders events: no reporting on adverse events Adverse that included adults, no details elements: for the SOTPs NOTE on SOTPs’ were provided on duration of program, selection criteria for participation in the program, assessment procedures and instruments, frequency duration of therapeutic sessions, timing of treatment, and treatment providers St ud t eris t ics y ’s charac Included studies: controlled outcome evalua - (experimentaltions of SOT and quasi-experimental, and French, published in English, German, Dutch, Swedish), reported up to June 2003, that included a minimum of 10 participants and reported out - comes in a way permitting calculation of ES estimates Excluded studies: uncontrolled studies; studies using only treatment dropouts as a control group; studies examining purely deterrent or punishing ap - studies focusing exclusively on changes proaches; in measures of personality or hormone levels, clini - cal ratings of improvement, and the like participants had to have been Participants: convicted of a sexual offence or to have committed lead have would that behaviour sexual illegal of acts to a conviction if officially prosecuted and/or organic Intervention: psychosocial treatment modes (hormonal medication or surgical incorporated castration), which therapeutic measures Comparator(s): no treatment, treatments “as another kind of treatment that differed from usual”, the evaluated treatment in content, intensity and specificity Outcome(s) and outcome measures: reduced recidivist behaviour (outcomes ranging from incarceration to lapse behaviour) St ud 11,91 t ic reviews (con t ’d) s t ema y le C 1: Sy le St ud and Losel, 2008 Schmucker Objective: to systematically review controlled outcome evaluations of and biological SOT psychosocial Methodology: basic study pool compiled from reference sections of previous reviews; computer searches of 14 electronic databases (includ - ing PsycINFO, MEDLINE, ERIC, The Social Services Library, Cochrane Abstracts, NCJRS abstract and full-text databases, Dissertation Abstracts International, and UK National Health Service National Register); hand search of journal pertaining- to topic; research field contacteders in SOT to locate of In - unpublished evaluations; searches ternet sites of pertinent institutions and department of corrections; all studies reported up to June 2003 were eligible; methodological quality of included stud - ies evaluated with an adapted Maryland scale; qualitative and quantitative analy - sis; study was carried out within the framework of the Campbell Collabora - tion Crime and Justice Group Country: Germany Support/funding: NR Conflict of interest: NR * Only main findings regarding the study population, intervention(s), and outcome of interest are summarized * Only main findings regarding the study population, intervention(s), and outcome of interest b Ta

70 Treatment of Convicted Adult Male Sex Offenders y ’s main findings* three RCTs; two RCTs reported reconviction two RCTs Selected studies: three RCTs; and >1 yr; one RCT included convicted rates by 6 months, 1 yr, adult male sex offenders and reported recidivism data over ≥ 2 yr follow-up period Recidivism: data from one large long-term RCT (N = 231 con - victed adult male sex offenders followed up for 10 yr) showed no significant difference in rearrest rate for offenders allocated to group plus probation and those receiving standardpsychotherapy care 95% CI(probation) (OR 0.8 to 4.37) = 1.87; events: no reporting on adverse events Adverse NOTE on SOTP’s elements: no details were provided on selec - tion criteria for participation in the program, assessment procedures and instruments, timing of treatment, treatment providers St ud t eris t ics y ’s charac RCTs(category A: randomization was Included studies: RCTs(category described; category B: randomization was stated but not described) Excluded studies: category C studies (no mention of randomization) individuals who have been convicted of Participants: sexual offences or who have disorders of preference Intervention: anti-libidinal management techniques interventions) psychological (drugs, surgery, Comparator(s): placebo or standard care Outcome(s) and outcome measures: recidivism (defined as the occurrence of additional thoughts, urges or acts relevant to the disorders of sexual preference during the period of treatment); and people lost to follow-up; other outcomes: death (suicide, all causes); other forms of crimi - nal offence, measures of mental state, patient satisfaction, measures of resource utilization or penile plethysmography, cost benefit, and side effects St ud t ic reviews (con t ’d) 58 s t ema y le C 1: Sy le St ud et al., 1998White Objective: to determine the effectiveness of to assist a range of management techniques people who have disorders of sexual prefer - ence and those who have been convicted of sexual offences (1966Methodology: computer searches to April 1998) of Biological Abstracts, the Group Register Schizophrenia Cochrane EMBASE, Library, Cochrane The of Trials, MEDLINE, using key terms and PsychLIT relating to sexual offence; further references sought from published trials and their authors; relevant pharmaceutical manufacturers were contacted; retrieved trials allocated to three quality categories (A, B, C) by considering randomization as key aspect of methodology; qualitative and quantitative analysis; although not clearly stated if publication language was used as a selection criterion, it appears study selection was restricted to studies in English Country: UK and Australia Review) (Cochrane Support/funding: Internal sources (Mental Ox - Australia; Health, Queensland Unit, Health UK) ford University Department of Psychiatry, Conflict of interest: None * Only main findings regarding the population, intervention(s) and outcome of interest are summarized * Only main findings regarding the population, intervention(s) and outcome of interest b Ta

Treatment of Convicted Adult Male Sex Offenders 71 Table C2: Methodological quality appraisal using AMSTAR tool

Brook- Han- Han- Pol- Sch- Gor- son son izzi muck- White Review Aos et Bilby don et et et et er et et 88 89 10 12 13 90 11 58 characteristics al. et al. al. al. al. al. al. al.

1 Was an a priori Yes Yes Yes Yes Yes Yes Yes Yes design provided?

2 Was there duplicate No C/A Yes No Yes No No Yes study selection and data extraction?

3 Was a No Yes Yes Yes Yes No Yes Yes comprehensive literature search performed?

4 Was the status of Yes No C/A Yes Yes No Yes Yes publication (i.e., grey literature) used as an inclusion criterion?

5 Was a list of studies No No No No Yes No No Yes (included and excluded) provided?

6 Were the characteristics No No Yes Yes Yes Yes Yes Yes of the included studies provided?

7 Was the scientific Yes No Yes No Yes Yes Yes Yes quality of the included studies assessed and documented?

8 Was the scientific quality C/A No Yes No Yes Yes Yes Yes of the included studies used appropriately in formulating conclusions?

9 Were the methods used Yes N/A C/A Yes Yes N/A Yes N/A to combine the findings of studies appropriate?

10 Was the likelihood No No Yes No No No Yes Yes of publication bias assessed?

11 Was the conflict of No No No No Yes Yes No Yes interest stated?

TOTALS Yes 4 2 7 5 10 5 8 10

No 6 7 2 6 1 5 3 0

Cannot 1 1 2 0 0 0 0 0 answer

Not 0 1 0 0 0 1 0 1 applicable

C/A – cannot answer; N/A – not applicable

72 Treatment of Convicted Adult Male Sex Offenders AMSTAR Quality Assessment Tool104

1. Was an a priori design provided? ? Yes The research question and inclusion criteria should be established ? No before the conduct of the review. ? Cannot answer ? Not applicable

2. Was there duplicate study selection and data extraction? ? Yes There should be at least two independent data extractors, and a ? No consensus procedure for disagreements should be in place. ? Cannot answer ? Not applicable

3. Was a comprehensive literature search performed? ? Yes At least two electronic sources should be searched. The report ? No must include years and databases used (e.g., Central, EMBASE, ? Cannot answer and MEDLINE). Key words and/or MeSH terms must be stated, and where feasible, the search strategy should be provided. All searches ? Not applicable should be supplemented by consulting current contents, reviews, textbooks, specialized registers, or experts in the particular field of study, and by reviewing the references in the studies found.

4. Was the status of publication (i.e., grey literature) used as ? Yes an inclusion criterion? ? No The authors should state that they searched for reports regardless of ? Cannot answer their publication type. The authors should state whether they excluded any reports (from the systematic review) based on their publication ? Not applicable status, language, etc.

5. Was a list of studies (included and excluded) provided? ? Yes A list of included and excluded studies should be provided. ? No ? Cannot answer ? Not applicable

6. Were the characteristics of the included studies provided? ? Yes In an aggregated form such as a table, data from the original studies ? No should be provided on the participants, interventions, and outcomes. ? Cannot answer The ranges of characteristics in all the studies analyzed (e.g., age, race, sex, relevant socioeconomic data, disease status, duration, ? Not applicable severity, or other diseases) should be reported.

7. Was the scientific quality of the included studies assessed ? Yes and documented? ? No A priori methods of assessment should be provided (e.g., for ? Cannot answer effectiveness studies if the author(s) chose to include only randomized, double-blind, placebo-controlled studies, or allocation ? Not applicable concealment as inclusion criteria); for other types of studies alternative items will be relevant.

8. Was the scientific quality of the included studies used ? Yes appropriately in formulating conclusions? ? No The results of the methodological rigor and scientific quality should ? Cannot answer be considered in the analysis and the conclusions of the review, and explicitly stated in formulating recommendations. ? Not applicable

Treatment of Convicted Adult Male Sex Offenders 73 9. Were the methods used to combine the findings of studies ? Yes appropriate? ? No For the pooled results, a test should be done to ensure that the ? Cannot answer studies were combinable, to assess their homogeneity (i.e., Chi test for homogeneity, I²). If heterogeneity exists, a random effects model ? Not applicable should be used and/or the clinical appropriateness of combining should be taken into consideration (i.e., is it sensible to combine?).

10. Was the likelihood of publication bias assessed? ? Yes An assessment of publication bias should include a combination of ? No graphical aids (e.g., funnel plot, other available tests) and/or statistical ? Cannot answer tests (e.g., Egger regression test). ? Not applicable

11. Was the conflict of interest stated? ? Yes Potential sources of support should be clearly acknowledged in the ? No systematic review. ? Can’t answer ? Not applicable

AMSTAR Guidelines adapted by PC and MO 1. Was an a priori design provided? Yes: if the research question and the inclusion criteria are clearly stated in the Abstract, Introduction, or Methods sections of the review No: no statement on question or inclusion criteria Cannot answer: the research question and the inclusion criteria are vague/ unclear, or they are stated/described in other sections of the review 2. Was there duplicate study selection and data extraction? Yes: two reviewers for selection and extraction and a consensus procedure No: at least one of the above is a “no” (e.g., one reviewer for selection, two for extraction, and a consensus procedure in place) Cannot answer: if at least one of the above is not mentioned 3. Was a comprehensive literature search performed? Yes: all four elements are there (two electronic sources, years and databases, key words or MeSH terms, additional sources) No: if any are missing 4. Was the status of publication (e.g., grey literature) used as an inclusion criterion? Yes: clear statement about publication type and language No: no statement on publication type or language Cannot answer: statement is unclear

74 Treatment of Convicted Adult Male Sex Offenders 5. Was a list of studies (included and excluded) provided? Yes: both included and excluded are presented (tables or lists), or only included studies are presented but it is mentioned that a list of excluded studies is available on request No: if one or both are not listed 6. Were the characteristics of the included studies provided? Yes: tables of included studies with all three elements (participants, interventions, and outcomes) No: no tables with information on these elements Cannot answer: if only some elements 7. Was the scientific quality of the included studies assessed and documented? Yes: if tool or checklist/tool for formal critical appraisal is mentioned/used, and critical appraisal is documented in tables or text No: no mention of a tool/checklist, or critical appraisal not documented in tables or text Cannot answer: tool/checklist was used to categorize/rate studies according to level of evidence 8. Was the scientific quality of the included studies used appropriately in formulating conclusions? Yes: conclusions make reference to quality of evidence No: no reference to quality of evidence Cannot answer: studies mentioned only by level of evidence 9. Were the methods used to combine the findings of studies appropriate? Yes: for quantitative analysis, tests for homogeneity/heterogeneity must be done No: no test for homogeneity/heterogeneity done, or not mentioned Cannot answer: test was done, but results not mentioned Not applicable: qualitative analysis

Treatment of Convicted Adult Male Sex Offenders 75 10. Was the likelihood of publication bias assessed? Yes: if anything is mentioned on publication bias (graphical aids not required, but a statement is required) No: no statement on publication bias 11. Was the conflict of interest stated? Yes: sources of support acknowledged for the review No: if source of support is not mentioned REFERENCES 1. Lane Council of Governments. Managing sex offenders in the community: a national overview. Eugene OR: Lane Council of Governments; 2001. 2. Center for Sex Offender Management. Fact sheet: what you need to know about sex offenders. Center for Sex Offender Management 2008. 3. Adi Y, Ashcroft D, Browne K, Beech A, Fry-Smith A, Hyde C. Clinical effectiveness and cost-consequences of selective serotonin reuptake inhibitors in the treatment of sex offenders. Health Technology Assessment 2002;6(28):1-66. 4. Association for the Treatment of Sexual Abusers. Facts about adult sex offenders. Beaverton, OR: ATSA; 2005. 5. Gordon H. Psychiatric aspects of the assessment and treatment of sex offenders. Advances in Psychiatric Treatment 2004;10(1):73-80. 6. Perkins D, Hammond S, Coles D, Bishopp D. Review of sex offender treatment programmes. London: High Security Psychiatric Services Commissioning Board; 1998. 7. Gelb K. Recidivism of sex offenders: research paper. Victoria, Australia: Sentencing Advisory Council; 2007. 8. Bijleveld C. Sex Offenders and Sex Offending. Crime & Justice: A Review of Research 2007;35:319-87. 9. John Howard Society of Alberta. Sex offender treatment programs. Edmonton, AB: John Howard Society of Alberta; 2002. 10. Brooks-Gordon B, Bilby C, Wells H. A systematic review of psychological interventions for sexual offenders I: Randomised control trials. Journal of Forensic Psychiatry & Psychology 2006;17(3):442-66. 11. Schmucker M, Losel F. Does sexual offender treatment work? A systematic review of outcome evaluations. Psicothema 2008;20(1):10-9. 12. Hanson RK, Gordon A, Harris AJ, Marques JK, Murphy W, Quinsey VL, et al. First report of the collaborative outcome data project on the

76 Treatment of Convicted Adult Male Sex Offenders effectiveness of psychological treatment for sex offenders. Sexual Abuse: Journal of Research & Treatment 2002;14(2):169-94. 13. Hanson RK, Bourgon G, Helmus L, Hodgson S. The principles of effective correctional treatment also apply to sexual offenders: A meta-analysis. Criminal Justice & Behavior 2009;36(9):891. 14. Grubin D. Second expert paper: sex offender research. Liverpool, UK: Forensic Mental Health; 2007.

15. Wolak J, Finkelhor D, Mitchell KJ, Ybarra M. Online “predators” and their victims: myths, realities, and implications for prevention and treatment. American Psychology 2008;63(2):111-28. 16. Webb L, Craissati J, Keen S. Characteristics of Internet child pornography offenders: a comparison with child molesters. Sexual Abuse: A Journal of Research & Treatment 2007;19(4):449-65. 17. Robertiello G, Terry KJ. Can we profile sex offenders? A review of sex offender typologies. Aggression and Violent Behavior 2007;12(5):508-18. 18. Davidson J. Current practice and research into internet sex offending. Paisley, UK: Risk Management Authority Research; 2007. 19. Lievore D. Recidivism of sexual assault offenders: Rates, risk factors, and treatment efficacy. Canberra, Australia: Australian Institute of Criminology; 2003. 20. Fortney T, Levenson J, Brannon Y, Baker JN. Myths and facts about sexual offenders: implications for treatment and public policy. Sexual Offender Treatment 2007;2(1):1-25. 21. Grubin D. Sexual offending and the treatment of sex offenders. Psychiatry 2007;6(11):439-43. 22. Fortney T, Baker JN, Levenson J. A look in the mirror: Sexual abuse professionals’ perceptions about sex offenders. Victims & Offenders 2009;4(1):42-57. 23. Studer LH, Aylwin AS. Pedophilia: the problem with diagnosis and limitations of CBT in treatment. Medical Hypotheses 2006;67(4):774-81. 24. Studer LH, Reddon JR. Treatment may change risk prediction for sexual offenders. Sexual Abuse: Journal of Research and Treatment 1998;10(3):175-81. 25. Durkin KF, Digianantonio AL. Recidivism among child molesters: a brief overview. Journal of Offender Rehabilitation 2007;45(1):249-56. 26. Barbaree HE, Blanchard R. Sexual deviance over the lifespan: reductions in deviant sexual behavior in the aging sex offender. In: Laws DR, O’Donohue WT, editors. Sexual Deviance: Theory, Assessment, and Treatment. New York: Guilford Press; 2008. p. 37-60.

Treatment of Convicted Adult Male Sex Offenders 77 27. Hanson RK. Recidivism and age: Follow-up data from 4,673 sexual offenders. Journal of Interpersonal Violence 2002;17(10):1046-62. 28. Whitaker DJ, Le B, Hanson RK, Baker CK, McMahon PM, Ryan G, et al. Risk factors for the perpetration of child sexual abuse: a review and meta-analysis. Child Abuse & Neglect 2008;32(5):529-48. 29. Stalans LJ. Adult sex offenders on community supervision: a review of recent assessment strategies and treatment. Criminal Justice & Behavior 2004;31(5):564-608. 30. Engle MJ, McFalls JAJ, Gallagher BJI. The attitudes of members of the association for the treatment of sexual abusers towards treatment, release, and recidivism of violent sex offenders: An exploratory study. Journal of Offender Rehabilitation 2007;44(4):17-24. 31. Statistics Canada. Sex offenders: integration and analysis program. Juristat 1999;19(3):Statistics Canada Catalogue no. 85-002-XIE-Ottawa, ON. 32. Bourget D, Bradford JMW. Evidential basis for the assessment and treatment of sex offenders. Brief Treatment and Crisis Intervention 2008;8(1):130-46. 33. The National Child Exploitation Coordination Centre (NCECC). Internet based sexual exploitation of children and youth environmental scan. Ottawa, ON: Royal Canadian Mounted Police; 2005. 34. Middleton D, Mandeville-Norden R, Hayes E. Does treatment work with internet sex offenders? Emerging findings from the Internet Sex Offender Treatment Programme (i-SOTP). Journal of Sexual Aggression 2009;15(1):5-19. 35. Bradford JMW, Fedoroff P, Firestone P. Sexual violence and the clinician. In: Simon RIE, Tardiff KE, editors. Textbook of Violence Assessment and Management. American Psychiatric Publishing, Inc.; 2008. p. 441-60. 36. Endrass J, Urbaniok F, Hammermeister LC, Benz C, Elbert T, Laubacher A. The consumption of Internet child pornography and violent and sex offending. BMC Psychiatry 2009;9(43). 37. Werner J, Werner MC. Child sexual abuse in clinical and forensic psychiatry: a review of recent literature. Current Opinion in Psychiatry 2008;21(5):499-504. 38. Hanson RK, Morton-Bourgon KE. The accuracy of recidivism risk assessments for sexual offenders: a meta-analysis of 118 prediction studies. Psychological Assessment 2009;21(1):1-21. 39. Harris AJR, Hanson K. Sex offender recidivism: a simple question. Ottawa, ON: Public Safety and Emergency Preparedness Canada; 2004. 40. Codispoti VL. Pharmacology of sexually compulsive behavior.

78 Treatment of Convicted Adult Male Sex Offenders Psychiatry Clinics of North America 2008;31(4):671-9. 41. Studer LH, Aylwin AS. Sexual offender subtyping: the incest offender question. Sexual Offender Treatment 2006;1(2):1-8. 42. Hall RC, Hall RC. A profile of pedophilia: definition, characteristics of offenders, recidivism, treatment outcomes, and forensic issues. Mayo Clinic Proceedings 2007;82(4):457-71. 43. Abracen J, Looman J. Developments in the assessment and treatment of sexual offenders: looking backward with a view to the future. Journal of Interpersonal Violence 2005;20(1):12-9. 44. Seto MC. Pedophilia. Annual Review of Clinical Psychology 2009;5:391-407. 45. Prentky R, Schwartz B. Treatment of adult sex offenders. Treatment of adult sex offenders, 2006. Available at: http://wwsearch.ebscohost.com/login.aspx. 46. Hanson RK, Morton-Bourgon KE. Predictors of sexual recidivism: an updated meta-analysis. Public Safety Canada; 2004. 47. Freeman-Longo R, Reback D. Myths and facts about sex offenders, 2000. Available at: http://www.csom.org/pubs/mythsfacts.pdf. 48. Bradford JMW, Kaye NS. The pharmacological treatment of sexual offenders. American Academy of Psychiatry and Law Newsletter 1999;24:16-7. 49. Briken P. Pharmacological treatments for paraphilic patients and sexual offenders. Current Opinion in Psychiatry 2007;20(6):609-13. 50. McGrath RJ, Cumming GF, Burchard BL, Zeoli S, Ellerby L. Current practices and emerging trends in sexual abuse management: The Safer Society 2009 North American Survey. Available at: http://www safersociety.org 2010. 51. Center for Sex Offender Management. Understanding treatment for adults and juveniles who have committed sex offenses. Washington, DC: US Department of Justice, Office of Justice Programs; 2006. 52. Andrews DA. The risk-need-responsivity model of assessment and human service in prevention and corrections: crime-prevention jurisprudence. (Canada) (Sentencing and Risk Assessment). Canadian Journal of Criminology and Criminal Justice 2007;49(4). 53. Wormith JS, Althouse R, Simpson M, Reitzel LR, Fagan TJ, Morgan RD. The rehabilitation and reintegration of offenders: The current landscape and some future directions for correctional psychology. Criminal Justice & Behavior 2007;34(7):879-92. 54. Moster A, Wnuk DW, Jeglic EL. Cognitive behavioral therapy interventions with sex offenders.

Treatment of Convicted Adult Male Sex Offenders 79 Journal of Correctional Health Care 2008;14(2):109-21. 55. Witt PH. Cognitive/behavioral approaches to the treatment adult sex offenders. Journal of Psychiatry and Law, 2008;36(2):245-69. 56. Hughes JR. Review of medical reports on pedophilia. Clinical Pediatrics 2007;46(8):667-82. 57. Grossman LS, Martis B, Fichtner CG. Are sex offenders treatable? A research overview. Psychiatric Services 1999;50(3):349-61. 58. White P, Bradley C, Ferriter M, Hatzipetrou L. Managements for people with disorders of sexual preference and for convicted sexual offenders. Cochrane Database of Systematic Reviews 2000;(2):000251. 59. Yates PM, Kingston DA. The self-regulation model of sexual offending: the relationship between offence pathways and static and dynamic sexual offence risk. Sexual Abuse: Journal of Research and Treatment 2006;18(3):259-70. 60. Seto MC, Marques JK, Harris GT, Chaffin M, Lalumiere ML, Miner MH, et al. Good science and progress in sex offender treatment are intertwined: A response to Marshall and Marshall (2007). Sexual Abuse: Journal of Research and Treatment 2008;20(3):247-55. 61. Looman J, Dickie I, Abracen J. Responsivity issues in the treatment of sexual offenders. Trauma, Violence, & Abuse 2005;6(4):330-53. 62. Ward T, Connolly M. A human rights-based practice framework for sexual offenders. Journal of Sexual Aggression 2008;14(2):87-98. 63. Ward T, Mann RE, Gannon TA. The good lives model of offender rehabilitation: Clinical implications. Aggression and Violent Behavior 2007;12(1):87-107. 64. Ward T, Gannon T, Yates PM. The treatment of offenders: current practice and new developments with an emphasis on sex offenders. International Review of Victimology 2008;15(2):179-204. 65. Bonta J, Andrews DA. Risk-Need-Responsivity model for offender assessment and rehabilitation. Ottawa, ON: Public Safety Canada; 2007. 66. Bourgon G, Hanson K, Bonta J. Risk, need, and responsivity: a heuristic for evaluating the “quality” of offender interventions. Public Safety Canada; 2008 p. 45-9. 67. Andrews DA. Extensions of the Risk-Need-Responsivity (RNR) model of assessment and correctional treatment. Ottawa, ON: Public Safety Canada; 2008 p. 7-11. 68. Giltay EJ, Gooren LJG. Potential side effects of androgen deprivation treatment in sex offenders. Journal of the American Academy of Psychiatry and the Law 2009;37(1):53-8. 69. Berlin FS. Risk/benefit ratio of androgen deprivation treatment for sex

80 Treatment of Convicted Adult Male Sex Offenders offenders. Journal of the American Academy of Psychiatry and the Law 2009;37(1):59-62. 70. Polak MA, Nijman H. Pharmacological treatment of sexually aggressive forensic psychiatric patients. Psychology, Crime & Law 2005;11(4):457-65. 71. Berlin FS. Commentary: Risk/benefit ratio of androgen deprivation treatment for sex offenders. Journal of the American Academy of Psychiatry & the Law 2009;37(1):59-62. 72. Berlin FS. Commentary: Risk/benefit ratio of androgen deprivation treatment for sex offenders. Journal of the American Academy of Psychiatry and the Law 2009 Mar;.37(1):-62. 73. Dauvergne M. Crime statistics in Canada 2007. Juristat 2008;28(7):Statistics Canada Catalogue no. 85-002-XPE-Ottawa. 74. Kong R, Johnson H, Beattie S, Cardillo A. Sexual offences in Canada. Juristat 2002;23(6)Ottawa. 75. Statistics Canada. Crimes and offences. Statistics Canada 2010. Available at: http://www40.statcan.gc.ca/l01/ind01/l3_2693_2102-eng.htm?hili_legal19. 76. Brennan S, Taylor-Butts A. Sexual assault in Canada 2004 and 2007. Ottawa, ON: Minister of Industry; 2008. 77. Statistics Canada. Sexual offenses. The Daily, Friday, July 25 2003. 78. Department of Justice Canada. Bill C-46: records applications post-mills, a caselaw review. Ottawa, ON: Department of Justice Canada; 2008. 79. Ontario Women’s Directorate. Sexual assault: dispelling the myths. Ontario Women’s Directorate 2010. 80. The Alberta Association of Sexual Assault Centres. Sexual assault/abuse. The Alberta Association of Sexual Assault Centres 2005. Available at: http://www.aasac.ca/fact-sexual-assault-abuse.htm. 81. Motiuk LL, Vuong B. Sex offenders. Correctional Service of Canada Research Briefs 2005;B-37. 82. Marshall WL, Williams S. Assessment and treatment of sexual offenders. Forum on Corrections Research 2000;12(2):41-4. 83. Studer LH, Aylwin AS. Male victims and post treatment risk assessment among adult male sex offenders. International Journal of Law and Psychiatry 2008;31(1):60-5. 84. Studer LH, Aylwin AS, Reddon JR. Testosterone, sexual offense recidivism, and treatment effect among adult male sex offenders. Sexual Abuse: Journal of Research and Treatment 2005;17(2):171-80. 85. Studer LH, Reddon JR, Roper V, Estrada L. Phoenix: an in-hospital treatment program for sex offenders. Journal of Offender Rehabilitation 1996;23(1-2):91-7.

Treatment of Convicted Adult Male Sex Offenders 81 86. Studer LH, Clelland SR, Aylwin AS, Reddon JR, Monro A. Rethinking risk assessment for incest offenders. International Journal of Law and Psychiatry 2000;23(1):15-22. 87. Clelland SR, Studer LH, Reddon JR. Follow-up of rapists treated in a forensic psychiatric hospital. Violence and Victims 1998;13(1):79-86. 88. Aos S, Miller M, Drake E. Evidence-based adult corrections programs: What works and what does not? Olympia, WA: unknown; 2006. 89. Bilby C, Brooks-Gordon B, Wells H. A systematic review of psychological interventions for sexual offenders II: Quasi-experimental and qualitative data. Journal of Forensic Psychiatry & Psychology 2006;17(3):467-84. 90. Polizzi DM, MacKenzie DL, Hickman LJ. What works in adult sex offender treatment? A review of prison- and non-prison-based treatment programs. International Journal of Offender Therapy and Comparative Criminology 1999;43(3):357-74. 91. Losel F, Schmucker M. The effectiveness of treatment for sexual offenders: a comprehensive meta-analysis. Journal of Experimental Criminology 2005;1(1):117-46. 92. Hanson RK, Public SC. A meta-analysis of the effectiveness of treatment for sexual offenders: risk, need, and responsivity. Ottawa: Public Safety Canada; 2009. 93. Beech A, Bourgon G, Hanson RK, Harris AJR, Langton C, Marques J, et al. Sexual offender treatment outcome research : CODC guidelines for evaluation. Part 1, Introduction and overview. Ottawa: Public Safety Canada; 2007. 94. Beech A, Bourgon G, Hanson K, Harris AJR, Langton CM, Marques JK, et al. The Collaborative Outcome Data Committee’s guidelines for the evaluation of sexual offender treatment outcome research. Part 2: CODC guidelines. Ottawa, ON: Public Safety Canada; 2007. 95. Harrisson K. Legal and ethical issues when using antiandrogenic pharmacotherapy with sex offenders. Sexual Offender Treatment 2008;3(2). 96. Rainey B, Harrison K. Pharmacotherapy and human rights in sexual offenders: best of friends or unlikely bedfellows. Sexual Offender Treatment 2008;3(2). 97. Marques JK, Wiederanders M, Day DM, Nelson C, van Ommeren A. Effects of a Relapse Prevention Program on Sexual Recidivism: Final Results From California’s Sex Offender Treatment and Evaluation Project (SOTEP). Sexual Abuse: Journal of Research and Treatment 2005;17(1):79-107. 98. ATSA Professional Issues Committee. Practice standards and guidelines for members of the Association for the Treatment of Sexual Abusers. Beaverton OR: The Association for the Treatment of Sexual Abusers; 2003.

82 Treatment of Convicted Adult Male Sex Offenders 99. Ward T, Eccleston L. Risk, responsivity, and the treatment of offenders: introduction to the special issue. Psychology, Crime & Law 2004;10(3):223-7. 100. Wilson RJ, Yates PM. Effective interventions and the Good Lives Model: Maximizing treatment gains for sexual offenders. Aggression and Violent Behavior 2009;14(3):161. 101. Ward T, Melser J, Yates PM. Reconstructing the Risk-Need-Resposivity model: a theoretical elaboration and evaluation. Aggression and Violent Behavior 2007;12:208-28. 102. Harris A, Phenix A, Hanson RK, Thornton D. Static-99 coding rules: Revised 2003. Ottawa, ON: Solicitor General Canada; 2003. 103. Cook DJ, Sackett DL, Spitzer WO. Methodologic guidelines for systematic reviews of randomized control trials in health care from the Potsdam Consultation on Meta-Analysis. Journal of Clinical Epidemiology 1995;48(1):167-71. 104. Shea B, Grimshaw JM, Wells GA, Boers M, Andersson N, Hamel C, et al. Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Medical Research Methodology 2007;7(10)7-10. 105. Shea BJ, Hamel C, Wells GA, Bouter LM, Kristjansson E, Grimshaw J, et al. AMSTAR is a reliable and valid measurement tool to assess the methodological quality of systematic reviews 132. Journal of Clinical Epidemiology 2010;62(10):1013-20.

Treatment of Convicted Adult Male Sex Offenders 83 IHE Consensus Statements Consensus Development Conference Statement on Fetal Alcohol Spectrum Disorder (FASD) – Across the Lifespan (2009) Consensus Statement on Depression in Adults: How to Improve Prevention, Diagnosis and Treatment (2008) Consensus Statement on Healthy Mothers, Healthy Babies: How to Prevent Low Birth Weight (2007) Consensus Statement on Self-monitoring in Diabetes (2006)

IHE Briefs

2009 Comparative Effectiveness: An Overview

IHE Book Series Parkinson Disease: A Health Policy Perspective (2009) Chronic Pain: A Health Policy Perspective (2008) Cost Containment and Efficiency in National Health Systems: A Global Comparison (2008) Financing Health Care: New Ideas for a Changing Society (2007)

IHE Reports

2010 Means Restriction for Suicide Prevention Sexual Exploitation of Children and Youth Over the Internet IHE In Your Pocket: A Handbook of Health Economic Statistics Human papillomavirus (HPV) Testing in Alberta 2009 Health Technology Assessment on the Net – 11th Edition Exercise Testing for the Prediction of Cardiac Events in Patients with Diabetes Effectiveness of Organizational Interventions for the Prevention of Occupational Stress Assistive Reproductive Technologies: A Literature Review and Database Analysis

84 Treatment of Convicted Adult Male Sex Offenders 2008 Air Ambulance with Advanced Life Support Effective Dissemination of Findings from Research - A Compilation of Essays Research - A Compilation of Essays Health Technology on the Net (Tenth Edition) IHE In Your Pocket: A Handbook of Health Economic Statistics Spousal Violence Against Women: Preventing Recurrence The Importance of Measuring Health related Quality of Life Using Fetal Fibronectin to Diagnose Preterm Labour How Much Should We Spend on Mental Health? CT and MRI Services in Alberta: Comparisons with Other Health Care Systems Islet Transplantation for the Treatment of Type 1 Diabetes – An Update Determinants and Prevention of Low Birth Weight: a Synopsis of the Evidence Comparative Effectiveness: An Overview 2007 World in Your Pocket: A Handbook of International Health Economic Statistics Mental Health Economic Statistics In Your Pocket (Revised) Cost-effectiveness in the Detection of Syphilis Economics of Childhood Immunization in Canada: Databook Evidence of Benefits from Telemental Health: A Systematic Review Health Technology on the Net (Ninth Edition) Routine Pre-operative Tests – Are They Necessary? Screening Newborns for Cystic Fibrosis Screening Newborns for Hearing The Use and Benefits of Teleoncology The Use of Nitric Oxide in Acute Respiratory Distress Syndrome The Use of Videoconferencing for Mental Health Services in Canada and Finland 2006 Health Technology on the Net (Eighth Edition) IHE In Your Pocket: A Handbook of Health Economic Statistics This health technology assessment report is an overview of systematic reviews reporting on the efficacy/effectiveness of psychotherapy and/or pharmacotherapy interventions delivered within Sex Offender Treatment (SOT) programs to reduce the risk of re–offending in convicted adult male sex offenders.

Institute of Health Economics 1200 - 10405 Jasper Avenue Edmonton AB Canada T5J 3N4 Tel. 780.448.4881 Fax. 780.448.0018 [email protected] www.ihe.ca

ISBN 978-1-897443-74-3 (print) ISSN 978-1-897443-75-0 (online)